Investigation of steer preview methods to improve predictive control methods on off-road vehicles with realistic actuator delays

被引:0
|
作者
Peenze, Andries J. [1 ]
Els, P. Schalk [1 ]
机构
[1] Univ Pretoria, Dept Mech & Aeronaut Engn, Vehicle Dynam Grp, Pretoria, South Africa
关键词
Steering preview; Predictive control; Off-road vehicle dynamics; Actuator delays; PATH TRACKING;
D O I
10.1016/j.jterra.2024.101027
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
This paper investigates improvements that can be made to predictive control methods for off-road vehicles by adding of realistic steering preview. The objective of this study is to improve the performance and efficacy of predictive controllers by accounting for significant time delays in active and semi-active systems on vehicles. Traditional zero-order and first-order hold methods for steer preview are compared to a more realistic steer preview method. Semi-active suspension, rear wheel steering, and individual brake actuation are used as the actuators on this off-road vehicle. The results show that the addition of a realistic steering preview improves the handling performance of the vehicle in a severe double lane change manoeuvre on rough roads. Up to 10% reduction in roll angle can be achieved with semi-active suspension control. A 34% reduction in side-slip angle is possible with rear wheel steering control and a 15% reduction in side-slip angle is achieved with differential braking control. The controllers can pre-empt and consider the effect of the actuator time delays, and the preview states from the predictive controller are more representative over the prediction horizon. The findings suggest that the addition of a realistic steering preview can improve the performance of predictive controllers on vehicles. Further investigation of other disturbances and their preview effects on the system should be conducted to find further improvements for predictive control strategies on vehicles. (c) 2024 ISTVS. Published by Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
引用
收藏
页数:18
相关论文
empty
未找到相关数据