A vehicle includes a sensor circuit configured to detect an obstacle in a first region which is located on the predetermined traveling route and in a second region which is adjacent to the first region on the predetermined traveling route, the second region being farther than the first region. The vehicle enters the first region in a case where: there is no obstacle in the first region; and there is no obstacle in the second region, and does not enter the first region and stops before the first region in a case where: there is no obstacle in the first region; and there is an obstacle in the second region.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for planning the future trajectory of an autonomous vehicle in an environment. In one aspect, a method comprises obtaining multiple types of scene data characterizing a scene in an environment that includes an autonomous vehicle and multiple agents; receiving route data specifying an intended route for the autonomous vehicle; for each data type, processing at least the data type using a respective encoder network to generate a respective encoding of the data type; processing a sequence of the encodings using an encoder network to generate a respective alternative representation for each data type; and processing the alternative representations using a decoder network to generate a trajectory planning output that comprises respective scores for candidate trajectories that represent predicted likelihoods that the candidate trajectory is closest to resulting in the autonomous vehicle successfully navigating the intended route.

A method for predicting traffic light information by using a LIDAR is provided. The method includes steps of: (a) on condition that each of metadata has been allocated for each of virtual boxes included in a region covered by the LIDAR, obtaining, by a server, at least part of start timing information and stop timing information of a plurality of vehicles for each of the virtual boxes; and (b) predicting, by the server, each of pieces of the traffic light information respectively corresponding to each of the virtual boxes by referring to at least part of the start timing information and the stop timing information of the vehicles for each of the virtual boxes.

An autonomous vehicle uses machine learning based models such as neural networks to predict hidden context attributes associated with traffic entities. The hidden context represents behavior of the traffic entities in the traffic. The machine learning based model is configured to receive a video frame as input and output likelihoods of receiving user responses having particular ordinal values. The system uses a loss function based on cumulative histogram of user responses corresponding to various ordinal values. The system identifies user responses that are unlikely to be valid user responses to generate training data for training the machine learning mode. The system identifies invalid user responses based on response time of the user responses.

A system for controlling platooning by a following vehicle includes a sensor located in or on the following vehicle configured to detect data corresponding to a shape of a leading vehicle. The system further includes an electronic control unit (ECU) located in or on the following vehicle, coupled to the sensor, and configured to determine an optimal distance from the following vehicle to the leading vehicle based on the shape of the leading vehicle, the optimal distance corresponding to a distance at which drag applied to the following vehicle is reduced based on a pressure wake from the leading vehicle.

In various examples, a set of object trajectories may be determined based at least in part on sensor data representative of a field of view of a sensor. The set of object trajectories may be applied to a long short-term memory (LSTM) network to train the LSTM network. An expected object trajectory for an object in the field of view of the sensor may be computed by the LSTM network based at least in part an observed object trajectory. By comparing the observed object trajectory to the expected object trajectory, a determination may be made that the observed object trajectory is indicative of an anomaly.

A control device included in each vehicle composing a vehicle group predicts a vehicle state of a subject vehicle before a traveling regulation point based on speed information of the subject vehicle, light color information of a traffic signal, and distance information between the subject vehicle and an entrance of the traveling regulation point. Based on the vehicle state, it is determined whether a vehicle of which the vehicle state corresponds to an unenterable state is included in the vehicle group. The unenterable state indicates a state where the subject vehicle cannot enter the traveling regulation point before the traffic signal is changed to the yellow light when its speed is maintained at a current speed. When it is determined that the vehicle in the unenterable state is included in the vehicle group, the subject vehicle is controlled such that it does not enter the traveling regulation point.

Aspects of the subject technology relate to a vehicle backup warning system. A rearview image is received from a rearview camera capturing images of an area behind an own vehicle. The rearview image is determined to include a plurality of white pixels each having a luminance value equal to or above a luminance threshold. Two or more white pixels within a first distance of one another are grouped from the plurality of white pixels. The rearview image is determined to include two groups of the two or more white pixels. A distance between centers of the two groups is determined to be equal to or less than a second distance of each other. The two groups are identified as a pair of illuminated backup lights of a vehicle in the area behind the own vehicle. A warning is provided to alert that the vehicle's intention to backup.

Disclosed is an apparatus for assisting driving of a vehicle including a camera provided in the vehicle, a lidar sensor provided in the vehicle, and a controller configured to control the vehicle to prevent to deviate from a lane based on lane information obtained by the camera and the lidar sensor, wherein the controller excludes the lane information obtained by the camera and controls the vehicle to prevent to deviate from the lane based on the lane information obtained by the lidar sensor, when the vehicle enters a tunnel.

An information processing apparatus according to an embodiment of the present technology includes a first acquisition unit, a second acquisition unit, and a generation unit. The first acquisition unit acquires peripheral information regarding a periphery of a first mobile object. The second acquisition unit acquires, from an apparatus different from the first mobile object, attribute information regarding an attribute of a second mobile object present in the periphery of the first mobile object. The generation unit generates learning data for extracting an attribute of a target mobile object, on the basis of the acquired peripheral information and the acquired attribute information.