Lane-change control of all-axle-steering virtual track train based on a novel predefined-time sliding mode method

It is necessary to share the right of way with other means of transportation during the operation of All-Axle-Steering Virtual Track Train (AASVTT), so ensuring the operation security of it in cluttered environments is significant. And lane-change maneuver is one of the major research contents. In this study, a novel lane-change control framework is proposed, which consists of path generation and path tracking for the midpoint of each axle. In order to generate the reference lane-change trajectories, the AASVTT kinematics model based trajectories are established to avoid planned-actual mismatching, and the Positive and Negative Trapezoidal Lateral Acceleration Method is used to assure the rapidly of maneuver. Additionally, a novel predefined-time sliding mode controller is proposed to achieve independent path tracking of each axle based on scalar Lyapunov function and the nonlinear dynamic model with consideration of internal uncertainties and external disturbances. The effectiveness and adaptability of the proposed controller is verified by simulation. The simulated results show that this strategy, comparing with the traditional sliding mode controller, has the advantages of fast convergence and good path tracking performance. In the end, the lane-change control algorithm is realized with expectable performance at both low-speed and high-speed.