基于强化学习算法的地下铲运机车速控制

王伯健; 战凯; 郭鑫; 石峰; 高泽宇

Speed control of LHD based on reinforcement learning Algorithm

Received:November 12, 2021 Revised:December 08, 2021 Accepted:December 30, 2021 Published Online:April 27, 2022

View Full Text View/Add Comment Download reader

DOI：

KeyWord:LHD; reinforcement learning; unmanned driving; speed control.

Author	Institution
wangbojian	Beijing General Research Institute of Mining and Metallurgy
zhankai	Beijing General Research Institute of Mining and Metallurgy
guoxin	Beijing General Research Institute of Mining and Metallurgy
shifeng	Beijing General Research Institute of Mining and Metallurgy
gaozeyu	Beijing General Research Institute of Mining and Metallurgy

Hits: 2272

Download times: 967

Abstract:

To solve the problem that it is difficult to control the change of the speed of LHD unmanned driving, the reinforcement learning algorithm is applied to control the speed of LHD unmanned driving, so that the vehicle speed can keep smooth and stable in various states. This article compares the reinforcement learning algorithm and the experience method, fuzzy control and traditional PID control, synovial control, inverse control, intelligent optimization algorithm and other algorithms, the analysis and design of the reinforcement learning strategy, deduced the reinforcement learning model which control the speed and the speed of the car on a moment, a moment on the course Angle deviation, a moment position deviation, the relationship between related parameters calculation, Simulation experiments are carried out to verify the correctness of the model. Experimental results show that compared with traditional fuzzy hierarchical control speed and experience method to set the speed of reinforcement learning algorithm to control well improved the stability of the speed change, speed changes according to the environment and their own state flexibly and properly adjusting the speed of change, to improve the dynamic performance of the vehicle and reduce the error.