气垫车辆土壤参数估值算法

上海交通大学学报（自然版） ›› 2011, Vol. 45 ›› Issue (04): 451-456.

气垫车辆土壤参数估值算法

许烁, 喻凡, 罗哲, Pham D T, 纪赜

（1.上海交通大学机械与动力工程学院，上海 200240；2.卡迪夫大学制造工程中心，英国卡迪夫 CF24 3LA）

出版日期:2011-04-29 发布日期:2011-04-29
基金资助:
国家自然科学基金资助项目（50675135）

An Estimation Algorithm for Soil Parameters for Air-Cushion Vehicles

XU Shuo, YU Fan, LUO Zhe, Pham D T, JI Ze

(1. School of Mechanical Engineering, Shanghai Jiaotong University, Shanghai 200240, China; 2. Manufacturing Engineering Centre, Cardiff University, Cardiff, CF24 3LA, UK)

Online:2011-04-29 Published:2011-04-29

摘要/Abstract

摘要： 针对气垫车辆具有附加的垂向力控制自由度的特点，提出了基于g函数和扩展卡尔曼滤波器（EKF）联合算法的土壤参数估值方案.g函数通过解耦土壤参数解决了多解问题，EKF通过减小测量不确定度提高了估值准确性.通过不同噪声水平下算法准确性和大噪声重复实验下算法稳定性的实验表明：g-EKF联合算法在各种噪声水平下具有应用鲁棒性，特别是在大噪声工况下，其估值准确性、精确性和稳定性具有明显优势.

关键词: 土壤参数估值, g-EKF算法, 气垫车, 状态噪声, 测量噪声

Abstract: By taking advantage of the additional degree of control freedom for vertical force, a new estimation algorithm, the hybrid g-EKF algorithm, was proposed for air-cushion vehicles. In g-EKF, the g-function solves the multi-solution problem, and the extended Kalman filter (EKF) improves estimation accuracy by decreasing measurement uncertainties. Its advantages are demonstrated in two experiments, namely, estimation accuracy for different noise levels and stability in repeated tests.

Key words: soil parameter estimation, g-EKF algorithm, air-cushion vehicle, state noise, measurement noise

中图分类号:

U461.5

许烁, 喻凡, 罗哲, Pham D T, 纪赜. 气垫车辆土壤参数估值算法[J]. 上海交通大学学报（自然版）, 2011, 45(04): 451-456.

XU Shuo, YU Fan, LUO Zhe, Pham D T, JI Ze. An Estimation Algorithm for Soil Parameters for Air-Cushion Vehicles[J]. Journal of Shanghai Jiaotong University, 2011, 45(04): 451-456.

参考文献

［1］Bekker M G. Introduction to terrainvehicle systems ［M］. Ann Arbour: University of Michigan Press, 1969.

［2］Wong J Y. Theory of ground vehicles［M］. 3rd ed. New York: John Wiley & Sons, Inc, 2001.

［3］Hutangkabodee S, Zweiri Y H, Seneviratne L D, et al. Multisolution problem for trackterrain interaction dynamics and lumped soil parameter identification［C］//Springer Tracts in Advanced Robotics. Berlin: Springer, 2006, 25: 517528.

［4］Ray L E. Estimation of terrain forces and parameters for rigidwheeled vehicles［J］. IEEE Transactions on Robotics, 2009, 25(3): 717726.

［5］Le A T, Rye D C, DurrantWhyte H F. Estimation of tracksoil interactions for autonomous tracked vehicles［C］//Proceedings of 1997 IEEE International Conference on Robotics and Automation. New York: IEEE Press, 1997: 13881393.

［6］Iagnemma K, Kang S, Shibly H, et al. Online terrain parameter estimation for wheeled mobile robots with application to planetary rovers ［J］. IEEE Transactions on Robotics, 2004, 20(5): 921927.

［7］Hutangkabodee S, Zweiri Y H, Seneviratne L D, et al. Soil parameter identification for wheelterrain interaction dynamics and traversability prediction［J］. International Journal of Automation and Computing, 2006, 3(3): 244251.

［8］Li L, Sandu C. On the impact of cargo weight, vehicle parameters, and terrain characteristics on the prediction of traction for offroad vehicles［J］. Journal of Terramechanics, 2007, 44(3): 221238.

［9］Golda D, Iagnemma K, Dubowsky S. Probabilistic modeling and analysis of highspeed roughterrain mobile robots［C］//Proceedings of 2004 IEEE International Conference on Robotics and Automation. New York: IEEE Press, 2004: 914919.

［10］罗哲, 喻凡, 许烁, 等. 半履带气垫车［P］. 中国发明专利: CN101007531A. 20070118.

［11］许烁, 罗哲, 周科. 基于功耗最优的半履带气垫车模糊PID控制仿真［J］. 上海交通大学学报, 2007, 41(6): 10261030.

XU Shuo, LUO Zhe, ZHOU Ke. Fuzzy PID control simulation for a semitrack aircushion vehicle based on power consumption optimization［J］. Journal of Shanghai Jiaotong University, 2007, 41(6): 10261030.

［12］Dumond D A, Ray L, Trautmann E. Evaluation of terrain parameter estimation using a stochastic terrain model［C］//Proceedings of SPIE. New York: SPIE Press, 2009, 7332: 73321F.

[1]	周林阳，王生捷. 基于反正切非线性函数的自抗扰控制[J]. 上海交通大学学报（自然版）, 2013, 47(07): 1043-1048.
[2]	许烁1, 罗哲2, 喻凡2. 气垫车并行双环控制系统设计[J]. 上海交通大学学报（自然版）, 2012, 46(08): 1284-1290.