Described is a system for learning object labels for control of an autonomous platform. Pseudo-task optimization is performed to identify an optimal pseudo-task for each source model of one or more source models. An initial target network is trained using the optimal pseudo-task. Source image components are extracted from source models, and an attribute dictionary of attributes is generated from the source image components. Using zero-shot attribution distillation, the unlabeled target data is aligned with the source models similar to the unlabeled target data. The unlabeled target data are mapped onto attributes in the attribute dictionary. A new target network is generated from the mapping, and the new target network is used to assign an object label to an object in the unlabeled target data. The autonomous platform is controlled based on the object label.

A vehicle includes a lane generator that is configured to receive information that describes a first lane portion and a second lane portion and determine that a discontinuity is present. The lane generator obtains a sensor output and detects presence of a nearby vehicle in the first lane portion. A classification for the nearby vehicle is identified by analyzing the sensor output, and is used to select a vehicle kinematics model. The lane generator determines paths for a simulated vehicle from the first lane portion to the second lane portion using the vehicle kinematics model. A third lane portion is determined based on the paths such that the third lane portion defines a traversable route from the first lane portion to the second lane portion in accordance with the vehicle kinematics model. An automated vehicle control system generates control outputs based in part on the third lane portion.

A method and apparatus for controlling a vehicle is disclosed. The method may include: determining center points of at least two frames of point clouds collected for an identified obstacle during travelling of the vehicle; performing curve fitting based on the determined center points to obtain a fitted curve; determining a moving velocity of the obstacle based on the fitted curve; predicting whether the vehicle is to be collided with the obstacle when the vehicle continues travelling at a current velocity, based on the moving velocity of the obstacle, the traveling velocity of the vehicle, and a distance between the obstacle and the vehicle; and sending control information to the vehicle, in response to predicting that the vehicle is to be collided with the obstacle when the vehicle continues travelling at the current velocity, the control information being used to control the vehicle to avoid collision with the obstacle.

Techniques and methods for performing collision monitoring using system data. For instance, a vehicle may generate sensor data using one or more sensors. The vehicle may then analyze the sensor data using systems in order to determine parameters associated with the vehicle and parameters associated with another object. Additionally, the vehicle may determine uncertainties associated with the parameters and then process the parameters using the uncertainties. Based at least in part on the processing, the vehicle may determine a distribution of estimated locations associated with the vehicle and a distribution of estimated locations associated with the object. Using the distributions of estimated locations, the vehicle may determine the probability of collision between the vehicle and the object.

Techniques for determining a response of a simulated vehicle to a simulated object in a simulation are discussed herein. Log data captured by a physical vehicle in an environment can be received. Object data representing an object in the log data can be used to instantiate a simulated object in a simulation to determine a response of a simulated vehicle to the simulated object. Additionally, one or more trajectory segments in a trajectory library representing the log data can be determined and instantiated as a trajectory of the simulated object in order to increase the accuracy and realism of the simulation.

Determining classification(s) for object(s) in an environment of autonomous vehicle, and controlling the vehicle based on the determined classification(s). For example, autonomous steering, acceleration, and/or deceleration of the vehicle can be controlled based on determined pose(s) and/or classification(s) for objects in the environment. The control can be based on the pose(s) and/or classification(s) directly, and/or based on movement parameter(s), for the object(s), determined based on the pose(s) and/or classification(s). In many implementations, pose(s) and/or classification(s) of environmental object(s) are determined based on data from a phase coherent Light Detection and Ranging (LIDAR) component of the vehicle, such as a phase coherent LIDAR monopulse component and/or a frequency-modulated continuous wave (FMCW) LIDAR component.

A moving object escape prevention method includes: controlling, by a processor of a moving object, to drive the moving object based on autonomous driving; detecting, by the processor, whether a collision occurred by the moving object; in response to detecting the collision, transmitting, by the processor, a collision occurrence notification signal and position information of the moving object to an Intelligent Transportation System Infrastructure (ITSI); receiving, by the processor, escape-related information from the ITSI. The receiving escape-related information includes: determining, by the ITSI, whether or not the moving object escapes based on position information of the moving object; receiving, by the processor, accident handling information from the ITSI upon determining that the moving object does not escape, and receiving, by the processor, an escape warning message from the ITSI when the position information of the moving object changes.

A method and an apparatus for controlling lane changing, and a storage medium includes: predicting a target pose of a vehicle changed to a second lane based on a current pose of the vehicle on a first lane in response to a trigger of changing the vehicle from the first lane to the second lane; and determining a lane changing preparation pose of the vehicle on the first lane based on the target pose and at least one parameter of the vehicle.

Systems and methods for operating a robot. The methods comprise: performing, by a processor, operations to detect an object that is moving; identifying, by the processor, detected behavior of the object that constitutes an unrecognized behavior; predicting, by the processor, future movement of the object based on a circle having a radius that is function of a velocity of the object; and controlling operations of the robot based on the predicting.

An autonomous driving control method carried out by an autonomous driving control system having an autonomous driving control unit that executes an autonomous driving control for causing a host vehicle to travel along a target travel route generated on a map, comprising setting one or a plurality of target passage gates through which the host vehicle is scheduled to pass during passage through a toll plaza, determining the presence or absence of a preceding vehicle that has the predicted passage gate that matches the target passage gate of the host vehicle from among a plurality of preceding vehicles, and carrying out following travel using the preceding vehicle that has the predicted passage gate that matches the target passage gate as a follow target.