Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining the likelihood that a particular event would occur during a navigation interaction using simulations generated by sampling from agent data. In one aspect, a method comprises: identifying an instance of a navigation interaction that includes an autonomous vehicle and agents navigating in an environment; generating multiple simulated interactions corresponding to the instance, comprising, for each simulated interaction: identifying one or more agents; for each identified agent and for each property that characterizes behavior of the identified agent, obtaining a probability distribution for the property; sampling a respective value from each of the probability distributions; and simulating the navigation interaction in accordance with the sampled values; and determining a likelihood that the particular event would occur during the navigation interaction based on whether the particular event occurred during each of the simulated interactions.

A detection method and a device based on a laser radar, and a computer readable storage medium are disclosed. The detection method includes: obtaining scanning data of the laser radar (S101); performing algorithm splitting on a feature algorithm for detection based on the scanning data to obtain at least one sub-algorithm capable of parallel processing in the feature algorithm (S102); and performing heterogeneous acceleration for the at least one sub-algorithm to process the scanning data and to obtain a processing result; and obtaining a detected position of an obstacle and a detected drivable area based on the processing result (S103).

Multiple trajectories for a vehicle are generated based on a road segment. Sensor data is received from at least one sensor. The vehicle is traveling the road segment in accordance with a first trajectory of the multiple trajectories. A potential collision is predicted between the vehicle and an object based on the sensor data and the first trajectory. A set of constraints is determined to avoid the potential collision. The set of constraints is determined based on the sensor data. A maneuver is determined for the vehicle by superimposing each constraint of the set of constraints on each other constraint of the set of constraints. The maneuver includes a second trajectory independent of the multiple trajectories. Instructions are transmitted to a control circuit of the vehicle to override the first trajectory and traverse the road segment according to the second trajectory to perform the maneuver.

The cloud includes a server and a storage device. The storage device includes a road surface information map. When a first sampling distance is equal to or longer than a first distance threshold, the server performs re-sampling to interpolate data in such a manner that sampling positions located at a second sampling distance and unsprung mass member displacements of the respective sampling positions exist so as to produce re-sampled data-for-producing-map. The server stores a sub-sectional unsprung mass displacement in a storage area corresponding to a sub-section of the road surface information map, based on the re-sampled data-for-producing-map.

An abnormality detection device includes: a time window process part configured to divide, by time windows, each of a plurality of vehicle signals each including a time-series data and to extract the data included in the time window; a learning part configured to create learned information by means of learning using the time-series data in the vehicle signal divided by the time windows; and an abnormality degree calculation part configured to reconstruct the time-series data in the vehicle signal divided by the time windows, based on the learned information, thereby calculate an error between before and after the reconstruction, and further calculate an abnormality degree of the vehicle signal based on the calculated error and the learned information.

An information processing method is provided to reduce an amount of data to be monitored in an onboard system of a vehicle. In the method, detection results that indicate whether an abnormality is included in communication data on an onboard network are obtained, and a first log transmission instruction is generated to cause periodic transmission of a first log from the onboard system to a server device. The first log is a log of some of the communication data. A second log transmission instruction is generated to cause transmission of a second log from the onboard system to the server device in a case of the detection results indicating the abnormality is included in the communication data. The second log is a log of more of the communication data than the first log.

The cloud includes a server and a storage device. The storage device includes a road surface information map. When a first sampling distance is equal to or longer than a first distance threshold, the server performs re-sampling to interpolate data in such a manner that sampling positions located at a second sampling distance and unsprung mass member displacements of the respective sampling positions exist so as to produce re-sampled data-for-producing-map. The server stores a sub-sectional unsprung mass displacement in a storage area corresponding to a sub-section of the road surface information map, based on the re-sampled data-for-producing-map.

A method and apparatus for planning a travelling path, and a vehicle are provided. The method includes: determining at least one reference curve covering a first length range, and selecting a target reference curve covering the first length range from the at least one reference curve covering the first length range; extracting a curve to be adjusted covering a second length range from the target reference curve covering the first length range; processing the curve to be adjusted based on a safety parameter within the second length range, to obtain an adjusted curve; and determining a travelling path covering the first length range based on the adjusted curve and the target reference curve. The complexity of an actual traffic scene is taken into account, and a travelling path planning is not affected by the accuracy of sampling points.

An apparatus of determining a position of a vehicle, may include a plurality of sensors to acquire raw data for vehicle information and surrounding information related to the vehicle, and a controller to generate a plurality of vehicle position point data based on the raw data, generate respective tracklets for the sensors by combining the plurality of vehicle position point data, fuse the tracklets for the sensors, and determine a final position of the vehicle using the fused tracklets for the sensors. The position is exactly estimated, and a computation amount is prevented from being excessively increased such that real-time position information is easily acquired

The present disclosure relates to systems and methods for determining a driving action in autonomous driving. The systems may obtain driving information associated with a vehicle; determine a state of the vehicle; determine one or more candidate driving actions and one or more evaluation values corresponding to the one or more candidate driving actions based on the driving information and the state of the vehicle by using a trained driving-action model; select a target driving action from the one or more candidate driving actions based on the one or more evaluation values; determine a target driving path based on the target driving action; and send signals to a control component of the vehicle to direct the vehicle to take the target driving action to follow the target driving path.