复杂工况下货运车辆质量和道路坡度联合估计  

Joint Estimation of Freight Vehicle Mass and Road Slope under Complex Conditions

作　　者：高林 吴青[1] 贺宜[2] GAO Lin;WU Qing;HE Yi(School of Transportation and Logistics Engineering,Wuhan University of Technology,Wuhan 430063;Intelligent Transportation Systems Research Center,Wuhan University of Technology,Wuhan 430063)

机构地区：[1]武汉理工大学交通与物流工程学院,武汉430063 [2]武汉理工大学智能交通系统研究中心,武汉430063

出　　处：《机械工程学报》2025年第2期281-295,共15页Journal of Mechanical Engineering

基　　金：国家自然科学基金资助项目(52072292)。

摘　　要：准确估计车辆质量和道路坡度,是实现车辆控制和节能的关键。针对现有道路坡度估计研究大多焦距纵向坡度,对于纵向道路坡度和连续转弯上下坡相耦合场景下的坡度估计研究较为缺乏。为此,首先针对货运车辆加速过程中的质量估计稳定性问题,探究了不同加速类型对质量估计的影响,提出了M估计和基于遗忘因子的递归最小二乘法(Forgetting factor recursive least square,FFRLS)联合估计方法,实现了货运车辆质量的鲁棒估计。进而,基于质量估计结果对复杂工况道路坡度估计问题进行分析,考虑连续转弯上下坡、纵向上下坡等耦合复杂工况,提出了一种自适应多模型融合框架,通过设计基于货车纵向动力学模型的最小模型误差(Minimum model error,MME)准则来构建货车误差补偿模型,解决连续转弯上下坡工况下货运车辆横向轮胎力和大纵向坡度下模型误差问题。通过提出基于卡方分布多分位点数据分类检测的鲁棒容积卡尔曼滤波(Robust cubature Kalman filter,RCKF)估计方法,解决道路坡度估计过程传感器异常误差干扰问题,实现道路坡度估计算法的鲁棒性。结果表明,所提方法具有辨识精度高且鲁棒性好,可以实现复杂工况下的车辆质量和道路坡度鲁棒估计。Freight vehicle mass and road slope estimation are the keys to improving vehicle control and energy conservation.Most of the existing road slope estimation studies focus on the longitudinal slope.There is a lack of research on slope estimation in the coupling scenario of longitudinal road slopes and continuous turning uphill and downhill.Therefore,aiming at the problem of mass estimation stability in the acceleration process of freight vehicles,the influence of different acceleration types on mass estimation is explored.A joint estimation method of M-estimation and forgetting factor recursive least square(FFRLS)is proposed to realize the robust estimation of freight vehicle mass.Furthermore,the problem of road slope estimation under complex conditions is analyzed based on the mass estimation results.Considering the coupled complex conditions,such as continuous turning uphill and downhill,longitudinal uphill and downhill,an adaptive multi-model fusion framework is proposed.By designing the Minimum Model Error(MME)criterion based on the vehicle longitudinal dynamics model,a freight vehicle error compensation model is built to solve the problems of transverse tire force and model error under a large longitudinal slope of freight vehicles under continuous turning uphill and downhill.A robust cubature Kalman filter(RCKF)estimation method based on Chi-square distribution multi-quantile data classification detection is proposed,which can solve the problem of sensor abnormal error interference in road slope estimation and realize the robustness of the road slope estimation algorithm.The results show that the proposed method has high recognition accuracy and satisfying robustness and can achieve a robust estimation of vehicle mass and road slope under complex working conditions.

关 键 词：货运车辆 异常误差 最小模型误差准则 M估计 自适应多模型融合策略 鲁棒容积卡尔曼滤波 

分 类 号：U469[机械工程—车辆工程]