基于混沌理论的无人驾驶车辆行驶轨迹量化分析

发布时间：2018-05-28 21:02

本文选题：无人驾驶车辆 + 混沌　；参考：《机械工程学报》2016年02期

【摘要】：针对无人驾驶车辆客观量化评价困难的问题,提出基于混沌理论的无人驾驶车辆行驶轨迹的量化分析方法。应用五次多项式方法结合优秀驾驶员驾驶车辆行驶的初始状态和目标状态设计无人驾驶车辆的理想轨迹,计算实际行驶轨迹与理想轨迹的偏差得到偏差时间数据序列;采用混沌理论的C-C方法对偏差时间数据序列重构相空间;计算偏差时间数据序列的Lyapunov指数,实现无人驾驶车辆行驶轨迹的量化表示。运用提出的方法对无人驾驶车辆避障换道的行驶轨迹偏差时间数据序列进行相空间重构,计算其Lyapunov指数为正值,表征人驾驶车辆系统的混沌性,还表征人驾驶车辆收敛到稳态响应的快慢程度。试验表明基于混沌理论的无人驾驶车辆行驶轨迹的量化分析是可行的,有效的。
[Abstract]:In view of the difficulty of objective quantitative evaluation of driverless vehicles, a quantitative analysis method of driverless vehicle trajectory based on chaos theory is proposed. The ideal trajectory of driverless vehicle is designed by combining the initial state and the target state of the driving vehicle with the method of quintic polynomial. The deviation between the actual track and the ideal track is calculated and the deviation time series is obtained. The C-C method of chaos theory is used to reconstruct the phase space and calculate the Lyapunov exponent of the deviation time data series to realize the quantization of the driving trajectory of driverless vehicles. The proposed method is used to reconstruct the time series of track deviation of driverless vehicles for obstacle avoidance, and the Lyapunov exponent is calculated to be positive, which indicates the chaos of driving vehicle system. It also shows the speed of the steady state response of the driving vehicle. Experiments show that the quantitative analysis of driverless vehicle trajectory based on chaos theory is feasible and effective.
【作者单位】：河北工程大学机电学院;北京理工大学智能车辆研究所;
【基金】：国家自然科学基金重点(90920304);国家自然科学基金培育(91120010) 河北省教育厅(QN2015129)资助项目
【分类号】：U463.6;U495

【相似文献】