Model-based predictive control of a motor vehicle
DRIVE
June 9, 2026
A method for model-based predictive control (MPC) of a motor vehicle includes executing an MPC algorithm having a high level solver module, a longitudinal dynamics model, and a cost function dedicated to the high level solver module. By executing the high level solver module for an upcoming route segment while taking the longitudinal dynamics model into account, a speed trajectory is calculated that minimizes the cost function, according to which the motor vehicle is to travel within a prediction horizon. The method includes sending the speed trajectory to a human-machine interface as an input value, processing the speed trajectory to obtain a control signal in the human-machine interface, and outputting the control signal to a driver of the motor vehicle with the human-machine interface, such that the driver can control the motor vehicle in accordance with the control signal.
Discussion in the ATmosphere