基于高斯混合-隐半马尔可夫模型的双侧独立电驱动无人履带机动平台纵向决策方法

doi:10.12382/bgxb.2021.0680

摘要/Abstract

摘要： 基于运动学和动力学模型的电驱动无人履带机动平台纵向决策研究存在自适应能力差、难以获得精确的模型参数等问题。针对无人履带机动平台直驶遇到障碍物并接近障碍物的行驶工况，根据驾驶数据提出一种驾驶员纵向决策机理，使用高斯混合-隐半马尔可夫模型对熟练驾驶员纵向决策过程进行建模。使用高斯混合模型对驾驶员在直驶接近障碍物过程中的意图进行辨识，并对驾驶行为进行聚类和量化；通过隐半马尔可夫模型描述驾驶员决策转移过程与相同决策的持续时间；进行不同地面条件下的实车验证。试验结果表明：所提出的驾驶员模型可以很好地模仿驾驶员纵向决策机理，使得最大加速度小于3.5 m/s²,最大减速度大于-4.5 m/s²，决策边界平均加速度绝对值趋近于0.8 m/s²；通过对不同地面条件下的决策持续时间分布进行再训练，该方法可以不依赖地面参数从而适应不同环境条件。

关键词: 电驱动履带平台, 无人驾驶, 驾驶员模型, 纵向决策系统

Abstract: Atpresent, theresearch on the kinematics- and dynamics-based longitudinal decision-making system of electric unmanned tracked vehicles are confronted with problems such as poor adaptability and difficulty to obtain accurate model parameters. Aiming at the driving scenarios of the unmanned tracked vehicle straight-linedriving and approaching obstacles, this study introduces the longitudinal decision-making mechanism for driversby analyzing the driving data and constructsa model usingthe combination of Gaussian Mixture Model (GMM) and Hidden Semi-Markov Model (HSMM) to simulate the longitudinal decision-making process of experienced drivers. In the GMM-HSMM system, the GMM is utilized to identify the driving intention as well as cluster and quantifythe driving behavior duringtheobstacle-approachingprocess;the HSMM is applied to model the decision transfer process and the duration of the same decision. This system is verified by a real platform under different road conditions. The experimental results indicate that the proposed driver model canwellsimulate the longitudinal decision-making mechanismfor drivers,where the acceleration is limited to 3.5 m/s², the deceleration is larger than -4.5 m/s², andthe average value of absolute acceleration at the decision boundary approaches 0.8 m/s². Meanwhile, the GMM-HSMM-basedsystem is shown to be able to adapt to different road conditions withoutrelying on accurate road parameters by retraining the decision durationdistribution.

Key words: electrictrackedvehicle, unmanneddriving, drivermodel, longitudinaldecision-makingsystem

中图分类号:

TJ810.1

刘庆霄，唐泽月，张超朋，刘海鸥，陈慧岩. 基于高斯混合-隐半马尔可夫模型的双侧独立电驱动无人履带机动平台纵向决策方法[J]. 兵工学报, 2022, 43(8): 1733-1743.

LIU Qingxiao, TANG Zeyue, ZHANG Chaopeng, LIU Hai'ou, CHEN Huiyan. Research on GMM-HSMM-based Longitudinal Decision-making System for Two-side Independent Electric Unmanned TrackedPlatform[J]. Acta Armamentarii, 2022, 43(8): 1733-1743.

参考文献

［1］陈慧岩,熊光明,龚建伟,等.无人驾驶汽车概论［M］. 北京:北京理工大学出版社, 2014.
CHEN H Y, XIONG G M, GONG J W, et al. Introduction to self-driving car［M］. Beijing: Beijing Institute of Technology Press, 2014. (in Chinese)
［2］孙逢春, 陈树勇, 郭汾. 基于转矩控制策略的电传动履带车辆驱动特性研究［J］. 兵工学报, 2007, 28(2):129-133.
SUN F C, CHEN S Y, GUO F. Research on driving performances of electric transmission tracked vehicles based on torque control strategy［J］. Acta Armamentarii, 2007, 28(2):129-133. (in Chinese)
［3］陈朝萌, 周晓军, 杨辰龙,等. 高速电驱动履带车辆紧急制动控制策略［J］. 浙江大学学报（工学版）, 2020, 54(3):442-449.
CHEN C M, ZHOU X J, YANG C L, et al. Emergency braking control strategy for high speed electric drive tracked vehicle［J］. Journal of Zhejiang University (Engineering Science), 2020, 54(3): 442-449. (in Chinese)
［4］曾庆含, 马晓军, 魏巍,等. 分布式电驱动履带车辆驱动力协调控制策略研究［J］. 兵工学报, 2017, 38(1):9-19.
ZENG Q H, MA X J, WEI W, et al. Research on coordination control strategy of driving force of distributed electric drive tracked vehicle［J］. Acta Armamentarii, 2017, 38(1):9-19. (in Chinese)
［5］刘瑞, 朱西产. 驾驶员加速度分布特性及其应用［J］. 汽车安全与节能学报, 2018, 10(1):37-45.
LIU R, ZHU X C. Acceleration distribution characteristics of the driver and its application［J］. Journal of Automotive Safety and Energy, 2018, 10(1):37-45. (in Chinese)
［6］沈沛鸿, 赵治国, 郭秋伊. 基于ANFIS的工况跟踪用纵向驾驶员模型开发［J］. 汽车工程, 2019, 41(7):815-822.
SHEN P H, ZHAO Z G, GUO Q Y. Development of longitudinal driver model for driving cycle tracking based on ANFIS［J］. Automotive Engineering, 2019, 41(7):815-822. (in Chinese)
［7］ QU T, WANG Q, CONG Y F, et al. Modeling driver's longitudinal and lateral control behavior: a moving horizon control modeling scheme［C］∥Proceedings of 2018 Chinese Control and Decision Conference. Shenyang: CCDC,2018:3260-3264.
［8］ QU T, ZHAO J W, GAO H H, et al. Multi-mode switching-based model predictive control approach for longitudinal autonomous driving with acceleration estimation［J］. IET Intelligent Transport Systems, 2020, 14(14): 2102-2112.
［9］ SEINA K, ARATA E,MASAHIRO O. Development of driver braking control model based on ride comfort index［J］. Artificial Life and Robotics, 2021, 26(3):347-353.
［10］ PENG L Q, SOTELO M A, HE Y, et al. Rough set based method for vehicle collision risk assessment through inferring driver's braking actions in near-crash situations［J］. IEEE Intelligent Transportation Systems Magazine, 2019, 11(2):54-69.
［11］ TIAN Z, LI Y G, CEN M, et al. Multi-vehicle tracking using an environment interaction potential force model［J］. IEEE Sensors Journal, 2020, 20(20):12282-12294.
［12］ TSELENTIS D T, GONIDI C, YANNIS G. Driving speed model development using driving data obtained from smartphone sensors［C］∥Proceedings of World Conference on Transport Research. Mumbai,India: WCTR, 2020:673-686.
［13］ LU C, GONG J W, L C, et al. A personalized behavior learning system for human-like longitudinal speed control of autonomous vehicles［J］. Sensors, 2019, 19(17):3672-3690.
［14］ MUMCUOGLU M E, ALCAN G, UNEL M, et al. Driver evaluation in heavy duty vehicles based on acceleration and braking behaviors［C］∥Proceedings of the 46th Annual Conference of the IEEE Industrial Electronics Society. Singapore: IEEE, 2020:447-452.
［15］ WANG W S, XI J Q, ZHAO D. Learning and inferring a driver's braking action in car-following scenarios［J］. IEEE Transactions on Vehicular Technology, 2018, 67(5):3887-3899.
［16］ LIU Q X,XU S H, LU C, et al. Early recognition of driving intention for lane change based on recurrent hidden semi-Markov model［J］. IEEE Transactions on Vehicular Technology, 2020, 69(10):10545-10557.
［17］王博洋, 龚建伟, 高天云,等. 基于双层驾驶员模型的履带车辆纵向与横向协同跟踪控制方法［J］. 兵工学报, 2018, 39(9): 1675-1682.
WANG B Y, GONG J W, GAO T Y, et al. Longitudinal and lateral path following coordinated control method of tracked vehicle based on double-layer driver model［J］. Acta Armamentarii, 2018,39(9):1675-1682. (in Chinese)
［18］ YU S Z. Hidden semi-Markov models［J］. Artificial Intelligence, 2010, 174(2):215-243.
［19］ DUONG T V, PHUNG D Q, BUI H H, et al. Human behavior recognition with generic exponential family duration modeling in the hidden semi-Markov model［C］∥Proceedings of International Conference on Pattern Recognition.Hong Kong,China: ICPR, 2006:202-207.
［20］ SUN S, WANG X Z, MORAN B, et al. A hidden semi-Markov model for indoor radio source localization using received signal strength［J］. Signal Processing, 2019, 166:107230-107241.

第43卷第8期2022 年8月
兵工学报ACTA ARMAMENTARII
Vol.43No.8Aug.2022

[1]	熊光明, 罗震, 孙冬, 陶俊峰, 唐泽月, 吴超. 基于红外相机和毫米波雷达融合的烟雾遮挡无人驾驶车辆目标检测与跟踪[J]. 兵工学报, 2024, 45(3): 893-906.
[2]	王博洋, 关海杰, 龚建伟, 陈慧岩, 赵卉菁. 面向异构履带车辆的统一运动规划方法[J]. 兵工学报, 2022, 43(2): 241-251.
[3]	张超朋，刘庆霄，董昊天，陈慧岩，席军强. 无人驾驶履带车辆机电联合制动的协调控制[J]. 兵工学报, 2022, 43(11): 2727-2737.
[4]	陈慧岩, 关海杰, 刘海鸥, 龚建伟, 吴贺禹. 履带平台无人驾驶系统基于语义信息的模块串联方法[J]. 兵工学报, 2022, 43(11): 2705-2716.
[5]	张瑞增，龚建伟，陈慧岩，王博洋，刘海鸥. 基于无人驾驶单流传动的中型履带车辆大坡道起步控制[J]. 兵工学报, 2020, 41(8): 1590-1599.
[6]	王博洋，龚建伟，熊光明，张瑞增，陈慧岩，席军强. 有级转向履带车辆的驾驶员操控行为模型[J]. 兵工学报, 2020, 41(12): 2379-2388.
[7]	胡家铭，胡宇辉，陈慧岩，刘凯. 基于模型预测控制的无人驾驶履带车辆轨迹跟踪方法研究[J]. 兵工学报, 2019, 40(3): 456-463.
[8]	王博洋，龚建伟，高天云，张瑞增，陈慧岩，席军强. 基于双层驾驶员模型的履带车辆纵向与横向协同跟踪控制方法[J]. 兵工学报, 2018, 39(9): 1675-1682.
[9]	王博洋，龚建伟，高天云，陈慧岩，席军强. 基于高斯混合-隐马尔可夫模型的速差转向履带车辆横向控制驾驶员模型[J]. 兵工学报, 2017, 38(12): 2301-2308.
[10]	朱敏，陈慧岩，熊光明. 无人驾驶轮式车辆电控气压制动技术研究[J]. 兵工学报, 2015, 36(11): 2017-2023.