基于模型预测路径积分和输出调节的层级运动控制  

A Cascade Control Scheme for Path Tracking with Model Predictive Path Integral and Output Regulator

作　　者：万航 聂士达 刘辉[1] 张发旺 项昌乐[1] 韩立金[1] Wan Hang;Nie Shida;Liu Hui;Zhang Fawang;Xiang Changle;Han Lijin(School of Mechanical Engineering,Beijing Institute of Technology,Beijing 100081)

机构地区：[1]北京理工大学机械与车辆学院,北京100081

出　　处：《汽车工程》2024年第10期1790-1803,共14页Automotive Engineering

基　　金：国家自然科学基金(52394262)资助。

摘　　要：由于车载平台计算能力的限制,长控制/预测时域和短控制步长之间存在难以调和的矛盾。本文基于时标分离思想推导了车辆平移运动与平面姿态运动解耦的必要条件,使得车辆的平移运动控制器可以获得更长的控制/预测时域,以便于生成类人跟踪轨迹。车辆的平面姿态调整则基于更为精确的动力学模型和更短的控制周期来调节。此外,利用基于并行采样及优化的模型预测路径积分策略来减轻非线性运动规划的计算负担。最后,提出了一种模型预测输出调节器来解决车辆横向动力学中的欠驱动控制问题,并抑制偏航角的稳态误差。理论分析和试验结果表明,所提出方法能够极大地提升计算效率、参数适应性和转向平顺性,在所有验证工况中将横向冲击度平均降低了50%。Due to the limitation of the computing power of the vehicle platform,there is an irreconcilable contradiction between the long-term control/prediction time domain and the short-term control step.In this paper,a necessary condition is derived to decouple the translational motion from yawing motion based on the time-scale separation.Consequently,the translational motion is regulated over an extended control horizon to generate a humanlike tracking trajectory.The yawing motion is regulated based on a more accurate dynamic model and a shorter control cycle.In addition,model predictive path integral(MPPI)strategy is used to mitigate the computational burden of nonlinear motion planning through sampling-based optimization.Finally,a model predictive output regulator is proposed to solve the underactuated control problem in vehicle lateral dynamics and reduce steady-state errors in yaw angel.Theoretical analysis and simulation results show that the proposed method enhances computing efficiency,improves the parameters adaptability and steering smoothness and reduces the lateral jerk by an average of 50%in all driving scenarios.

关 键 词：自动驾驶 路径积分 欠驱动 输出调节 

分 类 号：U463.6[机械工程—车辆工程] TP273[交通运输工程—载运工具运用工程]