基于人机共驾的车道保持辅助控制系统研究（双语出版）

车道保持控制系统是汽车安全辅助驾驶的重要组成部分，可有效提高汽车主动安全性、避免车辆无意识地偏离本车道。目前，大部分车道保持控制系统在工作时将驾驶人的操作视为外界干扰，没有考虑人机共驾阶段下驾驶人与控制系统的控制权分配问题，易造成人机冲突、影响驾驶人的驾驶感受。论文兼顾驾驶人与辅助控制系统各自优势，基于人机共驾技术对车道保持控制系统进行研究。构建基于安全行驶区域与最晚预警边界相结合的车道偏离决策模型，在保证其预警精度的同时降低计算复杂性，根据车辆行驶状态和路面附着系数动态调整预警阈值；研究串级MPC-PID控制策略实现对车辆横向位置的控制，将最优问题转化为二次规划求得目标前轮转角，利用PID算法完成对目标前轮转角的跟踪；引入共驾系数对车辆的控制权进行分配，研究共驾系数分配模型，以车辆状态误差和驾驶人转向力矩作为模糊控制的输入变量、共驾系数作为输出变量，降低辅助控制系统与驾驶人之间的冲突；最后，利用CarSim与Simulink联合仿真对所研究的控制策略进行仿真验证，结果表明共驾系数能够根据驾驶人的操作和车辆运行状态的变化实现动态调整，辅助控制力矩与驾驶人输入力矩变化趋势相同，在保留驾驶人一定操作的基础下可避免车辆偏离车道、降低人机冲突。

Lane-keeping Control Systems Based on Human-machine Cooperative Driving（in English）

Lane-keeping control systems can effectively improve the active safety of a vehicle and avoid unconscious deviation from the lane. At present, most of the lane-keeping control systems regard operation by the driver as an external disturbance when the system is activated. Such systems fail to take into account the allocation of control rights between the system and the driver, which results in human-machine conflict and affects the driving experience. To consider fully the advantages of both the driver and the control assist system, this paper presents a lane-keeping control system based on cooperative human-machine driving. To ensure accuracy and reduce computation complexity, a lane departure decision model was adopted by combining the safe driving area and the latest warning boundary. The decision threshold was adjusted dynamically according to the running state of the vehicle and road adhesion coefficient. The cascaded MPC (model predictive control)-PID (proportional-integral-derivative) control strategy was proposed to control the lateral position of the vehicle. The optimal problem was transformed into quadratic programming to obtain the target front-wheel steering angle, which was tracked by a PID control algorithm. A cooperative driving coefficient was put forward to distribute the control rights of the vehicle. The lateral position and the direction errors were used to define the error state of the vehicle. The error state of the vehicle and the steering torque of the driver were used as the input variables for the fuzzy controller. The cooperative driving coefficient was the output variable of the controller. CarSim and Simulink were used to simulate and verify the proposed lane-keeping control strategy. The results show that the cooperative driving coefficient can be adjusted dynamically according to the operation of the driver and a change in the running state of the vehicle. The assisted torque changes with the same tendency as that of the input torque from the driver. The vehicle can avoid deviating from the lane while preserving the rights of the driver to operate the vehicle and reduce human-machine conflict.