Nonlinear Model Predictive Control Using State Estimation for Stabilization of Vehicle Dynamics to Avoid the Secondary Accident

In recent years, various control problems of vehicle dynamics such as collision avoidance, rollover prevention, wheel slip control, driver assistance control have been studied. This study focuses on the stabilization problem of unstable vehicle dynamics caused by a collision accident. In this paper, we consider the model predictive control (MPC) problem for stabilization of vehicle nonlinear dynamics to avoid the second accident after the first collision accident. MPC is a well-established control method in which the current control input is obtained by solving a finite horizon open-loop optimal control problem using the current state of the system as the initial state. However, MPC method is inapplicable to systems whose all state variables are not exactly known. In general, it is usual that the state variables of systems are measured through output sensors. Thus, only limited parts of them can be used for designing control inputs. Therefore, automatic control systems must incorporate some type of state estimation. In order to apply the MPC method to the automatic control systems for nonlinear vehicle dynamics, we need to establish a state estimation method for nonlinear vehicle systems with limited measurable state variables. The objective of this study is to establish a control method for unstable vehicle dynamics caused by a collision accident by means of incorporating state estimation into model predictive control.