基于车载视觉和动力学模型的路面附着系数融合估计方法

doi:10.3901/JME.2024.24.254

机械工程学报 ›› 2024, Vol. 60 ›› Issue (24): 254-264.doi: 10.3901/JME.2024.24.254

扫码分享

基于车载视觉和动力学模型的路面附着系数融合估计方法

杨显通^1,2, 郑玲^1,2, 金彦林^1,2, 张涵柯^1,2, 曾迪^1,2, 冀杰³

1. 重庆大学机械与运载工程学院重庆 400044;
2. 重庆大学高端装备机械传动全国重点实验室重庆 400044;
3. 西南大学工程技术学院重庆 400715

收稿日期:2024-02-11 修回日期:2024-11-08 出版日期:2024-12-20 发布日期:2025-02-01
作者简介:杨显通，男，1998年出生，博士研究生。主要研究方向为智能汽车状态估计与动力学控制。E-mail：xiantongyang@cqu.edu.cn;郑玲(通信作者)，女，1963年出生，博士，教授，博士研究生导师。主要研究方向为智能汽车的环境感知、决策与动力学控制。E-mail：zling@cqu.edu.cn
基金资助:
国家自然科学基金(52375133, 52388102)和和轨道交通运载系统全国重点实验室自主课题(2024RVL-T04)资助项目。

Fusion Estimation Method of Tire-road Friction Coefficient Based on Vehicle Vision and Dynamical Model

YANG Xiantong^1,2, ZHENG Ling^1,2, JIN Yanlin^1,2, ZHANG Hanke^1,2, ZENG Di^1,2, JI Jie³

1. College of Mechanical and Vehicle Engineering, Chongqing University, Chongqing 400044;
2. State Key Laboratory of Mechanical Transmission for Advanced Equipment, Chongqing University, Chongqing 400044;
3. College of Engineering and Technology, Southwest University, Chongqing 400715

Received:2024-02-11 Revised:2024-11-08 Online:2024-12-20 Published:2025-02-01

摘要/Abstract

摘要： 综合车载视觉和车辆动态响应信息，提出一种路面附着系数融合估计算法，以实现路面附着条件变化时对路面附着系数快速准确的估计。首先，标注路面种类语义分割数据集，使用Deeplabv3+算法训练路面种类语义分割模型，根据张正友标定法对相机进行标定，将路面种类语义分割结果映射到车辆坐标系，采用时序加权法对路面种类语义分割结果进行融合，并将其转换为对应的路面附着系数范围。其次，设计状态约束平方根容积卡尔曼滤波(Constrained square-root cubature Kalman filter，CSCKF)算法。最后，构建基于车辆七自由度双轨动力学模型和Brush轮胎模型的系统状态方程和观测方程，将路面语义分割结果作为CSCKF算法的约束边界输入，实现对路面附着系数的估计。结果表明，所提出的路面附着系数融合估计算法在路面附着条件突变时能够快速收敛，并且具有较高的识别精度。

关键词: 路面种类语义分割模型, 融合估计, 状态约束平方根容积卡尔曼滤波

Abstract: A fusion estimation method is proposed to rapidly estimate tire-road friction coefficient based on vehicle vision and dynamic response information. Firstly, the semantic segmentation data set of road type is marked, the semantic segmentation model of road type is trained using the Deeplabv3+ algorithm, the camera is calibrated according to the Zhang Zhengyou calibration method, the results of semantic segmentation of road type are mapped to the vehicle coordinate system, and the results of semantic segmentation of road type are fused using the time-series weighting method. And it is converted to the corresponding range of road adhesion coefficient. Secondly, the state-constrained square-root cubature Kalman filter(CSCKF) algorithm is derived and realized. Finally, the state and observation equation of the system are obtained based on the vehicle dynamics model and the Brush tire model. The road surface semantic segmentation results are used as the constrained boundary input of the CSCKF algorithm to estimate the adhesion coefficient of the road surface. The results show that the fusion estimation algorithm of tire-road friction coefficient can converge quickly and has high recognition accuracy when the road adhesion condition changes abruptly.

Key words: road-type semantic segmentation model, fusion estimation, state-constrained square-root cubature Kalman filter

中图分类号:

U461

杨显通, 郑玲, 金彦林, 张涵柯, 曾迪, 冀杰. 基于车载视觉和动力学模型的路面附着系数融合估计方法[J]. 机械工程学报, 2024, 60(24): 254-264.

YANG Xiantong, ZHENG Ling, JIN Yanlin, ZHANG Hanke, ZENG Di, JI Jie. Fusion Estimation Method of Tire-road Friction Coefficient Based on Vehicle Vision and Dynamical Model[J]. Journal of Mechanical Engineering, 2024, 60(24): 254-264.

参考文献

[1] CHAND A，JAYESH S，BHASI A. Road traffic accidents：An overview of data sources，analysis techniques and contributing factors[J]. Materials Today：Proceedings，2021，47：5135-5141.
[2] WANG Y，HU J，WANG F A，et al. Tire road friction coefficient estimation：Review and research perspectives[J]. Chinese Journal of Mechanical Engineering，2022，35(1)：1-11.
[3] DU Y，LIU C，SONG Y，et al. Rapid estimation of road friction for anti-skid autonomous driving[J]. IEEE Transactions on Intelligent Transportation Systems，2019，21(6)：2461-2470.
[4] LENG B，JIN D，HOU X，et al. Tire-road peak adhesion coefficient estimation method based on fusion of vehicle dynamics and machine vision[J]. IEEE Transactions on Intelligent Transportation Systems，2022，23(11)：21740-21752.
[5] GUSTAFSSON F. Slip-based tire-road friction estimation[J]. Automatica，1997，33(6)：1087-1099.
[6] LEE C，HEDRICK K，YI K. Real-time slip-based estimation of maximum tire-road friction coefficient[J]. IEEE/ASME Transactions on Mechatronics，2004，9(2)：454-458.
[7] GAO X，YU Z，NEUBECK J，et al. Sideslip angle estimation based on input-output linearisation with tire- road friction adaptation[J]. Vehicle System Dynamics，2010，48(2)：217-234.
[8] SHAO L，JIN C，LEX C，et al. Nonlinear adaptive observer for side slip angle and road friction estimation[C]// 2016 IEEE 55th Conference on Decision and Control (CDC). New York：IEEE，2016：6258-6265.
[9] VILLAGRA J，D’ANDRÉA-NOVEL B，FLIESS M，et al. A diagnosis-based approach for tire-road forces and maximum friction estimation[J]. Control Engineering Practice，2011，19(2)：174-184.
[10] ZHANG Z，ZHENG L，WU H，et al. An estimation scheme of road friction coefficient based on novel tyre model and improved SCKF[J]. Vehicle System Dynamics，2022，60(8)：2775-2804.
[11] YAMADA M，UEDA K，HORIBA I，et al. Road surface condition detection technique based on image taken by camera attached to vehicle rearview mirror[J]. Review of Automotive Engineering，2005，26(2)：163-168.
[12] CHEN Y，ZHANG P，WANG S，et al. Image feature based machine learning approach for road terrain classification[C]// 2018 IEEE International Conference on Mechatronics and Automation (ICMA). New York：IEEE，2018：2097-2102.
[13] KIM M H，PARK J，CHOI S. Road type identification ahead of the tire using D-CNN and reflected ultrasonic signals[J]. International Journal of Automotive Technology，2021，22：47-54.
[14] ERDOGAN G，ALEXANDER L，RAJAMANI R. Estimation of tire-road friction coefficient using a novel wireless piezoelectric tire sensor[J]. IEEE Sensors Journal，2010，11(2)：267-279.
[15] ERGUN M，IYINAM S，IYINAM A F. Prediction of road surface friction coefficient using only macro-and microtexture measurements[J]. Journal of Transportation Engineering，2005，131(4)：311-319.
[16] HONG S，ERDOGAN G，HEDRICK K，et al. Tyre-road friction coefficient estimation based on tyre sensors and lateral tyre deflection：Modelling，simulations and experiments[J]. Vehicle System Dynamics，2013，51(5)：627-647.
[17] CHEN L，LUO Y，BIAN M，et al. Estimation of tire-road friction coefficient based on frequency domain data fusion[J]. Mechanical Systems and Signal Processing，2017，85：177-192.
[18] SVENSSON L，TÖRNGREN M. Fusion of heterogeneous friction estimates for traction adaptive motion planning and control[C]// 2021 IEEE International Intelligent Transportation Systems Conference (ITSC). New York：IEEE，2021：424-431.
[19] CHEN L C，ZHU Y，PAPANDREOU G，et al. Encoder- decoder with atrous separable convolution for semantic image segmentation[C]// Proceedings of the European Conference on Computer Vision (ECCV). 2018：801-818.
[20] ZHANG Z. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence，2000，22(11)：1330-1334.
[21] 公安部交通管理科学研究所，公安部道路交通管理标准化技术委员会. 典型交通事故形态车辆行驶速度技术鉴定：GA/T 643—2006[S]. 北京：中国标准出版社，2006. Traffic Management Research Institute of the Ministry of Public Security，Road Traffic Management Standardization Technical Committee of the Ministry of Public Security. The speed technical evaluation for vehicles involved in representative road accidents：GA/T 643—2006[S]. Beijing：Standards Press of China，2006.
[22] SIMON D. Kalman filtering with state constraints：A survey of linear and nonlinear algorithms[J]. IET Control Theory & Applications，2010，4(8)：1303-1318.

基于车载视觉和动力学模型的路面附着系数融合估计方法

Fusion Estimation Method of Tire-road Friction Coefficient Based on Vehicle Vision and Dynamical Model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价