Automatic driving method and device, electronic equipment and storage medium

Abstract

The invention discloses an automatic driving method and device, electronic equipment and a computer readable storage medium. The method comprises the steps that multi-mode sensing information and driving behavior data of a driving environment are acquired; extracting multi-scale features of the multi-modal perception information by using a convolutional neural network, and fusing the multi-scale features by using Transform to obtain fused feature data; the fusion feature data and the driving behavior data are combined into expert demonstration data, and an automatic driving process is modeled into a Markov decision process; obtaining a reward function of an automatic driving process by using expert demonstration data and maximum entropy inverse reinforcement learning, and optimizing a driving strategy model by using deep reinforcement learning; and outputting the optimized driving strategy model to the client, so that the client realizes automatic driving by using the optimized driving strategy model according to the environmental perception information. The reliability of automatic driving perception data is ensured, and the reasonability of decision planning in the automatic driving process is improved.

Description

technical field [0001] The present application relates to the field of computer technology, and more specifically, to an automatic driving method and device, an electronic device, and a computer-readable storage medium. Background technique [0002] Autonomous driving is a complex system integrating environmental perception, decision-making planning, control execution and other functions. With the development of artificial intelligence and information technology, autonomous driving has attracted great attention from academia, industry and national defense and military fields. [0003] The end-to-end driving method takes the perceptual information measured by the sensor (such as lidar point cloud, RGB image, etc.) as the input of the neural network, and the neural network directly outputs control signals, such as steering commands and acceleration. The main advantage of this framework is that it is easy to implement and can obtain labeled training data by recording the human ...

AI Technical Summary

Problems solved by technology

However, the existing DRL-based end-to-end driving methods mainly focus on driving scenarios with limited dynamic participants, and assume that the behavior of other participants in the scenario is close to ideal, and it is difficult to take into account various complex problems in the real world.

In addition, this type of method mostly focuses on a single input mode, that is, the acquired perceptual information is only based on images or radar. When the confrontation of participants in the scene increases, the system will fail due to the lack of key perceptual information.

[0005] Therefore, in the process of automatic driving, in order to prevent the loss of key perception information caused by the complexity of the driving scene and the increase of confrontation, it is a technical problem that those skilled in the art need to solve.

Images