A formation method for unmanned vehicles on highways based on multi-agent reinforcement learning

Abstract

The invention provides a formation method for unmanned vehicles on high-speed roads based on multi-agent reinforcement learning. The vehicle formation problem is regarded as a multi-agent cooperation problem, and each vehicle has the ability to make independent decisions, which can realize safe and fast driving Under the premise of flexible formation, that is, it is safe to avoid obstacles when the traffic flow is large, and it is not necessary to maintain the formation, and the formation is restored when the traffic flow is small; the end-to-end method that directly maps from the image input to the vehicle control quantity is trained due to the large action search space. It is very difficult, so the present invention only uses the multi-agent reinforcement learning method to learn the lane-changing strategy, and then combines the S-T graph trajectory optimization method to calculate the precise control amount, increase the control constraints, respect the principle of kinematics, and ensure safety. Comply with human driving habits.

Description

Formation method of unmanned vehicles on expressway based on multi-agent reinforcement learning technical field The invention belongs to the technical field of intelligent vehicles, in particular to a kind of expressway based on multi-agent reinforcement learning Unmanned vehicle formation method. Background technique [0002] Autonomous vehicles have been researched for decades and can replace humans. Cumbersome operations in complex scenarios such as high-density, long-cycle, and large traffic have high social and economic value. Freeway has Features such as clear topology, known traffic rules, clear constraints, and relative closure make it a typical field for autonomous driving to land. scene. Among them, the formation of intelligent logistics vehicles is a key problem worthy of research, which is important for reducing fuel consumption and improving fleet operation. Efficiency, reducing traffic congestion and other aspects play an important role. However, for...

AI Technical Summary

Problems solved by technology

First of all, the dynamic vehicle movement state on the expressway is complex, and it is difficult for the vehicle formation to coordinate; secondly, the vehicle perception is limited, and the stability of the formation system is poor; third, the fixed formation mode makes the system less flexible and has a greater impact on surrounding vehicles

[0003] The formation method based on traditional control requires complex controller design, and the system-level control method has high requirements for the stability of a single vehicle. If a vehicle fails during formation driving, the control program needs to be changed manually. In high-speed road scenarios, fixed control modes will also lose system flexibility and adaptability to environmental changes

Reinforcement learning is machine learning. With the development of artificial intelligence and machine learning, reinforcement learning has gradually been applied to autonomous driving tasks, but it is usually aimed at single-vehicle intelligence, and the advantages of reinforcement learning in the field of multi-agents have not been fully exploited.

Images