BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Luo, Meifei
Abrégé
The present disclosure provides a pick-up point recommendation method and system, an electronic device, a medium, and a program product. The pick-up point recommendation method comprises: acquiring travel information of a user, the travel information comprising initial travel starting point information and travel destination information; when the user enables a recommendation function, on the basis of the travel information, determining a recommended pick-up point from among a plurality of candidate pick-up points corresponding to the initial travel starting point information, the distance of a route between the recommended pick-up point and a travel destination of the user satisfying a preset condition; and displaying on a travel information display page position information of the recommended pick-up point, the position information of the recommended pick-up point being used to update the initial travel starting point information in the travel information.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Cheng
Abrégé
Embodiments of the present invention provide a meeting assistance method and system, and a storage medium. The method comprises: determining the distance between a service provider and a preset meeting position; when the distance is smaller than a distance threshold, instructing the service provider or a vehicle corresponding to the service provider to send a short-range wireless communication signal; instructing a service requester to establish a communication connection to the service provider or the vehicle on the basis of the short-range wireless communication signal; and guiding, on the basis of the communication connection, the service requester to meet the service provider.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Cui, Qiaoling
Yang, Xiaolong
Xiahou, Youjie
Wang, Jiahao
Abrégé
Embodiments of the present application provide a method and apparatus for managing combined travel. The method provided herein comprises: acquiring trip information, wherein the trip information at least indicates a start point and an end point of a trip; and presenting combined travel recommendation information on a shared travel service interface, wherein the combined travel recommendation information is used for describing a target combined travel route, the target combined travel route is determined on the basis of the trip information, and the shared travel service interface further provides at least one candidate shared travel service for the trip. On the basis of such a mode, the embodiments of the present application can provide recommendation information for a combined travel route on the shared travel service interface, so as to provide richer travel choices for a user.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Yunfan
Gao, Sheng
Huang, Datao
Abrégé
The present disclosure provides a carpooling method and system, an electronic device, a medium, and a program product. The method comprises: obtaining and displaying, on the basis of departure point information of a first client, description information of at least one available vehicle; in response to receiving a trigger operation for the description information of any available vehicle, displaying travel detail information corresponding to the any available vehicle, information about each station in a travel route of the any available vehicle being displayed in the travel detail information; and generating carpooling order information according to the selected destination station information in the information about each station of the any available vehicle, and sending the carpooling order information to a server.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Chen, Yi
Zhang, Hongzhi
Abrégé
Provided in the present application are an information pushing method and apparatus, and an electronic device and a storage medium. The information pushing method comprises: receiving a selection operation which is for a message pushing icon and is issued by a first user in a first display interface of a first target program, and displaying a second display interface, wherein the first display interface and the second display interface are two interfaces which both belong to the first target program; receiving a selection operation of the first user for a second user in a message user list to be pushed that is in the second display interface, and acquiring second user identification information corresponding to the second user selected by the first user; and in a second target program, pushing a target message to the second user by means of a second target program user identifier corresponding to the second user identification information, wherein the second target program and the first target program are two different programs. By means of the present application, the efficiency of a first user pushing a message to other users can be improved.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Mingwen
Zhang, Tianming
Zhao, Ningning
Abrégé
Embodiments of the present description disclose a target discrimination method and system. The target discrimination method comprises: determining a first image sequence on the basis of a plurality of target images, the image capture time of the plurality of target images having a correspondence with the irradiation time of a plurality of lights in a lighting sequence irradiated on a target object; determining a second image sequence on the basis of a plurality of color template images, the plurality of color template images being generated on the basis of the lighting sequence; and on the basis of the first image sequence and the second image sequence, determining the authenticity of the plurality of target images.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Mingwen
Zhang, Tianming
Zhao, Ningning
Abrégé
Disclosed in the embodiments of the present description are a target identification method and system. The target identification method comprises: acquiring a plurality of target images, wherein photographing times of the plurality of target images have a correlation with irradiation times of a plurality of illuminations in an illumination sequence that are irradiated to a target object, the plurality of illuminations have a plurality of colors, the plurality of colors comprise at least one reference color and at least one verification color, and each of the at least one verification color is determined on the basis of at least some of the at least one reference color; and determining the authenticity of the plurality of target images on the basis of the illumination sequence and the plurality of target images.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
G06K 9/62 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
G06Q 20/40 - Autorisation, p.ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasin; Examen et approbation des payeurs, p.ex. contrôle des lignes de crédit ou des listes négatives
8.
IMAGE VERIFICATION METHOD AND SYSTEM, AND STORAGE MEDIUM
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Mingwen
Zhang, Tianming
Zhao, Ningning
Abrégé
Disclosed in the embodiments of the present description are an image verification method, system and apparatus, and a storage medium. The image verification method comprises: instructing a terminal device to emit at least two beams of colored light to a target; during the process of emitting the at least two beams of colored light, instructing the terminal device to collect a plurality of images of the target; selecting, from the plurality of images, at least two target images that respectively correspond to the at least two beams of colored light; on the basis of the at least two target images, determining a target verification code; and performing image authenticity verification by comparing the target verification code with a reference verification code corresponding to the at least two beams of colored light.
G06V 10/60 - Extraction de caractéristiques d’images ou de vidéos relative aux propriétés luminescentes, p.ex. utilisant un modèle de réflectance ou d’éclairage
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Cheng, Bo
Zhang, Tianming
Abrégé
The embodiments of the present description provide a method and system for identifying a target. The method comprises: acquiring at least one image capture parameter related to an image capture frame; sending an image capture instruction to a client, the image capture instruction instructing the client to display the image capture frame on the basis of the at least one image capture parameter; receiving at least one image from the client; and determining the authenticity of the at least one image on the basis of the at least one image capture parameter.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Mingwen
Zhang, Tianming
Zhao, Ningning
Abrégé
The embodiments of the present specification disclose a method and system for target recognition. The method for target recognition comprises: determining an illumination sequence, the illumination sequence being used to determine multiple colors of multiple illuminations emitted by a terminal when irradiating a target object; acquiring multiple target images on the basis of the terminal, the image capture times of the multiple target images having a correspondence with the irradiation times of the multiple illuminations; and determining, on the basis of the illumination sequence and the multiple target images, the authenticity of the multiple target images.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
G06K 9/62 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
G06Q 20/40 - Autorisation, p.ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasin; Examen et approbation des payeurs, p.ex. contrôle des lignes de crédit ou des listes négatives
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Fang, Jun
Abrégé
An interaction method and an interaction apparatus. After acquiring a path navigation information of a terminal, a server (12) determines road condition information of each road section according to a location of a target object relative to a corresponding lane of a corresponding road section in a road condition collection sequence of each road section in the previously acquired path navigation information, and after determining a target image corresponding to a road section in the road sections of which road condition information meets preset road conditions, sends the target image to the terminal. After receiving the target image, the terminal may perform, on a navigation page, rendering display on a road condition display control for displaying the target image. The location of each target object can be accurately determined by means of image recognition, and the road condition information of each road section is determined according to the location of each target object. Meanwhile, a road condition of a specific road section is displayed by means of a real scene image. Thus, the accuracy and timeliness of road condition determination are improved, such that a user can avoid a congested road section in a timely fashion.
G01C 21/36 - Dispositions d'entrée/sortie pour des calculateurs embarqués
G01C 21/20 - Instruments pour effectuer des calculs de navigation
G01C 21/28 - Navigation; Instruments de navigation non prévus dans les groupes spécialement adaptés pour la navigation dans un réseau routier avec corrélation de données de plusieurs instruments de navigation
12.
METHOD AND APPARATUS FOR DOMAIN ADAPTATION LEARNING, AND DEVICE, MEDIUM AND PRODUCT
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Xu, Pengfei
Wang, Yatian
Song, Xiaolin
Zhao, Sicheng
Abrégé
The present disclosure relates to a method and apparatus for domain adaptation learning, and a device, a medium and a product. The method described herein comprises: acquiring a source video sample set of a source domain and a target video sample set of a target domain, wherein source video samples in the source video sample set are marked as belonging to one of a plurality of known categories; on the basis of a plurality of similarities between a plurality of target video samples in the target video sample set and the source video sample set, determining the probability of each of the plurality of target video samples belonging to an unknown category; and at least on the basis of the plurality of target video samples and the corresponding probabilities thereof, making a multi-classification model, which is previously pre-trained by using the source video sample set, adaptive to the target domain. Therefore, more reliable domain adaptation learning for video classification can be realized.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Che, Zhengping
Yin, Chengxiang
Wu, Kun
Xu, Zhiyuan
Tang, Jian
Abrégé
Embodiments of the disclosure provide a method for knowledge-based visual question answering, a system, and a computer program product thereof. The method includes storing the candidate supporting facts in memory slots. The method further includes generating a second visual-semantic representation for the image-question pair. The method additionally includes applying a first set of attentions over the memory slots storing the candidate supporting facts, to obtain a memory summary based on the second visual-semantic representation of the image-question pair. The method additionally includes applying a second set of attentions to generate a third visual-semantic representation of the image-question pair based on the obtained memory summary. The method additionally includes generating an updated visual-semantic representation of the image-question pair based on the third visual-semantic representation. The method additionally includes identifying a candidate supporting fact with a highest probability as a target supporting fact for the answer to the question related to the image.
G06K 9/72 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques utilisant une analyse de contexte basée sur l'identité provisoire attribuée à une série de formes successives, p.ex. d'un mot
G06K 9/62 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
14.
BATTERY STATE OF HEALTH PREDICTION METHOD AND APPARATUS, AND ELECTRONIC DEVICE AND READABLE STORAGE MEDIUM
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Yang, Jing
Liu, Feng
Huang, Zhixin
Yang, Lei
Zhou, Wencan
Zhang, Xing
Li, Dan
Abrégé
A battery state of health prediction method and apparatus, and an electronic device and a readable storage medium, which relate to the technical field of computers. The method may comprise: determining, by means of a preset first correlation, the current state of health corresponding to the current driving mileage of a target electric vehicle (22); and then, on the basis of a pre-trained machine learning model and by comprehensively taking the current battery state of health, the current driving mileage and the current battery parameter into consideration, predicting a target battery state of health corresponding to a target driving mileage of the target electric vehicle (23). During a prediction process, a prediction result is more accurate by combining a current battery state of health.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Bai, Bing
Xing, Tengfei
Wang, Zhihui
Xu, Pengfei
Abrégé
A method and apparatus for presenting road information, a device, a storage medium, and a program product. The method comprises: presenting a road visual element (184) associated with a target road (110) (402); and in response to determining that the congestion degree of the target road (110) is higher than a predetermined threshold, presenting at least one road condition image (182) in association with the road visual element (184), the at least one road condition image (182) being selected from a group of road condition images (140) that are used for determining the congestion degree, and the group of road condition images (140) being acquired by an image acquisition device (130) associated with a vehicle (120) traveling on the target road (110) (404). On the basis of this mode, more accurate road condition information can be provided for a user in a more intuitive mode.
G08G 1/0967 - Systèmes impliquant la transmission d'informations pour les grands axes de circulation, p.ex. conditions météorologiques, limites de vitesse
16.
METHOD FOR ENCRYPTING VERIFICATION CODE IMAGE, AND DEVICE, STORAGE MEDIUM AND COMPUTER PROGRAM PRODUCT
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Hao
Zhang, Tianming
Wang, Zhiheng
Zhou, Shiqi
Abrégé
The embodiments of the present disclosure relate to a method for encrypting a verification code image, and a device, a storage medium and a computer program product. The method comprises: generating an original verification code image, wherein the original verification code image corresponds to a first character string; determining scramble gradient information of the original verification code image according to a second character string, wherein the scramble gradient information comprises an adjustment gradient of each pixel in the original verification code image, and the second character string is a scramble character string of the first character string; and adjusting the original verification code image according to the scramble gradient information, so as to generate an encrypted verification code image, wherein a neural network recognition result of the encrypted verification code image is the second character string. By using the method, the security of a verification code image can be improved, thereby effectively preventing malicious registration behavior.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Hao
Zhang, Tianming
Xue, Taolue
Wang, Zhiheng
Zhou, Shiqi
Abrégé
The embodiments of the present disclosure relate to a keyboard encryption method, and a device, a storage medium and a computer program product. The method comprises: performing feature extraction on a keyboard character image, so as to determine a feature of the keyboard character image; determining a fusion feature according to the feature of the keyboard character image and a feature of a wrong keyboard character; determining an encrypted keyboard character image according to the fusion feature, wherein the wrong keyboard character is different from the character corresponding to the keyboard character image; and constructing an encrypted keyboard on the basis of the encrypted keyboard character image, wherein a neural network recognition result of the encrypted keyboard character image is a wrong keyboard character, and the encrypted keyboard character image has the same visual recognition result as the keyboard character image. By using the method, the security of a keyboard can be improved, thereby effectively preventing the occurrence of malicious registration behavior.
G06F 21/83 - Protection des dispositifs de saisie, d’affichage de données ou d’interconnexion dispositifs de saisie de données, p.ex. claviers, souris ou commandes desdits claviers ou souris
G06K 9/62 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Li, Zhendong
Ge, Fangfei
Chen, Xin
Zhang, Wentian
Yuan, Yifang
Abrégé
A ride-sharing system can automatically and autonomously determine whether a rider is an account holder associated with an account of a ride-sharing service that requested a ride or a guest rider. A guest rider detection system can use geolocation data to determine whether a rider is a guest rider or an account holder. The system can obtain location data for an account holder that requests a ride using a ride sharing application. The location data of the account holder can be compared with one or more locations (e.g., pickup or drop off location) associated with the requested ride. Based at least in part on the comparisons between the location data of the account holder and the one or more locations associated with the requested ride, the system can determine with a particular degree of certainty whether the rider is the guest rider or the account holder.
H04W 4/02 - Services utilisant des informations de localisation
H04W 4/60 - Services basés sur un abonnement qui utilisent des serveurs d’applications ou de supports d’enregistrement, p.ex. boîtes à outils d’application SIM
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Li, Ronghao
Xu, Pengfei
Wang, Liang
Xu, Bin
Ma, Chaowei
Cai, Chao
Zhang, Song
Zhang, Lei
Liu, Tao
Yang, Tao
Hu, Meng
Zhou, Kang
Ma, Li
Hu, Runbo
Abrégé
A navigation method and apparatus, a device, a storage medium and a program product. A method comprises: acquiring a destination position associated with a first object, wherein the destination position is also associated with a second object, and the destination position indicates a predetermined site at which the first object can meet the second object (202); determining the current position of the first object (204); and providing a navigation interface, the navigation interface comprising navigation elements superimposed on a real-scene image associated with the current position, the navigation elements being determined on the basis of the destination position and the current position (206). The present invention can provide intuitive navigation for objects to meet, thereby increasing the efficiency of meeting between objects.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Bai, Ning
Zhang, Lei
Liu, Tao
Shen, Chao
Abrégé
The embodiments of the present description relate to a method and apparatus for optimizing a journey of a user, an electronic device, a storage medium, and a computer program product. In the method, a desired drop-off place for a journey by a user is determined. If it is determined that the desired drop-off place is located at a no-stop road section, a new drop-off place for the journey is determined. Then, instead of the desired drop-off place, a new candidate drop-off place is displayed to the user and an indication relating to the desired drop-off place being located at a no-stop road section is displayed. Further provided are a corresponding apparatus, an electronic device, and a storage medium. By using the described implementation means, it is possible to optimize the journey of a user by displaying a new drop-off place to the user instead of a desired drop-off place.
G06Q 10/04 - Prévision ou optimisation spécialement adaptées à des fins administratives ou de gestion, p. ex. programmation linéaire ou "problème d’optimisation des stocks"
G06F 16/9537 - Recherche à dépendance spatiale ou temporelle, p.ex. requêtes spatio-temporelles
21.
METHOD AND APPARATUS FOR USE IN RECOMMENDING PICKUP SPOT, DEVICE, MEDIUM, AND PROGRAM PRODUCT
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Lei
Bai, Ning
Liu, Tao
Shen, Chao
Abrégé
Provided in embodiments of the present description are a method and apparatus for use in recommending a pickup spot, a device, a medium, and a program product. The method provided comprises: providing a first position located in a stopping-permitted road section to a service requester as a target pickup spot; if a determination is made that the service requester attempts to change the target pickup spot from the first position to a second position located in a stopping-prohibited road section, automatically changing the target pickup spot to a third position located in a stopping-permitted road section; and providing a stopping-prohibited prompt regarding stopping-prohibited road sections to the service requester. By such means, the service requester could easily understand and accept the transfer of the target pickup spot to a safe stopping-permitted zone.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Lei
Bai, Ning
Liu, Tao
Shen, Chao
Abrégé
Provided in the embodiments of the present description are a method, apparatus, and device for use in prompting a no-parking road section, a computer-readable storage medium, and a computer program product, applicable in the field of intelligent traffic. The method of the present description comprises: determining a stop associated with a trip; determining a set of no-parking road sections in the vicinity of the stop; and if the distance between a transport and at least one no-parking road section in the set of no-parking road sections is determined to be less than a first threshold distance, issuing a prompt related to the set of no-parking road sections by means of a terminal associated with the trip. The embodiments of the present description effectively reduce the probability of a user parking illegally, thus enhancing user experience.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Lei
Bai, Ning
Liu, Tao
Shen, Chao
Abrégé
A method and apparatus for optimizing a boarding point, a device, a medium, and a program product. The method proposed therein comprises determining a target boarding point of a service requester (110), a service provider (120) needing to arrive at the target boarding point to pick up the service requester (110); determining whether the target boarding point is located on a parking restriction road section; and if determining that the target boarding point is located on the parking restriction road section, providing prompt information for the service requester (110) on the basis of relative positioning states of the service provider (120) and the target boarding point, the prompt information at least comprising parking restriction information for indicating the parking restriction characteristics of the target boarding point. According to the method, by indicating the service requester (110) with the parking restriction characteristics, the service requester (110) can be guided to board as soon as possible within a reasonable time, such that a dwell time of a vehicle of the service provider (120) on the parking restriction road section can be avoided, thereby avoiding the triggering of a parking restriction penalty on the road section, and ensuring smooth and safe road traffic.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Li, Zhendong
Yu, Yunhan
Liu, Fengyi
Abrégé
A method for identity risk assessment may include receiving a target identity (ID) verification event of a first user account from a user terminal, the target ID verification event including a target ID number. The method may also include obtaining a first feature associated with the first user account and a second feature associated with the target ID number based on the target ID verification event. The method may also include determining whether the target ID verification event is a malicious ID verification event using a risk assessment model to analyze the first feature and the second feature. In response to determining that the target ID verification event is a malicious ID verification event, the method may further include blocking at least one of the first user account and the target ID number.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Liu, Lian
Sun, Yiqiang
Abrégé
A method, an apparatus, a device, and a medium for use in data processing. The method comprises: acquiring a target image associated with a target object, the target image having spots, the spots being generated by the target object reflecting a laser beam of a specific wavelength (210); determining from the target image a target sub-image associated with a specific part of the target object (220); and performing a living body detection with respect to the target object on the basis of the target sub-image (230). By such means, whether the target object is a live user can be determined accurately and efficiently.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
26.
SYSTEMS AND METHODS FOR OBTAINING ESTIMATED TIME OF ARRIVAL IN ONLINE TO OFFLINE SERVICES
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Luo, Qing
Wang, Zheng
Abrégé
A method may include receiving a service request; obtaining a route linking the starting location and the destination; obtaining one or more global features associated with the service request; using a first model to analyze the one or more global features associated with the service request to produce a global vector for the service request; using a second model to analyze the features of each of the road segment to produce a local segment vector for each of the road segments; determining a segment weight of the global vector for each road segment based on the global vector and its corresponding local segment vector; obtaining a global segment vector for each road segment based on its corresponding segment weight and the global vector; obtaining a segment vector for each road segment by combining its corresponding local and global segment vectors; obtaining the ETA for the service request based on the segment vectors.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Huang, Tao
Qin, Zhiwei
Li, Qingyang
Shang, Wenjie
Abrégé
A computer-implemented method for determining incentive distribution includes: obtaining, by a computing device, a computer model, the computer model being configured to: obtain state information of a user of a platform and an incentive action of the platform, and generate simulation results on at least one performance criterion of the platform with and without the incentive action being provided to the user; receiving a computing request comprising state information of one or more visiting users; determining, by feeding the state information and the incentive action to the computer model, an uplift on the at least one performance criterion by providing the incentive action. By evaluating the uplift effect based on statistical distribution of the order and reward, the activeness of the user, and counter-factual balance of the data, the method improves the accuracy of the uplift prediction, thereby improving efficiency and accuracy of the incentive distribution.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Farmer, Kenneth
Baruffaldi, Paola, Giovanna, Piacentini
Abrégé
Systems and methods are disclosed configured to receive a ride request from an application hosted on a device of a user. A vehicle is routed to a pickup location. A route is generated configured to navigate the vehicle to a destination. Sensor data indicative of ride quality is monitored while the vehicle is navigating the generated route, wherein the sensor data comprises data indicating at least how bumpy the ride is and/or whether a sudden braking event occurred. When an adverse ride condition is detected, a determination is made that a content item is to be transmitted to a display device. A geofence is generated about the destination using a user content interaction history. The first item of content is selected based at least in part on an identification of an establishments within the geofence. The selected item of content is transmitted to the display device and the user may be transported to the pickup location within the geofence.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Shang, Wenjie
Li, Qingyang
Qin, Zhiwei
Abrégé
A method includes: selecting a current discount strategy according to a simulation result of a simulator of a machine learning model, wherein the simulation result comprises simulations of future transportation order bubbling in response to discounts given to current transportation order bubbling; obtaining a plurality of bubbling features of a transportation plan of a user, wherein the plurality of bubbling features comprise (i) a bubble signal comprising time information and location information corresponding to the transportation plan (ii) a supply and demand signal comprising transportation supply-demand information corresponding to the transportation plan, and (iii) a transportation order history signal of the user; determining a discount signal according to the plurality of bubbling features and the current discount strategy; and transmitting the discount signal to a computing device of the user.
G06Q 10/06 - Ressources, gestion de tâches, des ressources humaines ou de projets; Planification d’entreprise ou d’organisation; Modélisation d’entreprise ou d’organisation
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Li, Qingyang
Qin, Zhiwei
Abrégé
A computer-implemented method includes: obtaining a model comprising an environment module, a resource allocation module, and a personal recommendation module, receiving a real-time online signal of visiting the platform from a computing device of a visiting user; determining a resource allocation action by feeding user contextual data of the visiting user to the model; and based on the determined resource allocation action, transmitting a return signal to the computing device to present the resource allocation action.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
He, Xudong
Wang, Yanke
Wang, Minkang
Abrégé
Systems and methods for map construction. The systems may include a processing circuit connected to a first sensor and a second sensor. The systems may perform the methods to receive first data including location-orientation information of a vehicle obtained by the first sensor (610); receive second data including road-surroundings information of the vehicle obtained by the second sensor (620); correct the first data based on the second data (630); and construct a map based on the corrected first data and the second data (640).
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zou, Yayi
Qin, Zhiwei
Abrégé
Methods, systems, and apparatus, including computer programs encoded on computer storage media for task control based on Bayesian Meta-Reinforcement learning. An exemplary method includes obtaining a base machine learning (ML) model trained based on historical data collected from historical tasks. The base ML model represents a prior distribution of model parameters in a neural network representing control policies. The exemplary method further includes receiving observed data from a new control task; training a task-level ML model based on the base ML model and the observed data, wherein the task-level ML model represents a posterior distribution of the model parameters; sampling, based on the posterior distribution of the model parameters, a set of the model parameters that represent a control policy; and applying the control policy in performing the new control task.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Ni, Zetian
Yi, Zihan
Yang, Kaidan
Chen, Xin
Abrégé
The present disclosure provides a system for improving security. The system may identify a query associated with a user account, and access an ID graph database to obtain an ID graph relating to the user account by a database driver. The system may also determine whether the user account is a target account type based at least on the ID graph. The ID graph may include a plurality of nodes and a plurality of edges. Each of the plurality of edges may connect two nodes. Each of the plurality of nodes may include at least one of a register ID, a login ID, a payment ID, a background check ID, or a face ID. Each edge that connects two nodes may include at least one of a user type associated with the two nodes, a timestamp when the edge is connected, or source information of the edge.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Farmer, Kenneth
Baruffaldi, Paola, Giovanna, Piacentini
Abrégé
A ride-sharing system can automatically and autonomously identify connections of a user by accessing one or more third-party networking services used by the user. Further, the system can determine which of the connections of the user spend at least a threshold amount of time in a geographic area where the user spends time, such as a residential area or workplace of the user. When a user requests a ride-share pool ride, the ride may be scheduled with a connection of the user improving the safety of the ride. Machine learning models may be used to predict the likelihood that a connection of the user will travel along the same route as the user and will improve the safety of the rider.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Yang, Jiantao
He, Minghui
Shu, Weihuan
Ma, Li
Abrégé
A system for pick-up location recommendation is provided. The system may receive a start location of a service order from a requester terminal, and determine a plurality of candidate pick-up locations around the start location base on a location information database. Each candidate pick-up location may have a relative position to a boundary of the start location. The system may determine one or more recommended pick-up locations among the plurality of candidate pick-up locations by applying a pick-up location recommendation model to the plurality of candidate pick-up locations. An input of the pick-up location recommendation model at least may include the relative positions of the candidate pick-up locations to the boundary of the start location. The system may further transmit the one or more recommended pick-up locations to the requester terminal for recommendation.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Liu, Bolong
Abrégé
The present disclosure relates to systems and methods for recommending pick-up locations. The method may include obtaining and storing a current location of the user, a plurality of historical locations associated with the current location of the user, and one or more potential pick-up locations corresponding to each of the plurality of historical locations of the user. The method may also include extracting feature data from the current location, the plurality of historical locations, and the one or more potential pick-up locations corresponding to each of the plurality of historical locations of the user. The method may further include inputting the extracted feature data into a pick-up location recommendation model to determine a transition probability for each potential pick-up location. The method may still include determining a recommended pick-up location for the user based on the determined transition probability for each potential pick-up location.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Liu, Qian
Chen, Mingquan
Shen, Chao
Abrégé
The present disclosure relates to systems and methods for displaying one or more points of interest (POIs) of a location-based service (LBS). The method may include receiving a request for an LBS sent by a user via a user terminal and obtaining a start location based on the request for the LBS. The method may also include determining one or more recommended POIs based on the start location and POI data associated with the start location. Further, the method may also include transmitting signals to the user terminal to display at least one of the one or more recommended POIs.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Shu, Weihuan
Abrégé
The embodiments of the present description provide an ultra-wideband-based positioning method and apparatus, an electronic device and a readable storage medium, relating to the technical field of computers. In the embodiments of the present description, as a short-range wireless communication connection is not easily disturbed, a first device and a second device can determine an accurate relative position on the basis of the short-range wireless communication connection, when the absolute position corresponding to the first device or the second device is not accurate (for example, when a position is determined by means of satellite positioning, a satellite positioning point may be inaccurate due to a weak satellite signal), the first device or the second device can determine the position of the other party according to the accurate relative position. In an application scenario of online ride-hailing, by means of the embodiments of the present description, an online ride-hailing driver and a passenger can find each other quickly.
H04W 4/02 - Services utilisant des informations de localisation
G01S 5/02 - Localisation par coordination de plusieurs déterminations de direction ou de ligne de position; Localisation par coordination de plusieurs déterminations de distance utilisant les ondes radioélectriques
39.
ARTIFICIAL INTELLIGENCE SYSTEM FOR TRANSPORTATION SERVICE RELATED SAFETY ISSUES DETECTION BASED ON MACHINE LEARNING
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Ma, Baochang
Zhou, Yulong
Han, Kun
Abrégé
An artificial intelligence system for detecting a transportation service related safety issue. The system includes a storage device configured to store a first sample description of an incident and a known safety issue of the incident. The first sample description includes first set of labeled texts. The system further includes at least one processor. The at least one processor is configured to segment the first sample description into word segments, determine word vectors of the respective word segments and determine features of the first set of labeled texts in the first sample description. The at least one processor is also configured to train the learning model based on the word vectors and the features. The system also includes a communication interface configured to provide the learning model for automatically detecting the transportation service related safety issue from an incident description.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Yi
Deng, Chengyun
Ma, Shiqian
Sha, Yongtao
Song, Hui
Abrégé
Methods, systems, and apparatus, including computer programs encoded on computer storage media for acoustic echo cancellation and suppression are provided. An exemplary method comprises receiving a far-end acoustic signal and a corrupted near-end acoustic signal, wherein the corrupted near-end acoustic signal is generated based on (1) an echo of the far-end acoustic signal and (2) a near-end acoustic signal; feeding the far-end acoustic signal and the corrupted near-end acoustic signal into a neural network as an input to output a time-frequency (TF) mask that suppresses the echo and retains the near-end acoustic signal, and generating an enhanced version of the corrupted near-end acoustic signal by applying the obtained TF mask to the corrupted near-end acoustic signal.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Li, Shuaiji
Tang, Xiaocheng
Qin, Zhiwei
Abrégé
A method includes: obtaining a plurality of first signals corresponding to a vehicle and a plurality of second signals corresponding to a plurality of candidate carpool combinations each comprising one or more unassigned transportation orders, wherein: the plurality of first signals comprise a current time, a location of the vehicle at the current time, and one or more static features corresponding to the vehicle, the plurality of second signals comprise timestamps, origins, and destinations of the unassigned transportation orders, and the vehicle has an on-going transportation order at the current time; inputting the plurality of first and second signals to a trained machine learning model; and obtaining, from an output of the trained machine learning model, a utility score of each of the plurality of candidate carpool combinations.
G06Q 10/06 - Ressources, gestion de tâches, des ressources humaines ou de projets; Planification d’entreprise ou d’organisation; Modélisation d’entreprise ou d’organisation
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Han, Shaoying
Abrégé
A floating window control method and system. The method comprises: obtaining a first operation of a user on a terminal device, the first operation being used for instructing the terminal device to execute a first application program in the background; when the first application program is executed in the background, automatically adding a first floating window on a display interface of the terminal device, the first floating window comprising a first state and a second state, when the first floating window is in the first state, the first floating window being hidden in the display interface, and when the first floating window is in the second state, the first floating window being displayed on the display interface; and after the first floating window is added, adding a second floating window associated with the first application program, the second floating window covering a local area of the display interface, and when the relative position relationship between the second floating window and the first floating window meeting a first preset condition, the first floating window being switched from the first state to the second state, wherein the first floating window is used for closing the second floating window.
G06F 3/0481 - Techniques d’interaction fondées sur les interfaces utilisateur graphiques [GUI] fondées sur des propriétés spécifiques de l’objet d’interaction affiché ou sur un environnement basé sur les métaphores, p.ex. interaction avec des éléments du bureau telles les fenêtres ou les icônes, ou avec l’aide d’un curseur changeant de comport
43.
METHOD AND SYSTEM FOR RECOGNIZING CERTIFICATE ON BASIS OF GRAPH NEURAL NETWORK
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Hao
Zhang, Tianming
Wang, Zhiheng
Wang, Shudong
Xue, Taolue
Zhou, Shiqi
Cheng, Bo
Bi, Xiao
Abrégé
A method and system for recognizing a certificate on the basis of a graph neural network. The method comprises: obtaining an image to be recognized (310); detecting content text comprised in the image to be recognized, and determining multiple bounding boxes (320); constructing a layout on the basis of the multiple bounding boxes (330), wherein the layout comprises a plurality of nodes and a plurality of edges, each node corresponds to a bounding box, and each edge correspond to the spatial position relationship between a bounding box and another bounding box; and processing the layout using a trained graph neural network model to determine field categories in the bounding boxes in the layout (340), and performing text recognition on key fields, so as to extract key information in the certificate.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Shen, Chao
Abrégé
A method and system for reminding a user. The method comprises: acquiring positioning information of a user terminal, wherein the positioning information reflects information of the current road section of a vehicle in which a user is located (310); when the current road section is a restricted parking road section, determining a stay duration of the vehicle on the current road section (320); determining whether the stay duration is greater than a first safe stay duration (330); and in response to the stay duration being greater than the first safe stay duration, sending first-level reminder information to the user terminal (340).
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Yang, Jing
Guan, Wei
Abrégé
Embodiments of the disclosure provide systems and methods for detecting abnormal charging events. The system (100) may include a communication interface (202) configured to receive multivariant charging data of a plurality of charging events (702, 703, 704, 705), the charging data of each charging event (702, 703, 704, 705) comprising a plurality of variants each corresponding to a charging characteristic. The system (100) may further include at least one processor (204). The at least one processor (204) may be configured to determine differences between every two charging events (702, 703, 704, 705) based on the multivariant charging data of the two charging events (702, 703, 704, 705). The at least one processor (204) may be further configured to cluster the plurality of charging events (702, 703, 704, 705) based on the determined differences. The at least one processor (204) may be also configured to detect an abnormal charging event based on the clustering results.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Xudong
Abrégé
An information exchange method and apparatus, a storage medium, and an electronic device. After receiving task information of a task to be processed, a task processing terminal sends its own position. A task issuing terminal controls a display apparatus to render and display a position modification control corresponding to the task and, in response to the control being triggered, displays a modification page to detect an input operation. When a specific input operation is detected, the task issuing terminal determines corresponding position information and sends to a server position modification information determined on the basis of the position information. Thus, the server acquires the task information and, when the task processing state of the task is unprocessed, sends to the task processing terminal a position modification prompt and the position modification information, such that the task processing terminal modifies the starting position of the task on the basis of the position information. The possibility of tasks being cancelled is reduced and the rate of utilisation of task processing resources is increased.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Jiang, Dongwei
Li, Xiangang
Abrégé
A voice recognition method and apparatus, and a computation device and a storage medium. The method comprises: receiving a voice sequence by means of a voice recognition engine, the voice sequence comprising multiple voice frames (205); and coding the voice sequence by means of the following operations by a voice coding engine in the voice recognition engine so as to update the voice recognition engine: masking voice frames behind one voice frame among the multiple voice frames of the voice sequence, and coding said voice frame on the basis of voice frames in front of said voice frame (210). In addition, when training the voice recognition engine, an unlabeled voice sequence is used to pre-train the voice coding engine, and a labeled voice sequence is used to update the voice recognition engine. The method can efficiently achieve voice streaming recognition.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Song, Xueqing
Ma, Ning
Abrégé
Disclosed in embodiments of the present description are a data processing method and apparatus, a readable storage medium, and an electronic device. In the embodiments of the present invention, the method comprises: acquiring a first image; receiving a text fill-in request, the text fill-in request being used to fill in text content in the first image on an interaction interface; displaying the location of the text content in a first example image, the first example image being an example of an image category to which the first image belongs; and uploading the filled in text content in response to receiving a text fill-in confirmation instruction, the text fill-in confirmation instruction being used to complete the filling in of the text content on the interaction interface.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
49.
IMAGE AUTHENTICITY DETERMINATION METHOD AND SYSTEM
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Zhiheng
Zhang, Tianming
Zhang, Chunwei
Xue, Taolue
Zhou, Shiqi
Wang, Hao
Jing, Haipeng
Zhao, Ningning
Zhang, Mingwen
Cheng, Bo
Feng, Mao
Meng, Hui
Zhang, Yuanyou
Abrégé
Disclosed in embodiments of the present description is an image authenticity determination method and system. The method is applied to a server, and the method comprises: obtaining an original image from a client; extracting, according to a preset extraction rule, multiple images or image parts from the original image as multiple extracted images; determining, by means of a trained machine learning model on the basis of the multiple extracted images, the degree of matching between the multiple extracted images and a preset sequence; and determining the authenticity of the original image on the basis of the degree of matching, the preset sequence corresponding to a camera device of the client, and determining the authenticity of the original image specifically comprising: determining the authenticity of an image from the camera device.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
G06K 9/62 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
H04L 9/32 - Dispositions pour les communications secrètes ou protégées; Protocoles réseaux de sécurité comprenant des moyens pour vérifier l'identité ou l'autorisation d'un utilisateur du système
50.
INFORMATION INTERACTION METHOD AND APPARATUS, AND ELECTRONIC DEVICE AND READABLE STORAGE MEDIUM
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Qiu, Linxiang
Chen, Qianhong
Abrégé
An information interaction method and apparatus, and an electronic device and a readable storage medium, which relate to the technical field of computers. The method comprises: in response to monitoring that an expected embarking point is within a preset coordinate range, presenting at least one embarking area within the preset coordinate range (100), wherein the expected embarking point is a map coordinate point; receiving a selection instruction for an embarking area, and determining a target embarking area (200); determining a target embarking point corresponding to the target embarking area (300); in response to receiving an order issuing instruction for the target embarking point, determining a recommended route between the current location and the target embarking point (400); and presenting route guidance information on the basis of the recommended route (500).
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Qin, Zhiwei
Jiao, Yan
Tang, Xiaocheng
Zhu, Hongtu
Ye, Jieping
Abrégé
Deep reinforcement learning may be used for vehicle (150) repositioning on mobility-on-demand platforms. Information may be obtained. The information may include a current location of a vehicle (150) on a ride-sharing platform. A set of paths originated from the current location of the vehicle (150) may be obtained. Each of the set of paths may have a length less than a preset maximum path length. A set of expected cumulative rewards along the set of paths may be obtained based on a trained deep value-network. A best path from the set of paths may be selected based on a heuristic tree search of the set of expected cumulative rewards. A next step along the best path may be recommended as a reposition action for the vehicle (150).
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Dai, Feng
Liu, Haijiang
Huang, Zhixin
Abrégé
A system (100) and a method (600) for adaptively charging a battery system (120) including a plurality of battery cells, the method (600) including: receiving a present value of at least one functioning metric indicative of a present status of each battery cell (S602); identifying two discrete values of the at least one functioning metric from a lookup table for each battery cell based on the present value of the at least one functioning metric of the respective battery cell (S606); determining a present value of the charging parameter for the respective battery cell based on values of the charging parameter corresponding to the two identified discrete values in the lookup table (S608); causing the battery system to be charged based on the present values of the charging parameter determined for the respective battery cells (S612).
H02J 7/00 - Circuits pour la charge ou la dépolarisation des batteries ou pour alimenter des charges par des batteries
B60L 53/00 - PROPULSION DES VÉHICULES À TRACTION ÉLECTRIQUE; FOURNITURE DE L'ÉNERGIE ÉLECTRIQUE À L'ÉQUIPEMENT AUXILIAIRE DES VÉHICULES À TRACTION ÉLECTRIQUE; SYSTÈMES DE FREINS ÉLECTRODYNAMIQUES POUR VÉHICULES, EN GÉNÉRAL; SUSPENSION OU LÉVITATION MAGNÉTIQUES POUR VÉHICULES; CONTRÔLE DES PARAMÈTRES DE FONCTIONNEMENT DES VÉHICULES À TRACTION ÉLECTRIQUE; DISPOSITIFS ÉLECTRIQUES DE SÉCURITÉ POUR VÉHICULES À TRACTION ÉLECTRIQUE Échange d'éléments d’emmagasinage d'énergie dans les véhicules électriques
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Tihui
Cao, Yuchu
Abrégé
An information display method, a first service request terminal, a system and an electronic device. A service request position of a first service requester is acquired when a carpooling request from the first service requester is received (S201); dynamic carpooling information during the carpooling process of a second service requester within a preset distance range of the service request position of the first service requester is determined (S202); the dynamic carpooling information of the second service requester is displayed on a display interface of the first service request terminal (S203). Thus, during the time period when the carpooling request is awaiting a response, the dynamic carpooling information of the second service requester within the preset distance range of the service request position of the first service requester is displayed to the first service requester. Display content on the terminal can be effectively enriched, transparency of related information is enhanced, filling the idle waiting time for the user, and the probability of order cancellation can be reduced.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Dai, Feng
Abrégé
A system and method for safety monitoring for an electric vehicle are disclosed. The method may include obtaining parameter values of one or more motion parameters associated with a target object (510). The method may also include identifying an abnormal status of the target object based on the parameter values (520). The method may further include obtaining property values of one or more performance properties of a power supply unit of the target object in response to identifying the abnormal status of the target object (530), wherein the power supply unit provides power for the target object. The method may still further include determining a status of the power supply unit based at least in part on the property values (540).
B60L 3/00 - Dispositifs électriques de sécurité sur véhicules propulsés électriquement; Contrôle des paramètres de fonctionnement, p.ex. de la vitesse, de la décélération ou de la consommation d’énergie
55.
METHOD AND SYSTEM FOR RECOMMENDING PICK-UP LOCATION
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Yang, Jiantao
Xiong, Ting
Chen, Wangting
Ma, Li
Abrégé
A method for recommending a pick-up location. The method comprises: acquiring a current location of a user; generating, on the basis of the current location of the user, one or more pieces of candidate pick-up location information; displaying, to the user, the one or more pieces of candidate pick-up location information; receiving one of the one or more pieces of candidate pick-up location information selected by the user; and displaying a recommended pick-up location to the user on the basis of the candidate pick-up location information selected by the user. In the method, recommendation of a pick-up location to a user is made on the basis of a current location of the user and a candidate pick-up location selected by the user, such that the user only needs to make a one-tap selection or perform other simple selection operations to obtain an appropriate recommended pick-up location without having to enter a search term. The invention can provide a more accurate recommended pick-up location conveniently accessible to the user, and save time for the user, thereby improving user experience.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Ye, Cichang
Wang, Liqun
Li, Yiyan
Abrégé
A method for providing a real-scene image for a user. The method comprises: obtaining at least one candidate real-scene point (A1, A2, A3) on the basis of a target position (L1) (510); determining, on the basis of the target position (L1), a movement direction of a user, and the at least one candidate real-scene point (A1, A2, A3), whether a preset condition is satisfied (520); in response to that the preset condition is satisfied, determining a target real-scene point from the candidate real-scene point (A1, A2, A3) corresponding to the fact that the preset condition is satisfied (530); and determining, on the basis of the target real-scene point, a real-scene image corresponding to the target position (L1), and displaying the real-scene image on a navigation interface related to the user (540). By displaying on the navigation interface of the user the real-scene image of the vicinity of the target position (L1), the accuracy of recognizing the target position (L1) is improved, and thus the task processing efficiency can be improved.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Ye, Cichang
Wang, Liqun
Sun, Beijia
Li, Yiyan
Abrégé
A method for providing a real-scene image for a user. The method comprises: on the basis of candidate positions, determining at least one first real-scene image corresponding to the candidate positions, and displaying the at least one first real-scene image on a navigation interface related to a user (510); in response to a confirmation operation of the user for one or more of the at least one first real-scene image, determining a candidate position, which corresponds to a confirmed first real-scene image, to be a target position (520); determining a target real-scene point at least on the basis of the target position (530); and at least on the basis of the target real-scene point, determining a second real-scene image corresponding to the target position, and displaying the second real-scene image on the navigation interface (540). According to the embodiments, a real-scene image of a target position is displayed on a navigation interface, such that the accuracy of a user in identifying the target position can be improved.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Che, Zhengping
Wu, Kun
Jiang, Bo
Yin, Chengxiang
Tang, Jian
Abrégé
Embodiments of the disclosure provide systems and methods for performing motion transfer using a learning model. An exemplary system may include a communication interface configured to receive a first image including a first movable object and a second image including a second movable object. The system may also include at least one processor coupled to the communication interface. The at least one processor may be configured to extract a first set of motion features of the first movable object from the first image using a first encoder of the learning model and extract a first set of static features of the second movable object from the second image using a second encoder of the learning model. The at least one processor may also be configured to generate a third image by synthesizing the first set of motion features and the first set of static features.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhou, Fan
Tang, Xiaocheng
Qin, Zhiwei
Zhang, Fan
Zhu, Hongtu
Abrégé
Multi-objective distributional reinforcement learning may be applied to order dispatching on ride-hailing platforms. A set of historical driver trajectories and a set of driver-order pairs may be obtained. A weight vector between a first reward of the set of historical driver trajectories and a second reward of the set of historical driver trajectories may be determined using inverse reinforcement learning (IRL). A first value function and a second value function may be jointly learned using distributional reinforcement learning (DRL) based on the historical driver trajectories and the weight vector. A set of scores comprising a score of each driver-order pair in the set of driver-order pairs may be determined based on the weight vector, the first value function, and the second value function. A set of dispatch decisions may be determined based on the set of scores that maximizes a total reward of the set of dispatch decisions.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Xia, Xing
Li, Guozhen
Gui, Wenbin
Abrégé
An image processing method, an electronic device, a vehicle traveling data recorder, and a server. The method comprises: training, according to a plurality of groups of image training samples, a machine learning model to obtain an image processing model, wherein each group of the image training samples comprises: an input feature parameter comprising a sample image in an abnormal state and an output feature parameter comprising a label for identifying the abnormal state (S101); and inputting an image acquired by an image sensor into the image processing model, and determining the abnormal state of the image according to an output result of the image processing model (S102). In the solution, an identification mode of an abnormal state of an image is easily implemented, is strong in repeatability, can be applied to image sensors in various scenarios, and has a very high practicability.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Teng, Xiaoqiang
Xu, Pengfei
Ma, Chaowei
Xu, Bin
Zhang, Jun
Meng, Yiping
Hu, Runbo
Chai, Hua
Abrégé
A new approach for identifying a stepping event is proposed in indoor navigation. The approach involves the following steps: obtaining a first signal fragment and a second signal fragment respectively within a first time window and a second time window in an acceleration signal stream, wherein the acceleration signal stream is collected from an acceleration sensor associated with a moving user, the first time window being shorter than the second time window (210); determining a first amplitude feature and a second amplitude feature for the first and second time windows based on the first and second signal fragments, respectively (220); identifying a stepping event of the user based on a deviation between the first and second amplitude features (230). With the above approach, the stepping event is identified in a more effective an accurate way, and thus accuracy of the indoor navigation is increased.
G01C 22/00 - Mesure de la distance parcourue sur le sol par des véhicules, des personnes, des animaux ou autres corps solides en mouvement, p.ex. en utilisant des odomètres ou en utilisant des podomètres
G06K 9/62 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
G01C 21/16 - Navigation; Instruments de navigation non prévus dans les groupes en utilisant des mesures de la vitesse ou de l'accélération exécutées à bord de l'objet navigant; Navigation à l'estime en intégrant l'accélération ou la vitesse, c. à d. navigation par inertie
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Teng, Xiaoqiang
Xu, Pengfei
Ma, Chaowei
Xu, Bin
Zhang, Jun
Meng, Yiping
Hu, Runbo
Chai, Hua
Abrégé
A new approach for determining a movement orientation of a user is proposed in indoor navigation. The approach involves: obtaining a device orientation of a terminal device based on at least one signal stream collected from the terminal device carried by a moving user (1110); determining a deviation degree based on the at least one signal stream, wherein the deviation degree represents a deviation between a movement orientation of the user and an actual device orientation of the terminal device (1120); determining the movement orientation based on the device orientation in accordance with a determination that the deviation degree is below a threshold degree (1130). With the approach, the movement orientation of the user is determined in a more effective an accurate way, and thus accuracy of the indoor navigation is increased.
G01P 13/02 - Indication de la direction uniquement, p.ex. par une girouette
G01P 21/00 - Essai ou étalonnage d'appareils ou de dispositifs couverts par les autres groupes de la présente sous-classe
G01C 22/00 - Mesure de la distance parcourue sur le sol par des véhicules, des personnes, des animaux ou autres corps solides en mouvement, p.ex. en utilisant des odomètres ou en utilisant des podomètres
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Xu, Yangyang
Wei, Xiaoying
Abrégé
An electric vehicle charging method may include receiving a charging request for charging an electric vehicle; sending a charging cost determination model and a charging instruction to a charger (150) that charges the electric vehicle, wherein the charging cost determination model is customized based by a user profile associated with the electric vehicle; determining a first cost based on the charging cost determination model; obtaining a second cost from the charger (150), wherein the second cost is determined by the charger (150) based on the charging cost determination model; and determining a final charging cost of the electric vehicle based on the first cost and the second cost.
G06Q 10/04 - Prévision ou optimisation spécialement adaptées à des fins administratives ou de gestion, p. ex. programmation linéaire ou "problème d’optimisation des stocks"
G06Q 50/06 - Fourniture d'électricité, de gaz ou d'eau
G06Q 20/14 - Architectures de paiement spécialement adaptées aux systèmes de facturation
64.
SYSTEMS AND METHODS FOR ONLINE TO OFFLINE SERVICES
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Cheng
Abrégé
Systems and methods for online to offline services are provided. The methods may include obtaining a first service request issued by a first service requester, the first service request including a request time point, a start location, and a destination. The methods may further include determining a target time point based on at least one of the request time point or an input by the first service requester, wherein the target time point is after the request time point. The methods may further include obtaining one or more candidate service requests, and determining a service request set and a target service provider that matches with the service request set by executing, starting at the target time point, a matching process based on the first service request and the one or more candidate service requests. The service request set may include the first service request.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Shang, Wenjie
Li, Qingyang
Qin, Zhiwei
Ye, Jieping
Yu, Yang
Meng, Yiping
Abrégé
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for constructing a virtual environment for a ride-hailing platform are disclosed. An exemplary method comprises: obtaining a plurality of historical interaction trajectories each comprising one or more interaction records between a driver and a ride-hailing platform, each interaction record comprising a program recommendation of the ride-hailing platform to the driver and a reaction of the driver in response to the program recommendation; training a simulator based on the plurality of historical interaction trajectories; and integrating a reward function with the simulator to construct the virtual environment, wherein the plurality of first program recommendations and the plurality of reactions form a plurality of simulated interactions, and a data distribution of the plurality of simulated interactions approximates a data distribution of a plurality of interaction records in the plurality of historical interaction trajectories.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Xu, Jixing
Abrégé
A method for object evaluation is provided. The method may include for each of one or more first virtual edges associated with a plurality of objects, obtaining first edge feature information of the first virtual edge. The first virtual edge may connect two objects of the plurality of objects. The first edge feature information of the first virtual edge may include feature information of an entity group that includes one or more entities associated with the two objects. The method may also include determining an object representation vector of at least one of the objects based on object feature information of the objects and the first edge feature information of the first virtual edges by applying an object representation model. The method may further include generating an evaluation result with respect to the at least one object based on the object representation vector by applying an object evaluation model.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Minkang
Wang, Fei
Abrégé
A high-precision map building method (500) and system (100, 400). The high-precision map building method (500) comprises: building local maps, and obtaining a global map on the basis of a plurality of local maps, wherein the building of each local map comprises: acquiring pose data of a vehicle and point cloud data related to a laser radar (510); obtaining a local point cloud set on the basis of the pose data and the point cloud data (520); performing registration and optimization processing on the basis of the local point cloud set, so as to obtain an optimized pose node and a local point cloud set corresponding to the optimized pose node (530); and generating the local map on the basis of the optimized pose node and the local point cloud set corresponding to the optimized pose node (540). A task of building a global map is divided into a plurality of subtasks of building local maps, and then, a plurality of instances of distributed processing are performed on a plurality of sub-steps of each subtask, such that the building of a high-precision map can be quickly completed, thereby improving efficiency and reducing cost.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Chen, Gong
Abrégé
Systems and methods for health evaluation of a storage medium are provided. The method may include obtaining monitoring data of a storage medium. The monitoring data may be obtained from one or more registers of the storage medium and indicate a current running status of the storage medium. The method may also include evaluating, based on the monitoring data of the storage medium, a health status of the storage medium to obtain an evaluation result.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Minkang
Wang, Fei
Abrégé
A system (100) and method for determining a two-dimensional map. The method may include obtaining a pose graph of a region. The pose graph may include a plurality of route-tracking datasets in the region. Each route-tracking dataset may include a plurality of nodes (710, 720). The method may also include determining at least one overlapped sub-region included in the region based on the plurality of route-tracking datasets. At least two of the plurality of route-tracking datasets may overlap at one of the at least one overlapped sub-region and the at least two of the plurality of route-tracking datasets may satisfy a condition. Further, the method may include determining a plurality of layers of roads in the region based on the at least one overlapped sub-region. The method may also include determining a target map based on the plurality of layers of roads and the pose graph.
G01C 21/32 - Structuration ou formatage de données cartographiques
G06F 17/00 - TRAITEMENT ÉLECTRIQUE DE DONNÉES NUMÉRIQUES Équipement ou méthodes de traitement de données ou de calcul numérique, spécialement adaptés à des fonctions spécifiques
70.
WRITE OPERATION IN OBJECT STORAGE SYSTEM USING ENHANCED META STRUCTURE
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Li
Zhang, Yiming
Shi, Mingya
Huang, Chunhua
Abrégé
Embodiments of the disclosure provide systems and methods for performing a write operation of an object. The exemplary method includes receiving, by a proxy server, a request for the write operation and the object comprising application meta-data and object data, and sending, by the proxy server, the application meta-data to a meta server. The method further includes generating, by the meta sever, allocation data including placement meta-data recording disk volumes on a data server and positions within the disk volumes allocated to the object. The method also includes atomically writing, by the meta server, the application meta-data, the allocation data, and meta-logs of the request to a local store of the meta server. The method additional includes writing, by the data server, the object data to the allocated disk volumes at the allocated positions according to the allocation data.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Li
Zhang, Yiming
Shi, Mingya
Huang, Chunhua
Abrégé
Embodiments of the disclosure provide systems and methods for performing a delete operation of an object. The exemplary method includes receiving, by a proxy server, a request for the delete operation, the request including an identification of the object. The method further includes sending, by the proxy server, a meta-data query to a primary meta server, the meta-data query including the identification of the object. The method also includes atomically performing, by the primary meta-server, update operations to meta-data associated with the identification of the object. The meta-data is stored in a local store of the primary meta server and includes allocation data of the object. The method additionally includes sending, by the primary meta server, the update operations to at least one backup meta server, performing, by the backup meta server, the update operations to corresponding backup meta-data stored in a local store of the backup meta server.
G06F 16/38 - Recherche caractérisée par l’utilisation de métadonnées, p.ex. de métadonnées ne provenant pas du contenu ou de métadonnées générées manuellement
72.
DATA SERVER CRASH RECOVERY IN OBJECT STORAGE SYSTEM USING ENHANCED META STRUCTURE
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Li
Zhang, Yiming
Shi, Mingya
Huang, Chunhua
Abrégé
Embodiments of the disclosure provide systems and methods for recovering a data server crash. The exemplary method includes detecting a crash of a first data server and marking an affected volume group associated with the first data server read-only. The method further includes determining that the first data server is not back online within a predetermined time. The method also includes replacing the first data server with a second data server and recovering object data of the affected volume group to the second data server.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Li
Zhang, Yiming
Shi, Mingya
Huang, Chunhua
Abrégé
Embodiments of the disclosure provide systems and methods for performing a read operation of an object. The exemplary method includes receiving, by a proxy server, a request for the read operation, the request including an identification of the object. The method further includes requesting, by the proxy server, meta-data associated with the identification of the object from a meta server. The method also includes receiving, by the proxy server, the requested meta-data from the meta server, the meta-data including allocation data indicating disk volumes on a data server and positions within the disk volumes allocated to the object. The method additionally includes requesting, by the proxy server, object data from the data server according to the allocation data, and receiving, by the proxy server, the object data stored at the allocated positions in the allocated disk volumes on the data server.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Li
Zhang, Yiming
Shi, Mingya
Huang, Chunhua
Abrégé
Embodiments of the disclosure provide systems and methods for recovering a meta server crash. The exemplary method includes detecting a crash of a first meta server, which is assigned to an object group. The method further includes determining that the first meta server is back online within a predetermined time. The method also includes retrieving local operation logs of the object group from a local store of the first meta server, determining missing operation logs by comparing the local operation logs and backup operation logs on a second meta server also assigned to the object group, and updating the local operation logs with the missing operation logs. The method additionally includes recovering additional meta-data associated with the object group based on the updated local operation logs.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Li
Zhang, Yiming
Shi, Mingya
Huang, Chunhua
Abrégé
Embodiments of the disclosure provide systems and methods for recovering a data server crash. The exemplary method includes detecting a crash of a proxy server. The method further includes looking up a request log in a meta server, the request log associated with a write request from the proxy server and looking up object meta-data associated with the write request in the meta server. The object meta-data includes allocation data. The method also includes obtaining object data from a data server according to the allocation data and validating the object data against the object meta-data. The method additionally includes deleting the request log associated with the write request from the meta server.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Li
Zhang, Yiming
Xu, Jiawei
Abrégé
Systems and methods for performing data storage in an object-based storage system that includes a first set of object storage devices (OSDs) are disclosed. An exemplary method may include expanding the object-based storage system by adding a second set of OSDs at a first time point. The method may also include receiving an object created at a second time point. The method may further include mapping the object to the OSDs of the expanded object-based storage system based on a chronological order between the first time point and the second time point.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhou, Tianyi
Yan, Xiaogang
Abrégé
The present disclosure relates to systems and methods for displaying a point of interest (POI). The methods may include determining a location of a terminal device; determining a target POI to be displayed based on the location of the terminal device; and displaying, via a user interface, a first window of the target POI including first information associated with the target POI, wherein the first window of the target POI is displayed with the target POI as an anchor point and does not overlap with a POI symbol of the target POI and a location symbol of the location of the terminal device.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Jiang, Bo
Shi, Yue
Che, Zhengping
Tang, Jian
Abrégé
A method for performing object detection in an image using a detection model, comprising: generating, by at least one processor, a first feature map of the image based on a base network of the detection model; generating, by the at least one processor, a second feature map by applying a feature fusion network of the detection model to the first feature map. The feature fusion network includes layers structured in lattice of a plurality of branches and a plurality of levels and layers of at least one level are dilated. The method may further include detecting an object within the image based on the second feature map.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
79.
SYSTEMS AND METHODS FOR EMOTION RECOGNITION IN USER-GENERATED VIDEO(UGV)
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Xu, Pengfei
Zhao, Sicheng
Ma, Yunsheng
Gu, Yang
Yang, Jufeng
Xing, Tengfei
Hu, Runbo
Chai, Hua
Abrégé
Disclosed herein are method and system for emotion recognition in a User-Generated Video (UGV). An exemplary method may include: receiving a UGV (S302); extracting visual representation (S304); estimating visual spatial attention and determining visual spatial attention weighted feature map (S306); estimating visual channel-wise attention and determining visual channel-wise attention weighted feature map (S308); estimating visual temporal attention and determining visual embedding based on the visual temporal attention (S310); extracting acoustic representation (S312); estimating acoustic temporal attention and determining audio embedding based on acoustic temporal attention (S314); concatenating visual embedding and audio embedding to construct semantic vector (S316); predicting emotion label based on semantic vector (S318).
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Shudong
Wang, Zhiheng
Xue, Taolue
Zhang, Tianming
Li, Jie
Meng, Hui
Chen, Tianyu
Dai, Guiting
Wu, Zhaohui
Zhou, Duoqing
Wang, Zhilong
Pei, Renwang
Liu, Xiaoyang
Peng, Kai
Yi, Junjie
Sun, Aiju
Jing, Haipeng
Wang, Xinyu
Abrégé
A method and system for verifying a vehicle. The method comprises: acquiring a vehicle image from a verified person (510), wherein the vehicle image comprises one or more types of vehicle information and vehicle environment background information; on the basis of one or more types of the vehicle information and the vehicle environment background information, determining qualification features of the vehicle image (520), wherein the qualification features of the vehicle image comprise feature data related to photography qualification of the vehicle image; and determining a vehicle verification result on the basis of the qualification features of the vehicle image (530).
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
G06K 9/62 - Méthodes ou dispositions pour la reconnaissance utilisant des moyens électroniques
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhao, Ningning
Zhang, Tianming
Guo, Yuanhao
Zhang, Xiubao
Shen, Haifeng
Abrégé
Systems and methods for image processing. The methods may include obtaining a first facial image (510); determining, based on a trained image quality determination model, a value of each of one or more quality features of the first facial image (520); determining, based on the one or more values of the one or more quality features, whether the first facial image satisfies a quality condition (530); obtaining a correction condition of the first facial image (540); obtaining a trained correction model, wherein the trained correction model includes a trained first generator (550); and in response to determining that the first facial image satisfies the quality condition, generating a second facial image that satisfies the correction condition by correcting the first facial image based on the trained correction model and the correction condition (560).
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
82.
SYSTEMS AND METHODS FOR VIDEO ANOMALY DETECTION USING MULTI-SCALE IMAGE FRAME PREDICTION NETWORK
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Che, Zhengping
Wang, Xuanzhao
Yang, Ke
Jiang, Bo
Tang, Jian
Abrégé
Systems, methods, and computer-readable media for video anomaly detection using a learning model. The method may include receiving a video including a plurality of image frames captured by a camera, the plurality of image frames including a current image frame corresponding to a selected time point and a set of prior image frames corresponding to time points prior to the selected time point. The method may also include extracting spatial features from each prior image frame at different resolutions and predicting an estimated image frame in each resolution based on the spatial features of the set of prior image frames in that resolution. The method may further include predicting an estimated current image frame based on the estimated image frames in the different resolutions and detecting the video anomaly based on a difference between the captured current image frame and the estimated current image frame.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Tian, Xiaoming
Zhao, Shuo
Abrégé
The method for monitoring an on-board equipment (10,130) includes: acquiring a relevant parameter of the on-board equipment, the relevant parameter reflecting the working state of the on-board equipment; determining an operation instruction information related to the on-board equipment based on the relevant parameter and a preset threshold. The system for monitoring an on-board equipment provides a sampling resistor (20) between the power supply of the on-board equipment and a processing unit (40) that actually implements the function of the on-board equipment. The voltage across the sampling resistor is measured in real time as a basis for determining the working current of the on-board equipment.
B60R 16/02 - Circuits électriques ou circuits de fluides spécialement adaptés aux véhicules et non prévus ailleurs; Agencement des éléments des circuits électriques ou des circuits de fluides spécialement adapté aux véhicules et non prévu ailleurs électriques
84.
LOW-VOLTAGE PROTECTION METHOD AND SYSTEM FOR VEHICLE-MOUNTED DEVICES
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Tian, Xiaoming
Zhao, Shuo
Abrégé
A low-voltage protection method and system for vehicle-mounted devices. The low-voltage protection method comprises: acquiring a target protection voltage of a vehicle storage battery at the current time, the vehicle storage battery supplying power to one or more vehicle-mounted devices (130) after the vehicle is powered off, and the value of the target protection voltage being dynamically updated over time; comparing the target protection voltage with the real-time voltage of the vehicle storage battery; and determining an operation instruction related to the one or more vehicle-mounted devices (130) according to the comparison result.
B60R 16/033 - Circuits électriques ou circuits de fluides spécialement adaptés aux véhicules et non prévus ailleurs; Agencement des éléments des circuits électriques ou des circuits de fluides spécialement adapté aux véhicules et non prévu ailleurs électriques pour l'alimentation des sous-systèmes du véhicule en énergie électrique caractérisé par l'utilisation de cellules électriques ou de batteries
H02H 7/12 - Circuits de protection de sécurité spécialement adaptés pour des machines ou appareils électriques de types particuliers ou pour la protection sectionnelle de systèmes de câble ou ligne, et effectuant une commutation automatique dans le cas d'un chan pour redresseurs pour convertisseurs ou redresseurs statiques
B60R 16/02 - Circuits électriques ou circuits de fluides spécialement adaptés aux véhicules et non prévus ailleurs; Agencement des éléments des circuits électriques ou des circuits de fluides spécialement adapté aux véhicules et non prévu ailleurs électriques
85.
SYSTEMS AND METHODS FOR DETECTING LANE MARKINGS FOR AUTONAMOUS DRIVING
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Wang, Shiying
Xia, Xuezhi
Abrégé
A system and method for detecting lane markings in a high-definition map is disclosed. An exemplary method includes receiving an image of the high-definition map that includes a road with at least one lane marking, and extracting road patches from the image, each road patch encompassing a section of the road. The method further includes transforming sub-images each containing a road patch to adjust an orientation of the road patch in the sub-image. The method also includes segmenting the transformed sub-images using a learning network and detecting the lane marking of the road based on the segmentation results of the transformed sub-images.
G06K 9/00 - Méthodes ou dispositions pour la lecture ou la reconnaissance de caractères imprimés ou écrits ou pour la reconnaissance de formes, p.ex. d'empreintes digitales
86.
EFFICIENT NETWORK-WIDE SIGNAL COORDINATION WITH MULTIPLE CYCLE LENGTHS AND TRAJECTORY DATA
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Lloret Batlle, Roger
Wang, Zihao
Liu, Minjun
Zheng, Jianfeng
Liu, Xianghong
Abrégé
Systems and methods for efficient network-wide signal coordination with multiple cycle lengths and trajectory data. A method may include determining a set of bandwidths for the traffic network based on green times of traffic lights in the traffic network (S710). The method may also include normalizing the set of bandwidths (S720). The method may further include weighting the normalized set of bandwidths (S730). The method may additionally include determining lengths of the green times that maximize a sum of the weighted normalized bandwidths subject to a plurality of predetermined constraints (S750). The method may also include controlling the green times of the traffic lights in the traffic network according to the determined lengths (S760).
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Sun, Jianwei
Zhao, Shuaijiang
Abrégé
Disclosed in an embodiment of the present invention is a method for man-machine interaction. The method for man-machine interaction comprises: extracting an associated feature of a target object on the basis of an interaction instruction directed at the target object, the associated feature being related to at least one of the interaction instruction and historical data of the target object; determining a response policy for the target object by processing the associated feature, the response policy being related to at least one of response content, a response style and a response emotion; and determining a response message for the target object on the basis of the response policy.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Hong, Huiting
Guo, Hantao
Yang, Xiaoqing
Li, Zang
Abrégé
A target task execution vehicle determination method and a system. The method comprises: receiving a task request comprising position information of a user (510); determining a road segment set on the basis of the position information, the road segment set comprising multiple road segments, and the multiple road segments comprising a target road segment and at least one associated road segment having a linkage relationship with the target road segment (520); acquiring road segment representation information corresponding to each of the multiple road segments in the road segment set (530), wherein the road segment representation information is generated by a road segment feature extraction model; determining, on the basis of the road segment representation information corresponding to each of the multiple road segments in the road segment set, at least one candidate road segment corresponding to the target road segment (540); and determining, from candidate vehicles located at the at least one candidate road segment, a target task execution vehicle corresponding to the task request (550).
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Li, Wei
Liu, Mu
Abrégé
A method for displaying information. The method comprises: displaying, on a screen display region, at least one calibration location of a journey (510); and displaying, on the screen display region, a target attribute corresponding to each calibration location in the at least one calibration location (520), displaying modes of the target attribute being related to a calibration display position of the at least one calibration location on the screen display region, wherein the displaying modes of the target attribute comprise: a displaying direction of the target attribute with respect to a corresponding calibration location, and a displaying distance of the target attribute with respect to the corresponding calibration location. The target attribute corresponding to the calibration location is displayed in the displaying modes of the target attribute, so that information shielding of the target attribute of at least one calibration location can be avoided, the integrity of information display is improved, and convenience is brought to a user to view the map when the user goes out.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Lyu, Ying
Han, Kun
Abrégé
In some cases, one or more heuristics can be automatically generated using a small dataset of segments previously labeled by one or more domain experts. The generated one or more heuristics along with one or more patterns can be used to assign training labels to a large unlabeled dataset of segments. A subset of segments representing an occurrence of verbal harassment can be selected using the assigned training labels. Randomly selected segments can be used as being indicative of a non-occurrence of verbal harassment. The selected subset of segments and randomly selected segments can be used to train one or more machine learning models for verbal harassment detection.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Lyu, Ying
Han, Kun
Abrégé
In some cases, lower quality, large scale training data can be automatically generated by automatic labeling. The generated training data can be used to pre-train a machine learning model. For instance, the model can be a model for detection of verbal harassment. Parameters of the pre-trained model can be refined or updated using another one or more higher-quality sets of training data, with which the model can be subsequently trained.
G10L 17/26 - Reconnaissance de caractéristiques spéciales de voix, p.ex. pour utilisation dans les détecteurs de mensonge; Reconnaissance des voix d’animaux
G10L 15/06 - Création de gabarits de référence; Entraînement des systèmes de reconnaissance de la parole, p.ex. adaptation aux caractéristiques de la voix du locuteur
G10L 15/10 - Classement ou recherche de la parole utilisant des mesures de distance ou de distorsion entre la parole inconnue et les gabarits de référence
G10L 15/26 - Systèmes de synthèse de texte à partir de la parole
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Lyu, Ying
Han, Kun
Abrégé
In some cases, a verbal harassment detection system may use machine learning models to detect verbal harassment in real-time or near real-time. The system may receive an audio segment comprising a portion of audio captured by a microphone located within a vehicle. Further, the system may convert the audio segment to a text segment. The system may provide at least the text segment to a prediction model associated with verbal harassment detection to obtain a harassment prediction. Further, the system may provide the audio segment to an emotion detector to obtain a detected emotion of a speaking user that made an utterance included in the audio segment. Based at least in part on the harassment prediction and the detected emotion, the system may automatically, and without user intervention, determine whether a user is being harassed.
G10L 17/26 - Reconnaissance de caractéristiques spéciales de voix, p.ex. pour utilisation dans les détecteurs de mensonge; Reconnaissance des voix d’animaux
G10L 15/26 - Systèmes de synthèse de texte à partir de la parole
G10L 25/63 - Techniques d'analyses de la parole ou de la voix qui ne se limitent pas à un seul des groupes spécialement adaptées pour un usage particulier pour comparaison ou différentiation pour estimer un état émotionnel
G10L 15/16 - Classement ou recherche de la parole utilisant des réseaux neuronaux artificiels
G10L 15/06 - Création de gabarits de référence; Entraînement des systèmes de reconnaissance de la parole, p.ex. adaptation aux caractéristiques de la voix du locuteur
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Lyu, Ying
Han, Kun
Abrégé
Certain aspects disclosed herein improve on inappropriate behavior detection by applying machine-learning techniques to identify inappropriate behavior in real-time, or near real-time. For example, one or more patterns can be generated that represent possible incidences of inappropriate behavior, such as a request for a user'scontact information. Audio segments captured by a wireless device in a vehicle can be converted into text using an automatic speech recognition system, and an adaptation model can apply the pattern(s) to the text to determine portions of the text that match at least one pattern and portions of the text that do not match any pattern, with the adaptation model labeling the text portions accordingly. The adaptation model can then pre-train a text classification model with the labeled text portions. The adaptation model obtains manually labeled data points and re-trains or updates the pre-trained text classification model using the manually labeled data points.
B60W 50/14 - Moyens d'information du conducteur, pour l'avertir ou provoquer son intervention
B60W 40/08 - Calcul ou estimation des paramètres de fonctionnement pour les systèmes d'aide à la conduite de véhicules routiers qui ne sont pas liés à la commande d'un sous-ensemble particulier liés aux conducteurs ou aux passagers
B60W 60/00 - Systèmes d’aide à la conduite spécialement adaptés aux véhicules routiers autonomes
B60W 50/00 - COMMANDE CONJUGUÉE DE PLUSIEURS SOUS-ENSEMBLES D'UN VÉHICULE, DE FONCTION OU DE TYPE DIFFÉRENTS; SYSTÈMES DE COMMANDE SPÉCIALEMENT ADAPTÉS AUX VÉHICULES HYBRIDES; SYSTÈMES D'AIDE À LA CONDUITE DE VÉHICULES ROUTIERS, NON LIÉS À LA COMMANDE D'UN SOUS-ENSEMBLE PARTICULIER - Détails des systèmes d'aide à la conduite des véhicules routiers qui ne sont pas liés à la commande d'un sous-ensemble particulier
94.
METHOD AND SYSTEM FOR IDENTIFYING ABNORMAL STAY OF VEHICLE
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Zhang, Tengjian
Chen, Ao
Abrégé
A method for identifying abnormal stay of a vehicle (150). The method comprises the following steps: obtaining real-time stay data of the vehicle (150) (501); and determining, on the basis of the real-time stay data, whether the vehicle (150) is in an abnormal stay state using an identification model (503), wherein the identification model comprises a machine learning model obtained by training historical stay data of a plurality of historical vehicles (150).
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Tang, Liang
Tan, Bo
Abrégé
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for bidding-based ridesharing are described. One exemplary method includes: obtaining, by a computing device of a ridesharing platform from a terminal device, a trip request for a rider, the trip request comprising an origin, a destination, and a price; identifying, by the computing device, a plurality of driver candidates to match with the trip request; determining, by the computing device, an acceptance probability for each of the plurality of driver candidates to accept the trip request; determining, by the computing device based on the plurality of determined acceptance probabilities, an estimated waiting time and an estimated matching probability for the trip request; and transmitting at least one of the estimated waiting time and the estimated matching probability to the terminal device.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Tang, Liang
Tan, Bo
Abrégé
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for recommending one or more alternative destinations are provided. One of the methods comprises: receiving, by a computing device of a ridesharing platform from a terminal device, a trip request that comprises an origin and a destination; classifying the trip request into one of the plurality of trip purpose categories based at least on the origin and the destination of the trip request, the rider's information, and a machine-learning classifier trained to predict the one trip purpose category of the trip request; in response to the one trip purpose category belonging to a preset group of trip purpose categories, determining one or more alternative destinations for the trip request; and sending to the terminal device, the one or more alternative destinations.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Dai, Feng
Abrégé
Embodiments of the disclosure provide systems and methods for performing heat management of a battery pack. An exemplary system may include a communication interface configured to receive from sensors thermal data associated with a plurality of cells of the battery pack. The system may also include at least one processor coupled to the communication interface. The at least one processor is configured to determine a first parameter indicative of a condition of each of the plurality of cells based on the thermal data. Moreover, the at least one processor is further configured to determine a first variation in the first parameters and adjust a flow direction of a temperature control substance applied to the battery pack based on a comparison between the first variation and a first predetermined threshold.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Yang, Lei
Abrégé
Embodiments of the disclosure provide systems and methods for performing a health evaluation of a battery system. An exemplary system may include a communication interface configured to receive operating data of the battery system acquired from one or more vehicles. The system may also include at least one processor coupled to the communication interface. The at least one processor is configured to determine a parameter indicative of a charging characteristic of the battery system based on the operating data. Moreover, the at least one processor is further configured to determine a trend of the parameter over a predetermined time period and evaluate a health of the battery system based on the trend of the parameter.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Liu, Minjun
Zheng, Jianfeng
Liu, Xianghong
Abrégé
A method for controlling traffic signals may include obtaining intersection information associated with a plurality of intersections of a route (510). The method may also include determining traffic parameters based on the intersection information and a traffic model (520). The traffic parameters may be used to control traffic signals of the plurality of intersections. The traffic model may be configured to take into account at least one of: (a) a spillover situation associated with congestion at red light of each of the plurality of intersections, (b) an arterial green bandwidth of the whole route in combination with a link green bandwidth between two of the plurality of intersections, and (c) a vehicle arrival distribution in the route.
BEIJING DIDI INFINITY TECHNOLOGY AND DEVELOPMENT CO., LTD. (Chine)
Inventeur(s)
Luan, Beiou
Abrégé
Provided are a lens calibration system (10) and method. The lens calibration system (10) comprises a lens non-reference measurement level device (100) and a lens angle adjustment device (200); the lens non-reference measurement level device (100) is used for measuring the pitch angle of the lens (102), and the lens angle adjustment device (200) is used for connecting with the lens (102) so as to adjust the pitch angle of the lens (102).
G01C 9/32 - Moyens pour faciliter l'observation de la position de la bulle, p.ex. moyens d'éclairage
F16M 11/04 - Moyens pour la fixation des appareils; Moyens permettant le réglage des appareils par rapport au banc
F16M 11/12 - Moyens pour la fixation des appareils; Moyens permettant le réglage des appareils par rapport au banc permettant la rotation dans plus d'une direction