一类基于轨迹预测的驾驶员方向控制模型

发布时间：2018-08-27 15:43

【摘要】：驾驶员方向控制模型在人-车-路闭环系统仿真、驾驶员辅助系统开发和智能汽车控制中具有重要作用。在假设驾驶员具有汽车轨迹预测能力的基础上,提出一类基于轨迹预测的驾驶员方向控制模型。分别假定汽车在将来一段时间内保持恒定的横摆角速度或横摆角加速度,并结合汽车状态参数预测汽车的行驶轨迹,采用期望式、增量式以及期望式与增量式集成的转角决策方法建立5种不同的驾驶员模型。在ve DYNA/Simulink联合仿真平台上对各驾驶员模型进行仿真试验,结果表明,增量式驾驶员模型表现出良好的路径跟踪精度和很强的鲁棒性,期望式模型的转向操纵更加平滑,而集成式模型则具备综合优势。在ve DYNA/Labview硬件在环实时试验台架上对所提出的驾驶员模型进行模拟试验,所得结论与仿真基本一致。
[Abstract]:Driver direction control model plays an important role in the simulation of man-vehicle-road closed-loop system, the development of driver-assisted system and the intelligent vehicle control. Based on the assumption that the driver has the ability to predict the vehicle trajectory, a kind of driver direction control model based on the trajectory prediction is proposed. It is assumed that the vehicle will maintain a constant yaw angular velocity or yaw acceleration for a period of time in the future, and the vehicle trajectory can be predicted by combining the vehicle state parameters, and the desired formula is adopted. Five different driver models are established by incremental and expected and incremental angle decision methods. The simulation results of each driver model on the ve DYNA/Simulink platform show that the incremental driver model has good path tracking accuracy and strong robustness, and the steering control of the expected model is smoother. The integrated model has comprehensive advantages. The proposed driver model is simulated on the ring real time test rig with ve DYNA/Labview hardware, and the results are in good agreement with the simulation results.
【作者单位】：合肥工业大学机械与汽车工程学院;
【基金】：国家自然科学基金资助项目(U1564201,51375131,51305118)
【分类号】：U463.6

【相似文献】