Controlling nonlinear vehicular motions by exploiting linearized feedback law under delay-tolerance: stability, gain-scheduling, and validation
摘要
The automation of transportation systems inevitably faces the challenge of enhancing both the safety and intelligence of passenger vehicles. In this transitional stage toward full automation, advanced driver assistance systems (ADAS) play a critical role in bridging the gap. A key component of ADAS is vehicle stability control (VSC), which ensures motion stability during highly nonlinear handling maneuvers. This paper addresses the system nonlinearity under critical driving conditions and the loop delays within feedback processes by proposing a delay-tolerant feedback structure for VSC. The approach utilizes only the linearized dynamics along the trajectory of the maneuver, where the target-tracking performance is optimized. A nonlinear vehicle model is first constructed, followed by an investigation of its open-loop characteristics through equilibrium analysis and local linearization. Time delays arising from control sampling and actuation are incorporated into the feedback torque, yielding a delayed nonlinear system. A semi-discretized method is developed to construct stability charts of the tunable control gains, whose aggregation yields a conservative delay-tolerant domain. Two gain scheduling strategies are proposed to achieve maximum target-tracking performance, tailored for either real-time (RT) or offline implementation. The proposed method is designed for stable tracking of dynamic motion references under nonlinear conditions and is validated using experimental data-based simulations. The results demonstrate that a linearized control law, when properly designed, can deliver high-performance VSC with strong adaptability across different control loops subject to varying delays.