Abstract

The invention relates to a method of training a model for use in semantic image segmentation. The model includes a first classifier neural network and a second classifier neural network. The method includes training the first network by inputting a first training image containing a target object to the first network, using the first network to identify and erase pixels of the first training image that are discriminative for the target object, inputting the first training image to the second network, using the second network to determine a likelihood of the first modified training image containing the target object and updating weights of the first network using a first loss function that is a monotonically-increasing function of the determined likelihood.

IPC Classes  ?

• G06V 10/82 - Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks

Abstract

A method for providing data to one or more vehicles (130) for controlling respective automated driving systems (210) of the one or more vehicles (130), the method comprising: receiving an indication of a vehicle fleet; obtaining, from a restriction repository (119) storing restriction data (192) that indicates one or more location-dependent driving automation restrictions for automated driving systems, restriction data that corresponds to a portion of a road network and that is applicable to the indicated vehicle fleet; generating, or performing an update process for, a driving automation restriction layer (193) for an amount of map data that corresponds to the portion of the road network based on the obtained restriction data (192); and providing the amount of map data, with the driving automation restriction layer (193), to one or more vehicles (130) in the vehicle fleet.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G01C 21/00 - Navigation; Navigational instruments not provided for in groups
• G05D 1/02 - Control of position or course in two dimensions
• G08G 1/00 - Traffic control systems for road vehicles

Abstract

A computer processing system is configured to train a model for use in semantic image segmentation.The model comprises a refinement neural network, a discriminator neural network. The refinement neural network is configured to receive a predicted label distribution for an image, obtain one or more random values from a random or pseudo-random noise source, use the one or more random values to generate a plurality of predicted segmentation maps from the received predicted label distribution and output the plurality of predicted segmentation maps to the discriminator neural network. The computer processing system is configured to train the refinement neural network using an objective function that is a function of an output of the discriminator neural network and that further includes a term representative of a difference between the predicted label distribution and an average of the plurality of predicted segmentation maps output by the refinement neural network for the predicted label distribution.

IPC Classes  ?

• G06K 9/46 - Extraction of features or characteristics of the image
• G06K 9/62 - Methods or arrangements for recognition using electronic means
• G06K 9/32 - Aligning or centering of the image pick-up or image-field
• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

Abstract

Disclosed herein is a technique for the generating and provision of digital map data that is safe and reliable. The technique enables the verification of the digital map data in a map-client using a simple and efficient data structure to check the correctness of the map data before in-vehicle delivery to components that rely on this map data.

IPC Classes  ?

• G01C 21/00 - Navigation; Navigational instruments not provided for in groups
• G06F 16/23 - Updating
• H04L 9/32 - Arrangements for secret or secure communications; Network security protocols including means for verifying the identity or authority of a user of the system
• G06F 21/64 - Protecting data integrity, e.g. using checksums, certificates or signatures
• G06F 16/29 - Geographical information databases

Abstract

Disclosed herein is a technique for generating and providing an indication to an autonomous vehicle regarding the confidence level for the accuracy or quality of the map data in which the indication is determined from observation data received from other vehicles.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• B60W 60/00 - Drive control systems specially adapted for autonomous road vehicles
• G01C 21/00 - Navigation; Navigational instruments not provided for in groups

Abstract

Disclosed herein is a technique for generating map change data relative to the current map data and providing these changes to an autonomous vehicle for use in updating the current map data that it retrieves or obtains from a local storage. The technique processes observation data received from other vehicles to determine the map change data.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• B60W 60/00 - Drive control systems specially adapted for autonomous road vehicles
• G01C 21/00 - Navigation; Navigational instruments not provided for in groups

Abstract

Disclosed herein is a technique for processing map change data in a map client that uses the map changes to apply changes and updates to current map data. The map client provides the updated map data to other components in an autonomous vehicle. The map change data describes the processing in the map client to obtain updated map data from the map data.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• B60W 60/00 - Drive control systems specially adapted for autonomous road vehicles
• G01C 21/00 - Navigation; Navigational instruments not provided for in groups

Abstract

The present disclosure generally relates to a methods and systems for compensating for changes in the absolute position of locations with respect to the Earth's surface which occur over time due to crustal dynamics. The invention is particularly, although not exclusively, concerned with such compensation in the context of methods using digital map data e.g. methods of localization of a vehicle.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G01C 21/00 - Navigation; Navigational instruments not provided for in groups

Abstract

A method of training a generative adversarial network for performing semantic segmentation of images. The generative adversarial network includes a generator neural network and a critic neural network. The method includes using the generator neural network to generate predicted image segmentation maps from input images, wherein each predicted image segmentation map includes a classification prediction for each of a plurality of pixels of a respective input image, providing predicted image segmentation maps generated by the generator neural network to the critic neural network, training the critic neural network to determine weights for respective pixels of a predicted image segmentation map generated by the generator neural network, wherein the weight for each pixel is used to weight a pixel-wise cross entropy term in an objective function for the critic neural network, and using the weights determined by the critic neural network to train the generator neural network.

IPC Classes  ?

• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
• G06K 9/34 - Segmentation of touching or overlapping patterns in the image field
• G06K 9/62 - Methods or arrangements for recognition using electronic means
• G06N 3/04 - Architecture, e.g. interconnection topology

Abstract

There is provided a method for identifying possible errors/inconsistencies within an electronic map representation of a network of navigable elements within a geographic area, the method comprising: obtaining at a server positional data relating to the movement over time of a plurality of devices travelling around the navigable network; processing the obtained positional data at the server with reference to the electronic map representing the navigable network in order to identify potential inconsistencies in the map in the form of one or more locations within the navigable network where an observed behaviour of devices travelling around the navigable network as indicated by the obtained positional data is not consistent with a behaviour that would be expected based on the electronic map. This information may then be relayed to navigation devices for supplementing the electronic map when generating navigation instructions.

IPC Classes  ?

• G01C 21/00 - Navigation; Navigational instruments not provided for in groups
• G01C 21/36 - Input/output arrangements for on-board computers
• G08G 1/00 - Traffic control systems for road vehicles
• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

Abstract

121211'); and generating map data using the unified shape.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G06F 16/29 - Geographical information databases
• G09B 29/10 - Map spot or co-ordinate position indicators; Map-reading aids
• G06T 7/12 - Edge-based segmentation
• G01C 21/36 - Input/output arrangements for on-board computers
• G06K 9/34 - Segmentation of touching or overlapping patterns in the image field

Abstract

1 n1212121212 2 of the first region (205).

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G06F 8/65 - Updates
• G09B 29/00 - Maps; Plans; Charts; Diagrams, e.g. route diagrams

Abstract

A method of determining a position of a vehicle when traversing a road network involves obtaining, from a database, data indicative of one or more traffic signs expected to be encountered by the vehicle in the future while traversing the road network. The data indicative of each sign includes position and appearance data, and is associated with algorithm data to be used in analysing an image of a scene obtained from a camera associated with a vehicle to attempt to detect an image of the sign in the image of the scene obtained from the camera. A image is obtained from a camera associated with the vehicle while traversing the road network, and the appearance and algorithm data associated with the obtained data indicative of a traffic sign is used to analyse at least a portion of the image of a scene obtained from the camera associated with the vehicle to attempt to detect an image of the object in the image of the scene. The results of the analysis are used in determining a position of the vehicle.

IPC Classes  ?

• G01C 21/30 - Map- or contour-matching
• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
• G01S 19/48 - Determining position by combining or switching between position solutions derived from the satellite radio beacon positioning system and position solutions derived from a further system

Abstract

Traffic sign data including the location and type of traffic sign is input into a electronic map potential update tool together with an electronic map database. The type of the sign is indicative of an attribute associated with the sign. The tool uses the location and attribute data for the sign to identify whether a corresponding attribute is included in the electronic map database. Where the attribute is not present, the tool outputs data flagging the location of the sign. This output data is then used in updating the electronic map.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G09B 29/10 - Map spot or co-ordinate position indicators; Map-reading aids

Abstract

There is disclosed a method for providing information regarding traffic conditions within a road network in an area covered by an electronic map, the electronic map comprising a plurality of segments representing roads of the road network. The method involves obtaining traffic data from a traffic data source (200) having a known position relative to the road network. The position of the traffic data source (200) is then used to identify within the electronic map a set of roads falling within a predefined distance of the traffic data source (200). From this set, a determination is then made as to which of the road names in the electronic map best matches a road name included in the traffic data for defining the location of the traffic conditions. The traffic point location (204) representing the location of the traffic conditions within the electronic map can then be determined so that information indicative of the traffic conditions may be provided for inclusion into the map. Also disclosed are systems, particularly servers, for performing such methods.

IPC Classes  ?

• G08G 1/01 - Detecting movement of traffic to be counted or controlled
• G08G 1/0967 - Systems involving transmission of highway information, e.g. weather, speed limits
• G01C 21/36 - Input/output arrangements for on-board computers
• G08G 1/0968 - Systems involving transmission of navigation instructions to the vehicle

Abstract

A method is provided of training a generative adversarial network for performing semantic segmentation of images. The generative adversarial network includes a generator neural network and a discriminator neural network. The method includes providing an image as input to the generator neural network, receiving a predicted segmentation map for the image from the generator neural network, providing i) the image, ii) the predicted segmentation map, and iii) ground-truth label data corresponding to the image, as distinct training inputs to the discriminator neural network, determining a set of one or more outputs from the discriminator neural network in response to said training inputs, and training the generator neural network using a loss function that is a function of said set of outputs from the discriminator neural network.

IPC Classes  ?

• G06T 7/11 - Region-based segmentation

Abstract

A method and system for determining a geographical location and orientation of a vehicle (200, 300) travelling through a road network is disclosed. The method comprises obtaining, from one or more cameras (202, 302) associated with the vehicle (200, 300) travelling through the road network, a sequence of images (500) reflecting the environment of the road network on which the vehicle (200, 300) is travelling, wherein each of the images has an associated camera location at which the image was recorded. A local map representation representing an area of the road network on which the vehicle (200, 300) is travelling is then generated using at least some of the obtained images and the associated camera locations. The generated local map representation is compared with a section of a reference map, the reference map section covering the area of the road network on which the vehicle (200, 300) is travelling, and the geographical location and orientation of the vehicle (200, 300) within the road network is determined based on the comparison. Methods and systems for generating and/or updating an electronic map using data obtained by a vehicle (200, 300) travelling through a road network represented by the map are also disclosed.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
• G06F 17/30 - Information retrieval; Database structures therefor

Abstract

Methods and systems for classifying data points of a point cloud indicative of the environment around a vehicle by using features of a digital map relating to a deemed current position of the vehicle. Such methods and systems can be used to detect road actors, such as other vehicles, around a vehicle capable of sensing its environment as a point cloud; preferably used by highly and fully automated driving applications.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G01S 17/06 - Systems determining position data of a target
• G01S 19/46 - Determining position by combining measurements of signals from the satellite radio beacon positioning system with a supplementary measurement the supplementary measurement being of a radio-wave signal type
• G01S 7/48 - RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - Details of systems according to groups , , of systems according to group
• G01S 13/06 - Systems determining position data of a target
• G06T 3/00 - Geometric image transformation in the plane of the image
• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

Abstract

- 54 - ABSTRACT METHODS AND SYSTEMS FOR GENERATING AND USING LOCALISATION REFERENCE DATA 5 Methods and systems for improved positioning accuracy relative to a digital map are disclosed, and which are preferably used for highly and fully automated driving applications, and which may use localisation reference data associated with a digital map. The invention further extends to methods and systems for the generation of localisation reference data associated with a digital map. 10 [Fig. 30]

IPC Classes  ?

• G01C 21/30 - Map- or contour-matching
• G01S 17/89 - Lidar systems, specially adapted for specific applications for mapping or imaging
• G01C 21/28 - Navigation; Navigational instruments not provided for in groups specially adapted for navigation in a road network with correlation of data from several navigational instruments
• G01S 13/86 - Combinations of radar systems with non-radar systems, e.g. sonar, direction finder

Abstract

Methods and systems for improved positioning accuracy relative to a digital map are disclosed, and which are preferably used for highly and fully automated driving applications, and which may use localisation reference data associated with a digital map. A vehicle localization is obtained by comparing real time depth map acquired by a sensor associated to the vehicle to a depth map associated to a digital map. The depth maps are indicative of an environment around the vehicle. Longitudinal and lateral offsets of the vehicle with respect to the digital map are determined. The invention further extends to methods and systems for the generation of localisation reference data associated with a digital map.

IPC Classes  ?

• G01C 21/30 - Map- or contour-matching
• G01S 17/89 - Lidar systems, specially adapted for specific applications for mapping or imaging
• G01C 21/28 - Navigation; Navigational instruments not provided for in groups specially adapted for navigation in a road network with correlation of data from several navigational instruments
• G01S 13/86 - Combinations of radar systems with non-radar systems, e.g. sonar, direction finder

Abstract

Methods and systems for improved positioning accuracy relative to a digital map are disclosed, and which are preferably used for highly and fully automated driving applications, and which may use localisation reference data associated with a digital map. The invention further extends to methods and systems for the generation of localisation reference data associated with a digital map.

IPC Classes  ?

• G01C 21/30 - Map- or contour-matching
• G01S 17/89 - Lidar systems, specially adapted for specific applications for mapping or imaging
• G01C 21/28 - Navigation; Navigational instruments not provided for in groups specially adapted for navigation in a road network with correlation of data from several navigational instruments
• G01S 13/86 - Combinations of radar systems with non-radar systems, e.g. sonar, direction finder

Abstract

Methods and systems for improved positioning accuracy relative to a digital map are disclosed, and which are preferably used for highly and fully automated driving applications, and which may use localisation reference data associated with a digital map. The invention further extends to methods and systems for the generation of localisation reference data associated with a digital map.

IPC Classes  ?

• G01C 21/28 - Navigation; Navigational instruments not provided for in groups specially adapted for navigation in a road network with correlation of data from several navigational instruments
• G01C 21/30 - Map- or contour-matching
• G01S 17/89 - Lidar systems, specially adapted for specific applications for mapping or imaging
• G01S 13/86 - Combinations of radar systems with non-radar systems, e.g. sonar, direction finder

Abstract

Methods and systems for improved positioning accuracy relative to a digital map are disclosed, and which are preferably used for highly and fully automated driving applications, and which may use localisation reference data associated with a digital map. The invention further extends to methods and systems for the generation of localisation reference data associated with a digital map.

IPC Classes  ?

• G01C 21/30 - Map- or contour-matching
• G01S 17/89 - Lidar systems, specially adapted for specific applications for mapping or imaging
• G01C 21/28 - Navigation; Navigational instruments not provided for in groups specially adapted for navigation in a road network with correlation of data from several navigational instruments
• G01S 13/86 - Combinations of radar systems with non-radar systems, e.g. sonar, direction finder

Abstract

A method of determining a longitudinal position of a vehicle (100) relative to a digital map is disclosed in which real time scan data (200, 202) is determined by scanning a lateral environment around the vehicle using at least one range-finder sensor, said real time scan data comprising one or more depth maps, each depth map representing the measured lateral distance to surfaces in the lateral environment for a plurality of longitudinal positions and elevations. Localisation reference data associated with the digital map for a deemed current longitudinal position of the vehicle (100) in relation to the digital map is retrieved, and compared to the real time scan data (200, 202) by calculating a cross-correlation to determine a longitudinal offset. The deemed current longitudinal position is adjusted based on said longitudinal offset to determine the longitudinal position of the vehicle (100) relative to the digital map.

IPC Classes  ?

• G01C 21/28 - Navigation; Navigational instruments not provided for in groups specially adapted for navigation in a road network with correlation of data from several navigational instruments

Abstract

A method of detecting the closure of a navigable element forming part of a network of navigable elements within a geographic area. A server obtains positional data relating to the movement of a plurality of devices along the navigable element with respect to time. The positional data is used to determine an elapsed time since a device was last detected on the navigable element, and the determined elapsed time is compared to an expected time interval between consecutive devices being detected on the navigable element. The navigable element is identified as being potentially closed, subject to one or more optional validation steps, when the determined elapsed time exceeds the expected time interval, e.g. by a predetermined amount.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data

Abstract

A method is disclosed for incrementally updating an existing digital map of a client device in a digital map update system. The digital map update system comprising a primary digital map update centre, a secondary digital map update centre and a plurality of remote client devices. The primary digital map update centre and the secondary digital map update centre are each able to provide a set of one or more incremental updates for communication to one or more of one of the plurality of remote client devices for use in incrementally updating existing digital maps of the client devices. The secondary digital map update centre provides data indicative of a set of one or more incremental digital map updates for communication to at least one of the plurality of remote client devices for use in incrementally updating an existing digital map of the device. The secondary digital map update centre additionally provides information relating to the set of one or more incremental digital map updates to the primary digital map update centre.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G06Q 30/02 - Marketing; Price estimation or determination; Fundraising
• G06F 17/30 - Information retrieval; Database structures therefor

Abstract

Probe data collected at times of low traffic density is analyzed to derive a Raw Road Design Speed Limit (RRDSL, 16) for each road segment or group of segments in a digital map. The RRDSL (16), comprised of longitudinally distributed speeds, is associated with the road segment and stored in a digital medium to indicate the limits of the road section in free flow traffic. The longitudinally distributed speeds may be limited by local speed limits or other business logic to establish a Legal Raw Road Design Speed Limit (LRRDSL, 17). Either the RRDSL (16) or the LRRDSL (17) can be further modified to smooth acceleration and deceleration rates between changes in the longitudinally distributed speeds to create an Optimal Longitudinal Speed Profile (OLSP, 18), which represents optimized energy consumption. A signal can be produced if a driver's current speed rises unacceptably above a longitudinally distributed speed in real time. The signal can be audible, visible and/or haptic. Real-time traffic density information can be inferred by comparing current speed data to the longitudinally distributed speed for that position. If the current speed is consistently lower than the longitudinally distributed speed for that position, an inference is drawn that the road section is inefficient. Road efficiency assessments can be transmitted to a service center and/or other vehicles, and used by navigation software.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G06F 17/30 - Information retrieval; Database structures therefor
• B60W 30/14 - Cruise control

Abstract

Probe data is analyzed to derive Longitudinal Speed Profiles (LSPs) and an Optimal Longitudinal Speed Profile (18) for each road segment or link in a digital map network. The Longitudinal Speed Profiles (LSPs) profiles are calculated during defined time spans whereas the Optimal Longitudinal Speed Profile (18) is based on the LSP for the time span corresponding only to free flow traffic conditions. All of the LSPs can used to create a respective energy cost for each time span, or only the OLSP (18) can be used (or alternatively the RRDSL 16 or LRRDSL 17) to calculate an energy cost for the free flow conditions only. The energy cost can be used to predict the energy required by a vehicle to traverse the link. Navigation software can use the energy cost to plan the most energy efficient route between two locations in the digital map. Sensory signals can be activated if a driver strays from the Optimal Longitudinal Speed Profile (18) to achieve extremely high levels of energy efficiency.

IPC Classes  ?

• G01C 21/32 - Structuring or formatting of map data
• G06F 17/30 - Information retrieval; Database structures therefor
• B60W 30/14 - Cruise control