基于体感的仿人机器人步态学习与控制

doi:10.11772/j.issn.1001-9081.2015.03.787

计算机应用 ›› 2015, Vol. 35 ›› Issue (3): 787-791.DOI: 10.11772/j.issn.1001-9081.2015.03.787

基于体感的仿人机器人步态学习与控制

周浩¹, 浦剑涛², 梁岚珍^1,2, 方建军², 郭浩¹

1. 新疆大学电气工程学院, 乌鲁木齐 830047;
2. 北京联合大学自动化学院, 北京 100101

收稿日期:2014-10-23 修回日期:2014-11-14 出版日期:2015-03-10 发布日期:2015-03-13
通讯作者: 梁岚珍
作者简介:周浩(1988-),男,辽宁沈阳人,硕士研究生,主要研究方向:计算机控制、机器人技术;浦剑涛(1977-),男,江苏溧阳人,讲师,博士,主要研究方向:机器人技术、人工智能;梁岚珍(1957-),女,山西岚县人,教授,硕士,主要研究方向:计算机控制、机器人技术;方建军(1970-),男,湖北罗田人,教授,博士,主要研究方向:特种机器人;郭浩(1989-),男,河北保定人,硕士研究生,主要研究方向:计算机控制
基金资助:
北京市自然科学基金资助项目(4142018)

Gait learning and control of humanoid robot based on Kinect

ZHOU Hao¹, PU Jiantao², LIANG Lanzhen^1,2, FANG Jianjun², GUO Hao¹

1. School of Electrical Engineering, Xinjiang University, Urumqi Xinjiang 830047, China;
2. College of Automation, Beijing Union University, Beijing 100101, China

Received:2014-10-23 Revised:2014-11-14 Online:2015-03-10 Published:2015-03-13

摘要/Abstract

摘要：

针对现有理想化步态动力学模型规划方法复杂、人为指定参数过多、计算量大的问题,提出一种基于体感数据学习人体步态的仿人机器人步态生成方法。首先,用体感设备收集人体骨骼信息,基于最小二乘拟合方法建立人体关节局部坐标系;其次,搭建人体与机器人映射的运动学模型,根据两者间主要关节映射关系,生成机器人关节转角轨迹,实现机器人对人类行走姿态的学习;然后,基于零力矩点(ZMP)稳定性原则,对机器人脚踝关节转角采用梯度下降算法进行优化控制;最后,在步态稳定性分析上,提出使用安全系数来评价机器人行走稳定程度的方法。实验结果表明,步行过程中安全系数保持在0~0.85,期望为0.4825,ZMP接近于稳定区域中心,机器人实现了仿人姿态的稳定行走,证明了该方法的有效性。

关键词: 仿人机器人, 步态规划, 步态学习, 零力矩点, 体感

Abstract:

To solve the problems of complex planning method, too many man-made specified parameters and huge computation in the existing gait dynamic model, the gait generation approach of humanoid robot based on the data collected by Kinect to learn human gait was proposed. Firstly, the skeleton information was collected by Kinect device, human joint local coordinate system was built by the least square fitting method. Next, the dynamic model of human body mapping robot was built, and robot joint angle trajectory was generated according to mapping relation between main joints, the studies of walking posture from human was realized. Then, Robot's ankle joint was optimized and controlled by gradient descent on the basis of Zero-Moment Point (ZMP) stability principle. Finally, on the gait stability analysis, safety factor was proposed to evaluate the stability of robot walk. The experimental results show that the safety factor of walking keeps in 0 to 0.85, experctation is 0.4825 and ZMP closes to stable regional centres, the robot realizes walking imitating human posture and gait stability, which proves the validity of the method.

Key words: humanoid robot, gait planning, gait learning, Zero-Moment Point (ZMP), Kinect

中图分类号:

TP242.6

周浩, 浦剑涛, 梁岚珍, 方建军, 郭浩. 基于体感的仿人机器人步态学习与控制[J]. 计算机应用, 2015, 35(3): 787-791.

ZHOU Hao, PU Jiantao, LIANG Lanzhen, FANG Jianjun, GUO Hao. Gait learning and control of humanoid robot based on Kinect[J]. Journal of Computer Applications, 2015, 35(3): 787-791.

参考文献

[1] WANG L, LIU Z, CHEN C L P,et al. Energy-efficient SVM learning control system for biped walking robots[J]. IEEE Transactions on Neural Networks and Learning Systems, 2013,24(5):831-837.
[2] TLALOLINI D, CHEVALLEREAU C, AOUSTIN Y. Human-like walking:optimal motion of a bipedal robot with toe-rotation motion[J]. IEEE/ASME Transactions on Mechatronics, 2011,16(2):310-320.
[3] PARK C S, HONG Y D, KIM J H. Evolutionary-optimized central pattern generator for stable modifiable bipedal walking[J]. IEEE/ASME Transactions on Mechatronics, 2014,19(4):1374-1383.
[4] TASKIRAN E, YILMAZ M, KOCA O,et al. Trajectory generation with natural ZMP references for the biped walking robot SURALP[C]//IEEE International Conference on Robotics and Automation. Piscataway:IEEE, 2010:4237-4242.
[5] TSUJI T, OHNISHI K. A control of biped robot which applies inverted pendulum mode with virtual supporting point[C]//IEEE International Workshop on Advanced Motion Control. Piscataway:IEEE, 2002:478-483.
[6] STASSE O, VERRELST B, VANDERBORGHT B,et al. Strategies for humanoid robots to dynamically walk over large obstacles[J]. IEEE Transactions on Robotics, 2009,25(4):960-967.
[7] YU G, LIU Y, LI X. Humanoid robot gait planning based on 3D linear inverted pendulum model[J]. Journal of Computer Applications, 2012,32(9):2643-2647.(于国晨,刘永信,李晓红.基于三维线性倒立摆的仿人机器人步态规划[J].计算机应用,2012,32(9):2643-2647.)
[8] LIM H-O, TAKANISHI A. Compensatory motion control for a biped walking robot[J]. Robotica, 2005,23(1):1-11.
[9] MOOSAVIAN S A A, ALGHOONEH M, TAKHMAR A. Stable trajectory planning, dynamics modeling and fuzzy regulated sliding mode control of a biped robot[C]//IEEE-RAS International Conference on Humanoid Robots. Piscataway:IEEE, 2007:471-476.
[10] ZHAO X, HUANG Q, PENG Z,et al. Kinematics mapping of humanoid motion based on human motion[J]. Robot, 2005,27(4):358-361.(赵晓军,黄强,彭朝琴,等.基于人体运动的仿人型机器人动作的运动学匹配[J].机器人,2005,27(4):358-361.)
[11] KAJITA S. Humanoid robots[M]. GUAN Y, translated. Beijing:Tsinghua University Press, 2007:17-90.(梶田秀司.仿人机器人[M].管贻生,译.北京:清华大学出版社,2007:17-90.)
[12] VUKOBRATOVIC M, BOROVAC B. Zero-moment point-thirty five years of its life[J]. International Journal of Humanoid Robotics, 2004,1(1):157-173.

基于体感的仿人机器人步态学习与控制

Gait learning and control of humanoid robot based on Kinect

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

编辑推荐

Metrics

[1]	杨文璐, 于孟孟, 谢宏. 基于关键姿势的双人交互行为识别[J]. 计算机应用, 2020, 40(8): 2231-2235.
[2]	易康, 赵玉婷, 齐新社. 基于Octomap的仿人机器人局部环境与能力图模型算法[J]. 计算机应用, 2019, 39(4): 1220-1223.
[3]	曾鹤杨宜民. 基于零力矩点预观控制的在线步态规划[J]. 计算机应用, 2014, 34(2): 514-518.
[4]	柯文德彭志平陈珂项顺伯. 基于分层Option的仿人机器人相似性关键姿势转换[J]. 计算机应用, 2013, 33(05): 1301-1304.
[5]	于国晨刘永信李晓红. 基于三维线性倒立摆的仿人机器人步态规划[J]. 计算机应用, 2012, 32(09): 2643-2647.
[6]	卫津津李素梅刘文娟臧艳军. 基于可控金字塔的立体图像质量评价方法[J]. 计算机应用, 2012, 32(03): 710-714.
[7]	王凡李龙澍. 实时解析法在NAO模型运动学求逆解中的应用[J]. 计算机应用, 2011, 31(10): 2825-2827.