关键词: 半挂汽车列车, 主动转向, 粒子群优化算法, 线性二次型调节器, 最优控制

Abstract: To address the poor tracking performance of semitrailers when tractors steering at low speeds, an active steering Linear Quadratic Regulator (LQR) based on Particle Swarm Optimization (PSO) is designed, and the impact of different weight matrix acquisition methods on the steering control effect of trailers is explored. First, the reliability of the kinematics model of the tractor-semitrailer with the trailer steering system is verified. Second, an LQR for low-speed trajectory tracking of the trailer is designed, and the weight matrix in the LQR is optimized by employing PSO algorithm. Finally, the controller performance with different weight matrix acquired by different methods is studied. Research results show the LQR controller optimized by PSO algorithm allows the trailer to enter a stable tracking state faster; when the weight matrix R is set to 0.1 and 1, compared to the trailer tracking error corresponding to the manually adjusted weight matrix Q, the trailer tracking error corresponding to the global optimal weight matrix is down by 26.1% and 19.4% respectively in a single U-shaped path, and down by 40.9% and 43.4% respectively in a spiral path on the ramp.