基于模糊双曲正切模型的车辆路径跟踪算法

发布时间：2018-08-20 10:18

【摘要】：为了实现铰接式车辆无人驾驶技术。针对路径跟踪问题,该文提出了基于模糊双曲正切模型的铰接式车辆路径跟踪控制算法。首先根据实地试验测得铰接式车辆的横向偏差、横向偏差变化率、航向角偏差、航向角偏差变化率和转向角的样本数据,建立其模糊双曲正切模型。在此基础上,采用改进的自适应反向传播(back propagation,BP)神经网络对模型进行参数辨识,并推导了基于Cauchy鲁棒误差估计器的权系数调解率公式。然后设计基于极点配置方法的控制器,得到转角的反馈控制率。从试验数据可以看出:车辆横向位置偏差、航向角偏差、转角控制量分别控制在0.008 m、0.07 rad(0.5°)、0.21 rad(12°)附近,各向偏差均稳定,误差控制在1%以内。该种路径跟踪控制算法的研究可为铰接式车辆无人驾驶提供参考。
[Abstract]:In order to achieve articulated vehicle driverless technology. To solve the problem of path tracking, this paper presents a control algorithm of articulated vehicle path tracking based on fuzzy hyperbolic tangent model. Firstly, the fuzzy hyperbolic tangent model is established based on the sample data of lateral deviation, lateral deviation change rate, heading angle deviation rate and steering angle of articulated vehicle. On this basis, the improved back propagation neural network is used to identify the parameters of the model, and the formula of the weight coefficient adjustment rate based on the Cauchy robust error estimator is derived. Then the controller based on pole assignment method is designed and the feedback control rate of rotation angle is obtained. From the test data, it can be seen that the vehicle lateral position deviation, heading angle deviation and rotation angle control amount are controlled in the range of 0.008 mm-0.07 rad (0.5 掳) 0.21 rad (12 掳) respectively. The deviation in each direction is stable and the error is less than 1%. The research of this path tracking control algorithm can be used as a reference for articulated vehicles.
【作者单位】：北京科技大学机械工程学院;
【基金】：863计划(2011AA060404) 中央高校基本科研业务费(FRF-TP-16-004A1)
【分类号】：TP273

【参考文献】