A sensor evaluation device evaluates a first sensor mounted on a sensor-mounting object. The sensor evaluation device is provided with a specific event detecting unit, and a recording unit. The specific event detecting unit detects a specific event which is at least one of (a) an unrecognized event where a second sensor mounted on the sensor-mounting object recognizes a first target whereas the first sensor does not recognize the first target, and (b) a misrecognized event where the first sensor recognizes a second target whereas the second sensor does not recognize the second target. The recording unit records information on the specific event when detecting the specific event.

Embodiments of this application provide a method and an apparatus for determining drivable region information. The method includes obtaining first information, where the first information includes information about an initial drivable region determined based on at least one image, and the at least one image is from at least one camera module. The method also includes obtaining second information, where the second information includes radar detection information. The method further includes determining first drivable region information based on the first information and the second information.

An end-to-end system for data generation, map creation using the generated data, and localization to the created map is disclosed. Mapstreams—or streams of sensor data, perception outputs from deep neural networks (DNNs), and/or relative trajectory data—corresponding to any number of drives by any number of vehicles may be generated and uploaded to the cloud. The mapstreams may be used to generate map data—and ultimately a fused high definition (HD) map—that represents data generated over a plurality of drives. When localizing to the fused HD map, individual localization results may be generated based on comparisons of real-time data from a sensor modality to map data corresponding to the same sensor modality. This process may be repeated for any number of sensor modalities and the results may be fused together to determine a final fused localization result.

A system can include a computer including a processor and a memory, the memory storing instructions executable by the processor to receive point cloud data. The instructions further include instructions to generate a plurality of feature maps based on the point cloud data, each feature map of the plurality of feature maps corresponding to a parameter of the point cloud data. The instructions further include instructions to aggregate the plurality of feature maps into an aggregated feature map. The instructions further include instructions to generate, via a feedforward neural network, at least one of a segmentation output or a classification output based on the aggregated feature map.

Vehicles and methods for tracking an object and controlling a vehicle based on the tracked object. A Radar-Doppler (RD) map is received from the radar sensing system of the vehicle and relative acceleration of an object with respect to the vehicle is detected based on the RD map so as to provide acceleration data. A current frame of detected object data is received from a sensing system of the vehicle. When the relative acceleration has been detected, a tracking algorithm is adapted to reduce the influence of the predictive motion model or the historical state of the object and the object is tracked using the adapted tracking algorithm so as to provide adapted estimated object data based on the object tracking. One or more vehicle actuators are controlled based on the adapted estimated object data.

To implement a configuration to calculate a manual driving recoverable time required for a driver who is executing automatic driving in order to achieve a requested recovery ratio (RRR) for each road section, and issue a manual driving recovery request notification on the basis of the calculated time. A data processing unit is included, which calculates a manual driving recoverable time required for a driver who is executing automatic driving in order to achieve a predefined requested recovery ratio (RRR) from automatic driving to manual driving and determines notification timing of a manual driving recovery request notification on the basis of the calculated time. The data processing unit acquires the requested recovery ratio (RRR) for each road section set as ancillary information of a local dynamic map (LDM), and calculates the manual driving recoverable time for each road section scheduled to travel, using learning data for each driver.

A driver assistance system for an ego vehicle, and a method for a driver assistance system is provided. The system is configured to refine a coarse geolocation method based on the detection of the static features located in the vicinity of the ego vehicle. The system performs at least one measurement of the visual appearance of each of at least one static feature located in the vicinity of the ego vehicle. Using the at least one measurement, a position of the ego vehicle relative to the static feature is calculated. The real world position of the static feature is identified. The position of the ego vehicle relative to the static feature is calculated, which is, in turn, used to calculate a static feature measurement of the vehicle location. The coarse geolocation measurement and the the static feature measurement are combined to form a fine geolocation position. By combining the measurements, a more accurate location of the ego vehicle can be determined.

Method for improving global positioning performance of a first road vehicle (10), the method comprising, by means of a data server (3, 4, 4″): acquiring data from onboard sensors (2a, 2b, 2c, 2d, 2e, 2f, 2g) arranged on the first road vehicle (10) and on at least two neighbouring road vehicles (10′, 10″, 10′″), the data comprising data on relative positions and data on heading angle and velocity of the road vehicles (10, 10′, 10″, 10′″), and acquiring global positioning data of at least two of the road vehicles (10, 10′, 10″, 10′″), processing (102) data comprising the global positioning data, the data, with corresponding timestamp, acquired from the onboard sensors (2a, 2b, 2c, 2d, 2e, 2f, 2g), and a motion model for each of the first road vehicle (10) and the at least two neighbouring road vehicles (10′, 10″, 10′″) using a data fusion algorithm, calculating adjusted global positioning data for the first road vehicle (10) and communicating (104) the adjusted global positioning data to a positioning system (6) of the first road vehicle (10).

A method of multi-sensor data fusion includes determining a plurality of first data sets using a plurality of sensors, each of the first data sets being associated with a respective one of a plurality of sensor coordinate systems, and each of the sensor coordinate systems being defined in dependence of a respective one of a plurality of mounting positions for the sensors; transforming the first data sets into a plurality of second data sets using a transformation rule, each of the second data sets being associated with a unified coordinate system, the unified coordinate system being defined in dependence of at least one predetermined reference point; and determining at least one fused data set by fusing the second data sets.