基于數據驅動(dòng)的冗余機器人末端執行器位姿控制方案

金龍; 張凡; 劉佰陽(yáng); 鄭宇

doi:10.16383/j.aas.c230273

基于數據驅動(dòng)的冗余機器人末端執行器位姿控制方案

doi: 10.16383/j.aas.c230273

1.
蘭州大學(xué)信息科學(xué)與工程學(xué)院蘭州 730000
2.
騰訊科技(深圳)有限公司Robotics X 深圳 518057

基金項目: 國家自然科學(xué)基金 (62176109), 甘肅省自然科學(xué)基金杰出青年項目 (21JR7RA531), 中央高?；究蒲袠I(yè)務(wù)費 (lzujbky-2023-ct05, lzujbky-2023-ey07), 甘肅省教育廳優(yōu)秀研究生“創(chuàng )新之星”項目 (2023CXZX-072), 騰訊Robotics X犀牛鳥(niǎo)專(zhuān)項研究計劃 (2021-01), 蘭州大學(xué)超算中心資助

詳細信息

作者簡(jiǎn)介:
金龍：蘭州大學(xué)信息科學(xué)與工程學(xué)院教授. 主要研究方向為神經(jīng)網(wǎng)絡(luò ), 機器人技術(shù)和智能信息處理. 本文通信作者. E-mail: jinlongsysu@foxmail.com

張凡：蘭州大學(xué)信息科學(xué)與工程學(xué)院碩士研究生. 主要研究方向為模型預測控制, 機器人技術(shù)和優(yōu)化. E-mail: zhangfanas@foxmail.com

劉佰陽(yáng)：2023年獲得蘭州大學(xué)信息科學(xué)與工程學(xué)院碩士學(xué)位. 主要研究方向為機器人技術(shù)和神經(jīng)網(wǎng)絡(luò ). E-mail: baiyang-liu@foxmail.com

鄭宇：騰訊科技(深圳)有限公司Robotics X首席研究員. 主要研究方向為多體機器人系統, 機器人抓取與操作和機器人算法. E-mail: petezheng@tencent.com

計量
- 文章訪(fǎng)問(wèn)數: 3834
- HTML全文瀏覽量: 227
- PDF下載量: 277
- 被引次數: 0
出版歷程
- 收稿日期: 2023-05-11
- 錄用日期: 2023-08-29
- 網(wǎng)絡(luò )出版日期: 2023-12-27
- 刊出日期: 2024-03-29

Position and Orientation Control Scheme for End-effector of Redundant Manipulators Based on Data-driven Technology

JIN Long^1
,,
ZHANG Fan^1
,,
LIU Bai-Yang^1
,,
ZHENG Yu^2
,

1.
School of Information Science and Engineering, Lanzhou University, Lanzhou 730000
2.
Robotics X, Tencent Technology (Shenzhen) Company Limited, Shenzhen 518057

Funds: Supported by National Natural Science Foundation of China (62176109), Natural Science Foundation of Gansu Province (21JR7RA531), Fundamental Research Funds for the Central Universities (lzujbky-2023-ct05, lzujbky-2023-ey07), Education Department of Gansu Province: Excellent Graduate Student “Innovation Star” Project (2023CXZX-072), CIE-Tencent Robotics X Rhino-Bird Focused Research Program (2021-01), and Supercomputing Center of Lanzhou University

More Information

Author Bio:
JIN Long　Professor at the School of Information Science and Engineering, Lanzhou University. His research interest covers neural networks, robotics, and intelligent information processing. Corresponding author of this paper

ZHANG Fan　Master student at the School of Information Science and Engineering, Lanzhou University. His research interest covers model predictive control, robotics, and optimization

LIU Bai-Yang　Received his master degree from the School of Information Science and Engineering, Lanzhou University in 2023. His research interest covers robotics and neural network

ZHENG Yu　Principal researcher at Robotics X, Tencent Technology (Shenzhen) Company Limited. His research interest covers multibody robotic systems, robotic grasping and manipulation, and algorithms for robotics

摘要

摘要: 模型未知的冗余機器人執行任務(wù)的過(guò)程中會(huì )產(chǎn)生較大的控制誤差, 其末端執行器的位置與姿態(tài)也需要針對不同任務(wù)進(jìn)行修正. 為解決該問(wèn)題, 提出一種基于數據驅動(dòng)的冗余機器人末端執行器位置與姿態(tài)控制方案. 該方案使用在線(xiàn)學(xué)習技術(shù), 能夠應用于模型未知的冗余機器人控制. 同時(shí)引入四元數表示法將控制機器人末端執行器姿態(tài)問(wèn)題轉化為基于四元數表示的控制方法. 隨后, 設計一種神經(jīng)動(dòng)力學(xué)求解器對所提方案進(jìn)行求解. 相關(guān)的理論分析、仿真及對比體現了所提方案的可行性、有效性與新穎性.
- 冗余機器人 /
- 數據驅動(dòng) /
- 位姿控制 /
- 軌跡跟蹤
Abstract: A redundant manipulator with unknown models produces a large control error during a task execution, and the position and orientation of its end-effector need to be corrected for different tasks. To solve this problem, a position and orientation control scheme for the end-effector of a redundant manipulator is proposed based on a data-driven technology. The proposed scheme utilizes an online learning technology, which is able to be applied to control a redundant manipulator with unknown models. By introducing the quaternion representation, the rotation matrix controlling the orientation of the end-effector of a redundant manipulator is transformed into a quaternion representation control method. In addition, a neural dynamics solver is designed to solve the proposed scheme. Theoretical analysis, simulations, and comparisons demonstrate the feasibility, validity, and novelty of the proposed scheme.
- Redundant manipulators /
- data-driven technology /
- position and orientation control /
- trajectory tracking

HTML全文

圖 1 采用所提方案(14)實(shí)現冗余機器人末端執行器位置跟蹤與姿態(tài)保持的仿真結果

Fig. 1 Simulation results of the redundant manipulator using the proposed scheme (14) to achieve position tracking and orientation maintenance

下載: 全尺寸圖片幻燈片

圖 2 采用所提方案(14)實(shí)現冗余機器人位置與姿態(tài)跟蹤的仿真結果

Fig. 2 Simulation results of the redundant manipulator using the proposed scheme (14) to achieve position and orientation tracking

下載: 全尺寸圖片幻燈片

圖 3 基于CoppeliaSim平臺冗余機器人實(shí)現位置與姿態(tài)跟蹤的對比結果

Fig. 3 Comparison results of the redundant manipulator achieving position and orientation tracking based on CoppeliaSim platform

下載: 全尺寸圖片幻燈片

表 1 所提冗余機器人控制方案的符號含義

Table 1 Definitions of variables of the proposed scheme for redundant manipulators

符號	含義
$ {{\boldsymbol{\theta}}} \in {\bf{R}}^a $	機器人關(guān)節角向量
$ \dot{\boldsymbol{\theta}}\in {\bf{R}}^a $	機器人關(guān)節角速度向量
$ \dot{\boldsymbol{\theta}}^{-}(\dot{\boldsymbol{\theta}}^{+}) $	關(guān)節角速度的下界(上界)
$ {\boldsymbol r}\in {\bf{R}}^b $	末端執行器的位置向量
$ \boldsymbol{r}^rf50c1hsl6\in {\bf{R}}^b $	末端執行器的期望位置向量
$ \dot{\boldsymbol r}\in {\bf{R}}^b $	末端執行器的速度向量
$ \dot{\hat{\boldsymbol r}}\in {\bf{R}}^b $	末端執行器的估計速度向量
$ f(\cdot): {\bf{R}}^a \rightarrow {\bf{R}}^b $	機器人非線(xiàn)性前向運動(dòng)學(xué)映射
$ J=\dfrac{\partial f({{\boldsymbol{\theta}}})}{\partial {{\boldsymbol{\theta}}}}\in {\bf{R}}^{b\times a} $	機器人雅可比矩陣
$ \hat{J}\in {\bf{R}}^{b\times a} $	機器人估計雅可比矩陣
$ {\dot{\hat{J}}}\in {\bf{R}}^{b\times a} $	機器人估計雅可比矩陣的導數
$ M(\boldsymbol \theta)\in {\bf{R}}^{3\times 3} $	末端執行器的方向旋轉矩陣
$ {\boldsymbol q}_{E}(\boldsymbol \theta)\in {\bf{R}}^{4} $	末端執行器的方向四元數
$ \boldsymbol{\overline{o}}(\boldsymbol \theta)\in {\bf{R}}^{5} $	末端執行器的方向向量
$ \tilde{\boldsymbol q}\in {\bf{R}}^{5} $	末端執行器的期望方向向量
$ H({\boldsymbol \theta})=\dfrac{\partial{\boldsymbol q}_{E}(\boldsymbol \theta)}{\partial{\boldsymbol \theta}}\in {\bf{R}}^{4\times a} $	$ {\boldsymbol q}_{E} $ 的雅可比矩陣
$ G({\boldsymbol{\theta}})=\dfrac{\partial{\boldsymbol{\overline{o}}({\boldsymbol{\theta}}})}{\partial{{\boldsymbol{\theta}}}}\in {\bf{R}}^{5\times a} $	$ \boldsymbol{\overline{o}}({\boldsymbol{\theta}}) $的雅可比矩陣
$ \kappa(\boldsymbol q)=\dfrac{\partial{{\tilde{\boldsymbol q}}}}{\partial{\boldsymbol q}}\in {\bf{R}}^{5\times 4} $	$ \tilde{\boldsymbol q} $ 的雅可比矩陣
$ \boldsymbol{u}\in {\bf{R}}^a $	方差為極小值的獨立同分布零均值隨機噪聲
${\boldsymbol{u} }_{0}\in {\bf{R} }^a$	$ \boldsymbol{u} $的上界
$ \hat{\dot{{\boldsymbol{\theta}}}}\in {\bf{R}}^a $	受噪聲驅動(dòng)的關(guān)節角速度
$ \Vert \cdot \Vert_2 $	向量的二范數
$ \mathrm{tr(\cdot)} $	矩陣的跡

下載: 導出CSV

表 2 冗余機器人不同軌跡跟蹤控制方案對比

Table 2 Comparison of different trajectory tracking control schemes for redundant manipulators

	方案層級	末端控制	結構信息	位置誤差(m)	姿態(tài)誤差
本文	速度層	位姿	未知	1.653 × 10^?3	3.956 × 10^?3
文獻[13]	速度層	姿態(tài)保持	未知	1.056 × 10^?3	4.635 × 10^?4
文獻[22]	加速度層	位置	已知	3.312 × 10^?3	—
文獻[23]	加速度層	位置	已知	1.423 × 10^?3	—
文獻[24]	速度層	位置	已知	2.734 × 10^?3	—
文獻[25]	速度層	位姿	已知	1.374 × 10^?3	3.461 × 10^?4

下載: 導出CSV

<button id="qm3rj"><thead id="qm3rj"></thead></button>

亚洲第一网址_国产国产人精品视频69_久久久久精品视频_国产精品第九页

參考文獻(25)

[1]	胡靜. 冗余自由度機器人的路徑規劃研究[碩士學(xué)位論文], 東南大學(xué), 中國, 2017. Hu Jing. Research on Path Planning of Redundant Robot [Master thesis], Southeast University, China, 2017.
[2]	李亞昕, 王國磊, 張劍輝, 田鑫亮, 安靜, 陳懇. 基于碰撞反饋的冗余機器人避障規劃算法. 清華大學(xué)學(xué)報(自然科學(xué)版), 2022, 62(03): 408-415 doi: 10.16511/j.cnki.qhdxxb.2021.25.022 Li Ya-Xin, Wang Guo-Lei, Zhang Jian-Hui, Tian Xin-Liang, An Jing, Chen Ken. Obstacle avoidance alorithm for redundant robots based on collision feedback. Journal of Tsinghua University(Science and Technology), 2022, 62(03): 408-415 doi: 10.16511/j.cnki.qhdxxb.2021.25.022
[3]	劉美嬌. 空間超冗余四足爬行機器人軌跡規劃方法研究[博士學(xué)位論文], 中國科學(xué)院大學(xué), 中國, 2023. Liu Mei-Jiao. Research on Trajectory Planning Method of Spatial Hyper-Redundant Quadruped Crawling Robot [Ph.D. dissertation], University of Chinese Academy of Sciences, China, 2023.
[4]	梁旭, 蘇婷婷, 侯增廣, 劉圣達, 章杰, 何廣平. 基于變阻抗控制的冗余驅動(dòng)并聯(lián)機器人多目標內力優(yōu)化. 自動(dòng)化學(xué)報, 2023, 49(5): 1099-1115 doi: 10.16383/j.aas.c210963 Liang Xu, Su Ting-Ting, Hou Zeng-Guang, Liu Sheng-Da, Zhang Jie, He Guang-Ping. A multi-objective internal preload optimization method of redundantly actuated parallel robots based on variable impedance control. Acta Automatica Sinica, 2023, 49(5): 1099-1115 doi: 10.16383/j.aas.c210963
[5]	鄺禹聰. 面向實(shí)驗教學(xué)的六自由度機器人開(kāi)發(fā)[碩士學(xué)位論文], 華南理工大學(xué), 中國, 2017. Kuang Yu-Cong. Develpment of 6-DOF Robot for Experiment Teaching [Master thesis], South China University of Technology, China, 2017.
[6]	Xu Zhi-Hao, Zhou Xue-Feng, Wu Hong-Min, Li Xiao-Xiao, Li Shuai. Motion planning of manipulators for simultaneous obstacle avoidance and target tracking: An RNN approach with guaranteed performance. IEEE Transactions on Industrial Electronics, 2022, 69(4): 3887-3897 doi: 10.1109/TIE.2021.3073305
[7]	Thakar Shantanu, Rajendran Pradeep, Kabir Ariyan, Gupta Satyandra. Manipulator motion planning for part pickup and transport operations from a moving base. IEEE Transactions on Automation Science and Engineering, 2022, 19(1): 191-206 doi: 10.1109/TASE.2020.3020050
[8]	Xie Zheng-Tai, Jin Long, Luo Xin, Sun Zhong-Bo, Liu Mei. RNN for repetitive motion generation of redundant robot manipulators: An orthogonal projection-based scheme. IEEE Transactions on Neural Networks and Learning Systems, 2022 33(2): 615-628 doi: 10.1109/TNNLS.2020.3028304
[9]	李中奇, 周靚, 楊輝. 高速動(dòng)車(chē)組數據驅動(dòng)無(wú)模型自適應控制方法. 自動(dòng)化學(xué)報, 2023, 49(2): 437-447 doi: 10.16383/j.aas.c211068 Li Zhong-Qi, Zhou Liang, Yang Hui. Data-driven model-free adaptive control method for high-speed electric multiple unit. Acta Automatica Sinica, 2023, 49(2): 437-447 doi: 10.16383/j.aas.c211068
[10]	Lee Uichin, Jung Gyuwon, Ma Eun-Yeol, Kim Jin-San, Kim Heepyung, Alikhanov Jumabek, Noh Youngtae, Kim Heeyoung. Toward data-driven digital therapeutics analytics: Literature review and research directions. IEEE/CAA Journal of Automatica Sinica, 2023, 10(1): 42-66 doi: 10.1109/JAS.2023.123015
[11]	梁正平, 黃錫均, 李燊鈿, 王喜瑜, 朱澤軒. 基于剪枝堆棧泛化的離線(xiàn)數據驅動(dòng)進(jìn)化優(yōu)化. 自動(dòng)化學(xué)報, 2023, 49(6): 1306-1325 doi: 10.16383/j.aas.c220387 Liang Zheng-Ping, Huang Xi-Jun, Li Shen-Tian, Wang Xi-Yu, Zhu Ze-Xuan. Offline data driven evolutionary optimization based on pruning stacked generalization. Acta Automatica Sinica, 2023, 49(6): 1306-1325 doi: 10.16383/j.aas.c220387
[12]	姜藝, 范家璐, 柴天佑. 數據驅動(dòng)的保證收斂速率最優(yōu)輸出調節. 自動(dòng)化學(xué)報, 2022, 48(4): 980-991 doi: 10.16383/j.aas.c200932 Jiang Yi, Fan Jia-Lu, Chai Tian-You. Data-driven optimal output regulation with assured convergence rate. Acta Automatica Sinica, 2022, 48(4): 980-991 doi: 10.16383/j.aas.c200932
[13]	Fan Jia-Liang, Jin Long, Xie Zheng-Tai, Li Shuai, Zheng Yu. Data-driven motion-force control scheme for redundant manipulators: A kinematic perspective. IEEE Transactions on Industrial Informatics, 2022, 18(8): 5338-5347 doi: 10.1109/TII.2021.3125449
[14]	Xie Zheng-Tai, Jin Long, Luo Xin, Hu Bin, Li Shuai. An acceleration-level data-driven repetitive motion planning scheme for kinematic control of robots with unknown structure. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2022, 52(9): 5679-5691 doi: 10.1109/TSMC.2021.3129794
[15]	Jin Long, Zheng Xin. Neural dynamics for distributed collaborative control of manipulators with time delays. IEEE/CAA Journal of Automatica Sinica, 2022, 9(5): 854-863 doi: 10.1109/JAS.2022.105446
[16]	Ma Bo-Yu, Xie Zong-Wu, Zhan Bo-Wen, Jiang Zai-Nan, Liu Yang, Liu Hong. Actual shape-based obstacle avoidance synthesized by velocity-acceleration minimization for redundant manipulators: An optimization perspective. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2023: 1-15 doi: 10.1109/TSMC.2023.3283266, to be published
[17]	Zhang Zhi-Jun, He Hao-Tian, Deng Xian-Zhi. An FPGA-implemented antinoise fuzzy recurrent neural network for motion planning of redundant robot manipulators. IEEE Transactions on Neural Networks and Learning Systems, 2023: 1-13 doi: 10.1109/TNNLS.2023.3253801, to be published
[18]	Li Shuai, Shao Zi-Li, Guan Yong. A dynamic neural network approach for efficient control of manipulators. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2019, 49(5): 932-941 doi: 10.1109/TSMC.2017.2690460
[19]	Guo Kun-Lin, Su Hang, Yang Cheng-Guang. A small opening workspace control strategy for redundant manipulator based on RCM method. IEEE Transactions on Control Systems Technology, 2022, 30(6): 2717-2725 doi: 10.1109/TCST.2022.3145645
[20]	張振國, 毛建旭, 譚浩然, 王耀南, 張雪波, 江一鳴. 重大裝備制造多機器人任務(wù)分配與運動(dòng)規劃技術(shù)研究綜述. 自動(dòng)化學(xué)報, 2024, 50(1): 21?41 Zhang Zhen-Guo, Mao Jian-Xu, Tan Hao-Ran, Wang Yao-Nan, Zhang Xue-Bo, Jiang Yi-Ming. A review of task allocation and motion planning for multi-robot in major equipment manufacturing. Acta Automatica Sinica, 2024, 50(1): 21?41
[21]	Khail H. Nonlinear Systems 3rd Edition. Englewood Cliffs: Prentice-Hall, 2001.
[22]	Yan Jing-Kun, Jin Long, Yuan Zhan-Ting, Liu Zhi-Yi. RNN for receding horizon control of redundant robot manipulators. IEEE Transactions on Industrial Electronics, 2022, 69(2): 1608-1619 doi: 10.1109/TIE.2021.3062257
[23]	Zhang Zhi-Jun, Chen Si-Yuan, Zhu Xu-Peng, Yan Zi-Yi. Two hybrid end-effector posture-maintaining and obstacle-limits avoidance schemes for redundant robot manipulators. IEEE Transactions on Industrial Informatics, 2020, 16(2): 754-763 doi: 10.1109/TII.2019.2922694
[24]	Jin Long, Zhang Yu-Nong. G2-type SRMPC scheme for synchronous manipulation of two redundant robot arms. IEEE Transactions on Cybernetics, 2015, 45(2): 153-164 doi: 10.1109/TCYB.2014.2321390
[25]	Liu M, Shang M S. Orientation tracking incorporated multi-criteria control for redundant manipulators with dynamic neural network. IEEE Transactions on Industrial Electronics, 2014, 71(4): 3801?3810