基于改进TD3的山地无人作业底盘姿态控制方法*

doi:10.16731/j.cnki.1671-3133.2025.05.001

现代制造工程 ›› 2025, Vol. 536 ›› Issue (5): 1-11.doi: 10.16731/j.cnki.1671-3133.2025.05.001

• 试验研究 • 下一篇

基于改进TD3的山地无人作业底盘姿态控制方法^*

李希明¹, 刘业通¹, 彭世康², 吴湘柠², 李恒强¹, 蒙艳玫¹

1 广西大学机械工程学院,南宁 530004;
2 广西机械工业研究院有限公司,南宁 530001

收稿日期:2024-11-06 出版日期:2025-05-18 发布日期:2025-05-30
通讯作者: 蒙艳玫,E-mail:1361064345@qq.com
作者简介:李希明,硕士研究生,主要研究方向为主动悬架控制策略研究。
基金资助:
*国家自然科学基金项目(52365001);广西科技重大专项项目(桂科AA23062040-3)

Attitude control method for mountain unmanned operation chassis based on improved TD3

LI Ximing¹, LIU Yetong¹, PENG Shikang², WU Xiangning², LI Hengqiang¹, MENG Yanmei¹

1 School of Mechanical Engineering,Guangxi University,Nanning 530004,China;
2 Guangxi Machinery Industry Research Institute Co.,Ltd.,Nanning 530001,China

Received:2024-11-06 Online:2025-05-18 Published:2025-05-30

摘要/Abstract

摘要： 针对山地无人作业底盘在复杂道路下姿态不平稳,传统控制方法适应性、鲁棒性差等问题,提出了一种基于牛顿-拉弗森优化(Newton-Raphson-Based Optimizer,NRBO)算法、极致梯度提升树(eXtreme Gradient Boosting,XGBoost)算法和双延迟深度确定性策略梯度(Twin Delayed Deep Deterministic policy gradient,TD3)算法的底盘姿态控制策略。首先,搭建七自由度主动悬架振动模型环境;然后,训练NRBO-XGBoost的状态预测模型,在TD3算法中加入状态预测模型并在网络中加入注意力机制,增强TD3智能体在复杂环境下的决策能力和适应能力,同时设计奖励函数并训练TD3智能体,实现在复杂道路环境下的底盘姿态控制;最后,基于Matlab 2023a/Simulink软件开展仿真。仿真结果表明,基于改进TD3的底盘姿态控制策略能够有效抑制无人作业底盘在复杂道路下的姿态变化,其俯仰角、侧倾角和垂向位移分别抑制了61.4 %、84.9 %和84.9 %,显著提高了平稳性;相比传统DDPG、PPO和TD3强化学习控制策略,改进TD3算法下的俯仰角分别改善了49.1 %、7.4 %和37.2 %,侧倾角分别改善了83.3 %、36.5 %和34.7 %,垂向位移分别改善了70.7 %、77.5 %和64.0 %,垂向位移加速度分别改善了67.7 %、42.1 %和49.7 %,控制效果更好,具有更好的适应性与鲁棒性。

关键词: 山地无人作业底盘, 主动悬架控制, 改进TD3算法, 自注意力机制

Abstract: Aiming at the problems of unstable attitude of mountain unmanned operation chassis under complex roads and poor adaptability and robustness of traditional control methods,a chassis attitude control strategy based on Newton-Raphson-Based Optimizer (NRBO) algorithm,eXtreme Gradient Boosting (XGBoost) algorithm and Twin Delayed Deep Deterministic policy gradient (TD3) algorithm was proposed. Firstly,the seven-degree-of-freedom active suspension vibration model environment was built;then the state prediction model of NRBO-XGBoost was trained,the state prediction model was added to the TD3 algorithm and the attention mechanism was added to the network to enhance the decision-making ability and adaptive ability of the TD3 intelligences in complex environments,and at the same time,the reward function was designed and the TD3 intelligences were trained to realize the chassis attitude control in complex road environments; finally,simulations were carried out based on Matlab 2023a/Simulink software. The simulation results show that the chassis attitude strategy method based on the improved TD3 can effectively suppress the attitude change of unmanned operation chassis under complex roads,and the pitch angle,lateral inclination angle,and vertical displacement are suppressed by 61.4 %,84.9 %,and 84.9 %,respectively,which significantly improves the smoothness;compared with the traditional DDPG,PPO,and TD3 reinforcement learning control strategies,with the improved TD3 algorithm the pitch angle is improved by 49.1 %,7.4 % and 37.2 %,respectively,the lateral inclination angle is improved by 83.3 %,36.5 % and 34.7 %,respectively,the vertical displacement is improved by 70.7 %,77.5 % and 64.0 %,respectively,and the vertical displacement acceleration is improved by 67.7 %,42.1 % and 49.7 %,respectively,which provides a better control effect with better adaptability and robustness.

Key words: mountain unmanned operation chassis, active suspension control, improved TD3 algorithm, self-attention mechanism

中图分类号:

TP273

李希明, 刘业通, 彭世康, 吴湘柠, 李恒强, 蒙艳玫. 基于改进TD3的山地无人作业底盘姿态控制方法^*[J]. 现代制造工程, 2025, 536(5): 1-11.

LI Ximing, LIU Yetong, PENG Shikang, WU Xiangning, LI Hengqiang, MENG Yanmei. Attitude control method for mountain unmanned operation chassis based on improved TD3[J]. Modern Manufacturing Engineering, 2025, 536(5): 1-11.

参考文献

[1] 郭庆.浅析如何补齐丘陵山区农机化短板[J].农机质量与监督,2023(10):14,33.
[2] 王玮举,李华,郑荣华.发展智能农机助力丘陵山区农业现代化[J].农机科技推广,2020(11):53-55.
[3] 吕凤玉,李晓康,贺成柱,等.全向姿态调整农机履带底盘设计与试验[J].中国农机化学报,2024,45(8):132-137.
[4] 赖晓,程健华,李尚平,等.丘陵履带式甘蔗收获机底盘调平机构设计与试验[J].农业机械学报,2024,55(12):100-109.
[5] 刘凤霞,逄焕晓,翟国强,等.丘陵山区单行玉米收获机自适应姿态调整系统研究[J/OL].农机化研究:1-8[2024-10-29].http://kns.cnki.net/kcms/detail/23.1233.S.20241012.1503.002.html.
[6] 杨怡婷,李广棵,吴磊.车辆半主动悬架模糊变权重因子自适应控制研究[J].现代制造工程,2024(9):73-82.
[7] 张艳兵,徐鹏跃,李卓,等.农用机器人底盘控制系统的设计[J/OL].农机化研究:1-6[2024-10-29].https://doi.org/10.13427/j.issn.1003-88X.2025.06.032.
[8] 张喜清,王博,连晋毅,等.模糊PID控制的步履式底盘姿态自适应调整研究[J].工程机械,2023,54(9):50-57,9.
[9] 刘平义,彭凤娟,李海涛,等.丘陵山区农用自适应调平底盘设计与试验[J].农业机械学报,2017,48(12):42-47.
[10] 吴伟斌,李泽艺,洪添胜,等.基于山地果园路谱的轮式运输车钢板弹簧悬架优化设计[J].华中农业大学学报,2018,37(4):7-14.
[11] 董朝闻,黄龙.基于模糊控制的汽车半主动悬架研究[J].工程机械,2024,55(2):60-64,8.
[12] 金贤建,王佳栋,徐利伟,等.轮毂电机驱动电动汽车主动悬架μ综合鲁棒控制研究[J].机械工程学报,2024,60(16):259-269.
[13] 贾继良,赵清海,杨景周,等.汽车半主动座椅悬架自适应模糊神经滑模控制[J].机械设计,2024,41(4):28-35.
[14] 高晋,李晖,杨秀建.7自由度半主动悬架整车模型PID-LQR控制研究[J].车辆与动力技术,2023(3):1-6,23.
[15] 王刚,李昆鹏,景晖,等.基于Q学习的整车主动悬架免参数H_∞控制[J].汽车工程,2023,45(12):2260-2271.
[16] LEE D,JIN S,LEE C. Deep reinforcement learning of semi-active suspension controller for vehicle ride comfort[J]. IEEE Transactions on Vehicular Technology,2022,72(1):327-339.
[17] FARES A,BANI Y A. Online reinforcement learning-based control of an active suspension system using the actor critic approach[J]. Applied Sciences,2020,10(22):8060-8073.
[18] HAN S Y,LIANG T. Reinforcement-learning-based vibration control for a vehicle semi-active suspension system via the PPO approach[J]. Applied Sciences,2022,12(6):3078-3095.
[19] LIU M,LI Y,RONG X,et al. Semi-active suspension control based on deep reinforcement learning[J]. IEEE Access,2020,8:9978-9986.
[20] 陈盟,龙海洋,琚立颖,等.随机路面时域模型的建模与仿真[J].机械工程与自动化,2017,52(2):40-41.
[21] 黄健.丘陵山地拖拉机路面谱测试及再现方法研究[D].长春:吉林大学,2018.
[22] 刘国辉,郝称意,李民赞,等.半主动悬架山地拖拉机姿态控制系统设计与仿真[J].农业机械学报,2022,53(S2):338-348.
[23] SOWMYA R,PREMKUMAR M,JANGIR P. Newton-Raphson-based optimizer:A new population-based metaheuristic algorithm for continuous optimization problems[J]. Engineering Applications of Artificial Intelligence,2024,128:107532.
[24] CHEN T,GUESTRIN C. Xgboost:A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.[S.l.]:[s.n.],2016:785-794.
[25] YU M,EVANGELOU S A,DINID. Advances in Active Suspension Systems for Road Vehicles[J]. Engineering,2024,33:160-177.
[26] SHEN D. A Study on Active Suspension System with Reinforce-ment Learning[D]. Sydney:University of Technology,2022.
[27] 潘公宇,范菲阳,冯鑫.基于主动悬架的整车车身姿态控制策略研究[J].电子测量技术,2024,47(2):79-88.

基于改进TD3的山地无人作业底盘姿态控制方法^*

Attitude control method for mountain unmanned operation chassis based on improved TD3

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	曹胜杰, 吴海, 程壹涛, 任泽生. 执行器故障下机械臂有限时间容错控制[J]. 现代制造工程, 2025, 536(5): 82-90.
[2]	周国诚, 陶翼飞, 何毅, 李立山, 吴佳兴. 自动装箱系统中多AGV在线调度问题研究^*[J]. 现代制造工程, 2025, 533(2): 17-25.
[3]	靳俊霞, 王彩玲, 于峰. 轮式移动工业机器人非奇异终端滑模轨迹跟踪控制^*[J]. 现代制造工程, 2024, 531(12): 61-68.
[4]	项亚南, 邓小龙, 潘丰. 锂电池电芯电极自动装配系统的设计与实现^*[J]. 现代制造工程, 2024, 528(9): 136-143.
[5]	梁彪, 周德强, 盛卫锋, 左文娟, 何长江, 奚青, 陈曲燕. 基于分数阶滑模控制的AGV纠偏系统研究^*[J]. 现代制造工程, 2024, 527(8): 51-60.
[6]	刘宇, 韩钧, 曲嘉伟, 张诗奇, 陈海静. 基于信噪比和灰色关联分析的电火花喷爆加工工艺参数优化^*[J]. 现代制造工程, 2024, 526(7): 9-16.
[7]	杨怀磊, 姚云磊. 基于故障观测器的机器人多关节手臂最优容错控制^*[J]. 现代制造工程, 2024, 526(7): 61-68.
[8]	李鑫帅, 毛汉领, 王鹏康, 李思言. 无人战车轮毂电机稳定性控制策略研究^*[J]. 现代制造工程, 2024, 525(6): 9-14.
[9]	张俊, 胡俊峰. 基于视觉的软体机器人位姿测量与控制^*[J]. 现代制造工程, 2024, 524(5): 39-47.
[10]	孙建民, 杨世虎, 赵磊, 姚德臣. 基于动态惯性权重的电子节气门改进PSO-BP优化控制^*[J]. 现代制造工程, 2024, 521(2): 45-52.
[11]	刘阳，徐巧玉，陈翰文，张正. 凿岩机械臂轨迹规划及电液跟踪控制研究[J]. 现代制造工程, 2023, 519(12): 61-67.
[12]	王洪波;姚嘉凌. 桶装水码垛机器人关节空间最优运动控制[J]. 现代制造工程, 2023, 518(11): 33-39.
[13]	刘琛华. 基于S型速度曲线的液压伺服绞车自抗扰控制[J]. 现代制造工程, 2023, 517(10): 145-154.
[14]	郑凯强;李薇. 重载液压平台的论域自适应模糊耦合同步控制[J]. 现代制造工程, 2023, 515(8): 141-147.
[15]	楚雪平. 移动装配机器人手臂的滑模鲁棒控制方法[J]. 现代制造工程, 2023, 513(6): 39-45.