Comprehensive Predictive Control Method for Automated Vehicles With Delays

This paper considers the control problem of automated vehicles with delays. We develop a comprehensive control method within the framework of model predictive control (MPC) with compensation for the input delay and the state delay for automated vehicles on the multi-lane road, where the road boundaries and lane markings are modeled as potential functions. The original vehicle dynamics model is modified to formulate the augmented prediction model to compensate for delays. We also present a parallel strategy for the executions of the perception system, the controller, and the actuator, which can directly cut delays in the loop. The simulations show that this method can greatly reduce the impact of delays on the performance of the controller and, therefore, enhance its capabilities.