A method for combating a stop-and-go wave problem using deep reinforcement learning based autonomous vehicles includes selecting one of a plurality of deep reinforcement learning algorithms for training an autonomous vehicle and a reward function in a roundabout environment in which autonomous vehicles and non-autonomous vehicles are driving, determining a deep neural network architecture according to the selected deep reinforcement learning algorithm, learning a policy which enables the autonomous vehicle to drive at a closest velocity to a constant velocity based on state information including a velocity of the autonomous vehicle and a relative velocity and a relative position between the autonomous vehicle and an observable vehicle by the autonomous vehicle at a preset time interval and reward information, using the selected deep reinforcement learning algorithm, and driving the autonomous vehicle based on the learned policy to determine an action of the autonomous vehicle.

While operating a vehicle, a candidate marker is detected via first image data from a first image sensor. Upon failing to identify the candidate marker, vehicle exterior lighting is actuated to illuminate the candidate marker. Then the candidate marker is determined to be one of a real marker or a projected marker based on determining whether the candidate marker is detected via second image data from the first image sensor. Upon determining the candidate marker is the real marker, the vehicle is operated based on the real marker.

While operating a vehicle, a candidate marker is detected via first image data from a first image sensor. Upon failing to identify the candidate marker, vehicle exterior lighting is actuated to illuminate the candidate marker. Then the candidate marker is determined to be one of a real marker or a projected marker based on determining whether the candidate marker is detected via second image data from the first image sensor. Upon determining the candidate marker is the real marker, the vehicle is operated based on the real marker.

A device includes sensing circuitry, compression circuitry, and a memory. The sensing circuitry, in operation, generates sensor data. The compression circuitry is coupled to the sensing circuitry, and, in operation, determines environmental contexts based on variation rates of sensor data and compresses sensor data based on determined environmental contexts. The compressed data is stored in the memory.

A device includes sensing circuitry, compression circuitry, and a memory. The sensing circuitry, in operation, generates sensor data. The compression circuitry is coupled to the sensing circuitry, and, in operation, determines environmental contexts based on variation rates of sensor data and compresses sensor data based on determined environmental contexts. The compressed data is stored in the memory.

A method for selecting and executing at least one reactive action of a vehicle includes a control unit receiving sensor data from a vehicle sensor system; detecting an emergency situation based on the sensor data; performing an evaluation; and selecting and implementing a reactive action for minimizing an accident risk of the vehicle based on the evaluation, where, in the evaluation, sensors that are uninvolved in the detection of the emergency situation are not taken into account or are taken into account at a lower weighting, are operated at a reduced performance, and/or are operated with a reduced scanning range. In addition, a control unit, computer program, and machine-readable memory medium can be provided for implementing the method.

The present disclosure provides a method, system, and apparatus for processing parking and a vehicle controller, and relates to the field of intelligent transportation technology, specifically to the field of automated parking technology. The method is executed by a parking system deployed on a vehicle controller, the parking system including a perception module and other modules except the perception module; the perception module being deployed on a first operating system in the vehicle controller; and the other modules being deployed on a second operating system in the vehicle controller; the method includes: processing an image collected by an image collector through the perception module to obtain perception result data; and controlling a vehicle based on the perception result data obtained from the perception module by the other modules.

The present disclosure provides a method, system, and apparatus for processing parking and a vehicle controller, and relates to the field of intelligent transportation technology, specifically to the field of automated parking technology. The method is executed by a parking system deployed on a vehicle controller, the parking system including a perception module and other modules except the perception module; the perception module being deployed on a first operating system in the vehicle controller; and the other modules being deployed on a second operating system in the vehicle controller; the method includes: processing an image collected by an image collector through the perception module to obtain perception result data; and controlling a vehicle based on the perception result data obtained from the perception module by the other modules.

An automatic steering system for an agricultural vehicle includes an assisted steering device and a controller or estimator operatively coupled to the assisted steering device and configured to adjust a calibration range. Adjusting the calibration range includes generating an estimated curvature for the vehicle based on a pose of the agricultural vehicle; generating an estimated encoder position of the encoder based on the estimated curvature; determining a current encoder position of the encoder; and adjusting the calibrated encoder range based on the estimated encoder position and the current encoder position.

An automatic steering system for an agricultural vehicle includes an assisted steering device and a controller or estimator operatively coupled to the assisted steering device and configured to adjust a calibration range. Adjusting the calibration range includes generating an estimated curvature for the vehicle based on a pose of the agricultural vehicle; generating an estimated encoder position of the encoder based on the estimated curvature; determining a current encoder position of the encoder; and adjusting the calibrated encoder range based on the estimated encoder position and the current encoder position.