Ramp cooperative control system and method based on driving style

Abstract

The invention discloses a ramp cooperative control system and method based on a driving style. The system comprises a vehicle-mounted control subsystem, an inter-vehicle communication subsystem and a cooperative decision-making subsystem based on the driving style. According to the method, states and reaction decisions of competitive vehicles in the lane merging process are fully considered, a ramp collaborative merging interaction decision model is established, and a multi-vehicle deep reinforcement learning framework is designed; in order to consider the influence of the random driving style on the decision process, an optimal merging strategy is defined based on Bayesian Nash equilibrium, an experience pool is established through interactive learning with the environment, and the optimal merging strategy is trained and learned. According to the method, driving performance weights such as safety and high efficiency can be adjusted based on the driving style, so that personalized driving is realized, and the acceptability of an automatic driving vehicle is improved; by interactively learning a decision strategy with the environment, the adaptability of a decision algorithm to a dynamic scene is improved.

Description

technical field [0001] The invention belongs to the technical field of intelligent driving, and in particular relates to a ramp cooperative control system and method based on driving style. Background technique [0002] Among the various causes of traffic congestion, merging of vehicles in an on-ramp area is a typical one, especially when an on-ramp vehicle meets a vehicle behind on the main line. In the merging area, the on-ramp vehicles try to merge with the main line, and there is a right-of-way competition between the on-ramp vehicles and the main line vehicles, resulting in the vehicles having to slow down, stop and go, or even conflict with each other. [0003] In the current research on ramp cooperative control technology, most of the research focuses on optimizing the merging sequence of connected vehicles to improve the efficiency of traffic flow; [0004] As far as cooperative driving strategies are concerned, it can be divided into three categories, including: 1)...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies of the above-mentioned prior art, the purpose of the present invention is to provide a ramp cooperative control system and method based on driving style to solve the problems of poor adaptability and difficulty in realizing safe and personalized driving in the existing ramp control technology; For ramp merging conditions, the present invention fully considers the state and response decision-making of competing vehicles during the merging process, establishes a ramp collaborative merging interactive decision-making model, and designs a framework for multi-vehicle deep reinforcement learning; in order to consider the impact of random driving styles on the decision-making process, based on Bayesian Nash Equilibrium defines the optimal merging strategy, and establishes an experience pool through interactive learning with the environment, and trains and learns the optimal merging strategy

Images