Heavy-duty vehicle longitudinal automation with hydraulic retarder via H infinity control and off-policy reinforcement learning
by Chaoxian Wu; Xuexun Guo; Bo Yang; Xiaofei Pei; Zhenfu Chen
International Journal of Vehicle Design (IJVD), Vol. 82, No. 1/2/3/4, 2020

Abstract: A novel hierarchical heavy-duty vehicle (HDV) longitudinal control strategy with a hydraulic retarder is proposed in this paper to achieve the HDV down-hill longitudinal automation in the deceleration process. The upper level controller generates the optimal desired retarder torque through the H infinity control and the off-policy reinforcement learning (RL), in which the H infinity control is able to attenuate those disturbances and the off-policy RL can solve the H infinity control with completely unknown system dynamics. Then, according to the optimal desired retarder torque, the lower level controller can calculate the desired control pressure for the retarder to control the HDV. The effectiveness of this HDV longitudinal control strategy is verified by simulations based on an experimentally verified retarder model. Compared to the sliding-mode-control (SMC) based controller, the simulation result shows the proposed control strategy has better capability to attenuate the disturbances and guarantee the longitudinal speed tracking performance.

Online publication date: Thu, 01-Apr-2021

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Vehicle Design (IJVD):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com