Kalman Filter-Based Model Predictive Control for an Adaptive Cruise Control System Considering Measurement Noise

Sensor measurement noise has a great influence on adaptive cruise control (ACC) systems. To improve the robust performance of ACC application in the real world, a Kalman filter-based model predictive control (MPC) method was proposed in this study, in which the Kalman filter was adopted to deal with the measurement noise of state variables, and the MPC controller was developed to improve the ACC system performance of the longitudinal car-following accuracy, safety, and riding comfort. In the proposed MPC controller, a state feedback correction algorithm was applied to improve the accuracy of the predictive model in cases of parameter uncertainty and external disturbances. Then, relaxation factors were introduced to soften and extend the scope of the constraints to obtain a feasible solution. Further, the controlled variable of the MPC controller, which was defined as the desired acceleration of the host vehicle, was solved and rolling was optimized by using an optimization method based on a quadratic programming algorithm. Software simulations and real vehicle tests were conducted to verify the effectiveness of the proposed MPC controller with the Kalman filter, demonstrating that the proposed method significantly improves the safety, comfort, and performance of an ACC system.