Described herein are a method of training a learning-based critic for tuning a rule-based motion planner of an autonomous driving vehicle, a method of tuning a motion planner using an automatic tuning framework that with the learning-based critic. The method includes receiving training data that incudes human driving trajectories and random trajectories derived from the human driving trajectories; training a learning-based critic using the training data; identifying a set of discrepant trajectories by comparing a first set of trajectories, and a second set of trajectories; and refining, at the neural network training platform, the learning-based critic based on the set of discrepant trajectories. The automatic tuning framework can remove human efforts in tedious parameter tuning, reduce tuning time, while retaining the physical and safety constraints of the ruled-based motion planner. Further, the automatic tuning framework can create personalized motion planners when the learning-based critic is trained using different human driving datasets.

A vehicle control device includes a processor. The processor is configured to: output a torque command value related to a rotation speed of a wheel of a vehicle; specify an estimated value which is a value obtained by estimating the rotation speed of the wheel based on the torque command value; and determine a parameter based on an error between the estimated value and a measured value which is a value obtained by measuring the rotation speed of the wheel. The torque command value is determined by a feedforward control using a target value which is a value as a target of the rotation speed of the wheel and the parameter.

Systems and methods are provided herein for controlling the speed on each wheel of a vehicle, possibly operating a vehicle in a speed control mode. In response to receiving input to engage speed control mode and receiving an accelerator pedal input, the system determines a target wheel speed based on the accelerator pedal input, monitors wheel speed of each of a plurality of wheels and determines, for each monitored wheel, a difference based on the monitored wheel speed and the target wheel speed. A torque is provided to each of the plurality of wheels based on the respective difference to achieve the target wheel speed.

Embodiments of this application disclose a vehicle control method and device, where the method includes: calculating a longitudinal force interference compensation torque and a lateral force interference compensation torque of a vehicle when a flat tire occurs in the vehicle; calculating a feedback control torque of the vehicle; determining an additional yaw moment based on the longitudinal force interference compensation torque, the feedback control torque, and the lateral force interference compensation torque; and controlling, based on the additional yaw moment, a wheel in which the flat tire occurs.

To control a hybrid dynamical system, a predictive feedback controller formulates a mixed-integer nonlinear programming (MINLP) problem including nonlinear functions of continuous optimization variables representing the continuous elements of the operation of the hybrid dynamical system and/or one or multiple linear functions of integer optimization variables representing the discrete elements of the operation of the hybrid dynamical system. The MINLP problem is formulated into a separable format ensuring that the discrete elements of the operation are present only in the linear functions of the MINLP problem. The MINLP problem is solved over multiple iterations using a partial convexification of a portion of a space of the solution including a current solution guess. The partial convexification produces a convex approximation of the nonlinear functions of the MINLP without approximating the linear functions of the MINLP to produce a partially convexified MINLP.

A control device for electric motor vehicle configured to decelerate by a regenerative braking force of the motor detects an accelerator operation amount, calculates a motor torque command value and controls the motor on the basis of the calculated motor torque command value. Further, a speed parameter proportional to a traveling speed is detected, and a feedback torque for stopping the electric motor vehicle is calculated on the basis of the detected speed parameter. Furthermore, the speed parameter is estimated in accordance with a state of the electric motor vehicle, and a feedforward torque is calculated on the basis of the estimated speed parameter. When accelerator operation amount is not larger than a predetermined value and the electric motor vehicle stops shortly, the motor torque command value is converged to zero on the basis of the feedback torque and the feedforward torque with a reduction in the traveling speed.

A method for the closed-loop and/or open-loop control of a lateral guidance of a vehicle with the aid of a lane-keeping assist. In the process, a detection signal is read in which represents hands-off and/or hands-on driving of the vehicle. If the detection signal represents the hands-off driving, then a closed-loop control signal is provided for controlling the lateral guidance in closed loop. On the other hand, if the detection signal represents the hands-on driving, then an open-loop control signal is provided for controlling the lateral guidance in open loop.

A system for a vehicle can include a computer having a processor and a memory, the memory storing instructions executable by the processor, including instructions to determine a curvature command for a vehicle based on a feedforward control action based on a road curvature, a feedback control action based on at least one of a path offset or a heading offset of a vehicle with respect to a line defining a target lane; and to operate the vehicle with respect to the line defining the target lane based on the curvature command.

A vehicle travel control apparatus includes: a sensor that obtains preceding vehicle information representing a status of a preceding vehicle; a communication apparatus that obtains preceding vehicle acceleration/deceleration information, which is generated in the preceding vehicle, via communication with the preceding vehicle; and a controller that generates a first target value related to a target acceleration/deceleration value of a host vehicle based on the preceding vehicle information and a second target value related to the target acceleration/deceleration value of the host vehicle based on the preceding vehicle acceleration/deceleration information, and controls acceleration/deceleration of the host vehicle based on the generated first and second target values, wherein the controller corrects the preceding vehicle acceleration/deceleration information according to a travel scene to generate the second target value.

Provided are an autonomous driving system and a correction learning method for autonomous driving. The autonomous driving system includes a sensor configured to collect and output data required for autonomous driving, a first processor configured to output autonomous driving data on the basis of data input from the sensor, a second processor configured to output a driving data adjustment value on the basis of differences between the data input from the sensor, the autonomous driving data input from the first processor, and driving data input from driving by a human driver, and a driving part configured to perform driving on the basis of the autonomous driving data output from the first processor and the driving data adjustment value output from the second processor.