The invention provides a vehicle
adaptive control method based on PSO-BP optimization, and the method comprises the steps: building a transverse
tracking error model based on a two-degree-of-freedom
vehicle dynamics model, building a transverse controller through building a prediction model, building a constraint condition, and setting a target function, and then building a longitudinal controller based on a first-order inertial
system to solve a longitudinal acceleration, and the information is transmitted to a lower-layer controller to control the speed of the vehicle. In view of the problems of large
lateral deviation and insufficient driving stability in lateral control, in view of the important effect of
time domain parameters on
vehicle control, a PSO-BP method is provided to optimize an MPC controller,
simulation is performed by selecting different speeds, different prediction time domains and different control time domains, a
data set is obtained, and an MPC controller is optimized. And optimal
time domain parameter values corresponding to different vehicle speeds are obtained by using a PSO-BP method based on the
data set, so that the vehicle can adaptively select the optimal
time domain parameter values according to the vehicle speeds in the
overtaking and lane changing process of the expressway. The control precision and the driving stability are effectively improved.