Lane changing decision model generation method and unmanned vehicle lane changing decision method and device

Abstract

The invention discloses a lane changing decision model generation method and an unmanned vehicle lane changing decision method and device.The lane changing decision model generation method comprises the steps that a training sample set of vehicle lane changing is obtained, the training sample set comprises a plurality of training sample groups, and the training sample groups are used for training the vehicle lane changing; each training sample group comprises training samples under each time step length in the process that a vehicle completes lane changing according to a planned lane changing track, each training sample comprises a group of state quantities and corresponding control quantities, and the state quantities comprise the pose, the speed and the acceleration of a target vehicle; the pose, the speed and the acceleration of a vehicle in front of the lane of the target vehicle and the pose, the speed and the acceleration of a following vehicle on the target lane; the control quantity comprises the speed and the angular speed of the target vehicle; and training a decision-making model based on a deep reinforcement learning network through the training sample set to obtain a lane changing decision-making model which enables the state quantity of the target vehicle to be associated with the corresponding control quantity.

Description

technical field [0001] The present invention relates to the technical field of unmanned driving, in particular to a method for generating a lane-changing decision model and a method and device for unmanned vehicle lane-changing decision-making. Background technique [0002] In the field of unmanned driving, the architecture of the autonomous system of unmanned vehicles can usually be divided into a perception system and a decision-making control system. Traditional decision-making control systems use optimization-based algorithms. However, most classic optimization-based methods are computationally complex , resulting in the inability to solve complex decision-making tasks. In reality, the driving situation of vehicles is complex, and unmanned vehicles in unstructured environments use complex sensors, such as cameras and laser rangefinders. Since the sensing data obtained by the above sensors usually depends on the complex and unknown environment, the above After the sensin...

AI Technical Summary

Problems solved by technology

[0002] In the field of unmanned driving, the architecture of the autonomous system of unmanned vehicles can usually be divided into a perception system and a decision-making control system. Traditional decision-making control systems use optimization-based algorithms. However, most classic optimization-based methods are computationally complex , resulting in the inability to solve the problem of complex decision-making tasks

In reality, the driving situation of vehicles is complex, and unmanned vehicles in unstructured environments use complex sensors, such as cameras and laser rangefinders. Since the sensing data obtained by the above sensors usually depends on the complex and unknown environment, the above After the sensing data obtained by the sensor is directly input into the algorithm framework, it is difficult for the algorithm to output the optimal control amount

In the traditional method, the slam algorithm is usually used to draw the environment, and then obtain the trajectory in the result map, but this model-based algorithm increases the instability due to the high degree of uncertainty (such as road bumps) when the vehicle is driving. factor

Images