Multi-objective and multi-stage trajectory tracking controller for high-speed train based on weight optimization

The trajectory tracking is the core function to realize automatic operation of high-speed train. This paper designed a model predictive controller under consideration of multiple optimal objectives to achieve comprehensive control performance of accurate tracking, passenger comfort and energy conservation. By linearizing the cost function and constraints, the control optimization model is transformed into a mixed-integer quadratic programming model, which is solved by the Gurobi solver. There are several stages in the whole process of train operation with different priorities of multiple objectives. To adapt the different priorities of multiple objectives, this paper proposed a multi-objective and multi-stage (MOMS) MPC controller based on multi-objective particle swarm optimization algorithm and a selection strategy of optimal weight coefficient. The simulation results show that the controller designed in this paper can optimize the weight in different scenarios to improve comprehensive control performance.