Shared control for lane keeping assistance system based on multiple-phase handling inverse dynamics

Abstract

To achieve user-friendly design, a novel shared control scheme for Lane Keeping Assistance (LKA) system is proposed in this paper. Instead of adjusting the control authority allocation, the proposed scheme achieves shared control by adjusting the displacement interval between each phase. Through this scheme, we aim to design a LKA system which can realize continuously shared control between assistant controller and human driver with a fixed driving authority allocation strategy. Firstly, the shared control problem is formulated as a multiple-phase inverse dynamics problem; then we covert the inverse dynamics problem into an Optimal Control Problem (OCP). Secondly, by applying Radau Pseudospectral Method (RPM), the OCP is converted into a Nonlinear Programming Problem (NLP), which is then solved by Sequential Quadratic Programming (SQP). Finally, simulations under two typical conditions are implemented to validate the effectiveness of the proposed shared control approach.