考虑关节角加速度约束的仿人机器人偏摆力矩控制方法
|
摘要
针对仿人机器人步行过程中存在的机器人关节角加速度约束影响控制性能的问题,提出一种考虑关节角加速度约束的仿人机器人偏摆力矩控制方法.该方法充分考虑了双臂在摆动过程中对偏摆力矩的影响,根据力矩平衡条件得到需要抵消的偏摆力矩的大小与方向,将偏摆力矩的控制问题转化为带约束条件的二次规划问题,并设计了一种在线变步长迭代算法计算得到优化后的双臂摆动轨迹.实验表明,该方法能有效抵消机器人步行中产生的偏摆力矩,避免控制过程中的"削峰"现象,有效提高机器人的步行稳定性.
To cancel the effect on control performance due to joint acceleration constraints during humanoid robot walks,a new yaw moment control approach with consideration of joint acceleration constraints is proposed. After analyzing the effect on stability caused by the motion of arms swing, the undesired yaw moment which needs to be counteracted is computed. Then, the yaw moment control problem is formulated as a quadratic programming problem with inequality constraints.To solve this optimization problem effectively, an online variable step-size iteration method is designed to generate the trajectories of arms swinging. The experiment and simulation results verified that the proposed approach can effectively compensate yaw moment, avoid peak cut-off phenomenon and guarantee the stable walking.
To cancel the effect on control performance due to joint acceleration constraints during humanoid robot walks,a new yaw moment control approach with consideration of joint acceleration constraints is proposed. After analyzing the effect on stability caused by the motion of arms swing, the undesired yaw moment which needs to be counteracted is computed. Then, the yaw moment control problem is formulated as a quadratic programming problem with inequality constraints.To solve this optimization problem effectively, an online variable step-size iteration method is designed to generate the trajectories of arms swinging. The experiment and simulation results verified that the proposed approach can effectively compensate yaw moment, avoid peak cut-off phenomenon and guarantee the stable walking.
引文
[1]曹政才,赵应涛,付宜利.车式移动机器人轨迹跟踪控制方法[J].电子学报,2012,40(4):632-635.CAO Zheng-cai,ZHAO Ying-tao,FU Yi-li. Trajectorytracking control approach of a car-like mobile robot[J].Acta Electronica Sinica,2012,40(4):632-635.(in Chi-nese)
[2]王保防,张瑞雷,郭健,等.纵向打滑状态下的轮式移动机器人编队控制[J].电子学报,2017,45(1):206-212.WANG Bao-fang,ZHANG Rui-lei,GUO Jian,et al. Forma-tion control for car-like mobile robots under slip conditions[J]. Acta Electronica Sinica,2017,45(1):206-212.(inChinese)
[3]YANG L,LIU Z,ZHANG Y. Online walking control sys-tem for biped robot w ith optimized learning mechanism:anexperimental study[J]. Nonlinear Dynamics,2016,86(3):2035-2047.
[4]PIAO S,ZHONG Q,LIU Y. The research of optimal mo-tion planning for robot in complex environment[J]. Chi-nese Journal of Electronics,2011,20(4):637-640.
[5]钟秋波,朴松昊,杨滨.基于参数优化的类机器人运动规划研究[J].电子学报,2012,40(6):1154-1158.ZHONG Qiu-bo,PIAO Song-hao,YANG Bin. Research ofmotion control for humanoid robot based on parametric op-timum[J]. Acta Electronica Sinica,2012,40(6):1154-1158.(in Chinese)
[6]LEE B J,STONIER D,KIM Y D,et al. Modifiable walkingpattern of a humanoid robot by using allow able ZM P varia-tion[J]. IEEE Transactions on Robotics,2008,24(4):917-925.
[7]李建,陈卫东,王丽军,等.未知不平整地面上的双足步行稳定控制[J].电子学报,2010,38(11):2669-2674.LI Jian,CHEN Wei-dong,WANG Li-jun,et al. Stabilitycontrol for biped w alking on unknow n rough surface[J].Acta Electronica Sinica,2010,38(11):2669-2674.(inChinese)
[8] ZHU C,Tomizawa Y,LUO X,et al. Biped walking withvariable ZM P,frictional constraint,and inverted pendulummodel[A]. Proceeding of the 2014 IEEE InternationalConference on Robotics and Biomimetics[C]. China:IEEE,2004. 425-430.
[9]ERBATUR K,KURT O. Natural ZMP trajectories for bipedrobot reference generation[J]. IEEE Transactions on Indus-trial Electronics,2009,56(3):835-845.
[10]LIU Z,WANG L,CHEN C,et al. Energy-efficiency-basedgait control system architecture and algorithm for biped ro-bots[J]. IEEE Transactions on Systems,M an,and Cyber-netics(Part C:Applications and Review s),2012,42(6):926-933.
[11]蔚伟,包钢,王祖温,等.基于腰关节力矩补偿的仿人机器人快速步行模式生成[J].机器人,2010,32(2):219-225.YU W,BAO G,WANG Z-W,et al. Fast w alking patterngeneration for humanoid robot using w aist joint momentcompensation[J]. Robot,2010,32(2):219-225.(inChinese)
[12]UGURLU B,SAGLIA J A,TSAGARAKIS N G,et al. Yawmoment compensation for bipedal robots via intrinsic angu-lar momentum constraint[J]. International Journal of Hu-manoid Robotics,2012,9(04):1250033. 1-1250033. 27.
[13]XING D P,SU J B. Arm/trunk motion generation for hu-manoid robot[J]. Science in China(Series F:InformationSciences),2010,53(8):1603-1612.
[14]付根平,陈建平,杨宜民.基于双臂摆动的仿人机器人偏摆力矩矫正方法[J].机器人,2012,34(4):498-504.FU G-P,CHEN J-P,YANG Y-M. A yaw moment coun-teracting method for humanoid robot based on arms sw ing-ing[J]. Robot,2012,34(4):498-504.(in Chinese)
[15]HE B. A new method for a class of linear variational ine-qualities[J]. M athematical Programming,1994,66(1-3):137-144.
[16] ZHANG Y,GE S S,Lee T H. A unified quadratic-pro-gramming-based dynamical system approach to jointtorque optimization of physically constrained redundantmanipulators[J]. IEEE Transactions on Systems,M an,and Cybernetics(Part B:Cybernetics),2004,34(5):2126-2132.
[17] ZHANG Y,WU H,ZHANG Z,et al. Acceleration-levelrepetitive motion planning of redundant planar robotssolved by a simplified LVI-based primal-dual neural net-w ork[J]. Robotics and Computer-Integrated M anufactur-ing,2013,29(2):328-343.
[18]付根平,杨宜民,陈建平,等.基于ZMP误差校正的仿人机器人步行控制[J].机器人,2013,35(1):39-44.FU G-P,YANG Y-M,CHEN J-P,et al. Walking controlfor humanoid robot based on ZM P error correction[J].Robot,2013,35(1):39-44.(in Chinese)
[19]杨亮,傅瑜,何怀文.基于腿部关节控制的仿人机器人偏摆力矩控制方法[J].控制与决策,2016,31(1):79-83.YANG L,FU Y,HE H-W. A yaw moment control methodfor humanoid robot based on leg joints control[J]. Controland Decision,2016,31(1):78-83.(in Chinese)
[20]CHEN C,WEN C,LIU Z,et al. Adaptive asymptotic con-trol of multivariable systems based on a one-parameter es-timation approach[J]. Automatica,2017,83:124-132.
[21]YANG C,JIANG Y,HE W,et al. Adaptive parameter es-timation and control design for robot manipulators w ith fi-nite-time convergence[J]. IEEE Transactions on IndustrialElectronics,2018,65(10):8112-8123.
[22]YANG L,LIU Z,ZHANG Y. Online walking control sys-tem for biped robot w ith optimized learning mechanism:an experimental study[J]. Nonlinear Dynamics,2016,86(3):2035-2047.
[23]KANEHIRA N,KAWASAKI T U,OHTA S,et al. Designand experiments of advanced leg module(HRP-2L)for hu-manoid robot(HRP-2)development[A]. Proceeding ofthe 2002 IEEE International Conference on Intelligent Ro-bots and Systems,Sw itzerland[C]. US:IEEE,2002. 2455-2460.
[24] HU K,OTT C,LEE D. Learning and generalization ofcompensative zero-moment point trajectory for bipedw alking[J]. IEEE Transactions on Robotics,2016,32(3):717-725.
[2]王保防,张瑞雷,郭健,等.纵向打滑状态下的轮式移动机器人编队控制[J].电子学报,2017,45(1):206-212.WANG Bao-fang,ZHANG Rui-lei,GUO Jian,et al. Forma-tion control for car-like mobile robots under slip conditions[J]. Acta Electronica Sinica,2017,45(1):206-212.(inChinese)
[3]YANG L,LIU Z,ZHANG Y. Online walking control sys-tem for biped robot w ith optimized learning mechanism:anexperimental study[J]. Nonlinear Dynamics,2016,86(3):2035-2047.
[4]PIAO S,ZHONG Q,LIU Y. The research of optimal mo-tion planning for robot in complex environment[J]. Chi-nese Journal of Electronics,2011,20(4):637-640.
[5]钟秋波,朴松昊,杨滨.基于参数优化的类机器人运动规划研究[J].电子学报,2012,40(6):1154-1158.ZHONG Qiu-bo,PIAO Song-hao,YANG Bin. Research ofmotion control for humanoid robot based on parametric op-timum[J]. Acta Electronica Sinica,2012,40(6):1154-1158.(in Chinese)
[6]LEE B J,STONIER D,KIM Y D,et al. Modifiable walkingpattern of a humanoid robot by using allow able ZM P varia-tion[J]. IEEE Transactions on Robotics,2008,24(4):917-925.
[7]李建,陈卫东,王丽军,等.未知不平整地面上的双足步行稳定控制[J].电子学报,2010,38(11):2669-2674.LI Jian,CHEN Wei-dong,WANG Li-jun,et al. Stabilitycontrol for biped w alking on unknow n rough surface[J].Acta Electronica Sinica,2010,38(11):2669-2674.(inChinese)
[8] ZHU C,Tomizawa Y,LUO X,et al. Biped walking withvariable ZM P,frictional constraint,and inverted pendulummodel[A]. Proceeding of the 2014 IEEE InternationalConference on Robotics and Biomimetics[C]. China:IEEE,2004. 425-430.
[9]ERBATUR K,KURT O. Natural ZMP trajectories for bipedrobot reference generation[J]. IEEE Transactions on Indus-trial Electronics,2009,56(3):835-845.
[10]LIU Z,WANG L,CHEN C,et al. Energy-efficiency-basedgait control system architecture and algorithm for biped ro-bots[J]. IEEE Transactions on Systems,M an,and Cyber-netics(Part C:Applications and Review s),2012,42(6):926-933.
[11]蔚伟,包钢,王祖温,等.基于腰关节力矩补偿的仿人机器人快速步行模式生成[J].机器人,2010,32(2):219-225.YU W,BAO G,WANG Z-W,et al. Fast w alking patterngeneration for humanoid robot using w aist joint momentcompensation[J]. Robot,2010,32(2):219-225.(inChinese)
[12]UGURLU B,SAGLIA J A,TSAGARAKIS N G,et al. Yawmoment compensation for bipedal robots via intrinsic angu-lar momentum constraint[J]. International Journal of Hu-manoid Robotics,2012,9(04):1250033. 1-1250033. 27.
[13]XING D P,SU J B. Arm/trunk motion generation for hu-manoid robot[J]. Science in China(Series F:InformationSciences),2010,53(8):1603-1612.
[14]付根平,陈建平,杨宜民.基于双臂摆动的仿人机器人偏摆力矩矫正方法[J].机器人,2012,34(4):498-504.FU G-P,CHEN J-P,YANG Y-M. A yaw moment coun-teracting method for humanoid robot based on arms sw ing-ing[J]. Robot,2012,34(4):498-504.(in Chinese)
[15]HE B. A new method for a class of linear variational ine-qualities[J]. M athematical Programming,1994,66(1-3):137-144.
[16] ZHANG Y,GE S S,Lee T H. A unified quadratic-pro-gramming-based dynamical system approach to jointtorque optimization of physically constrained redundantmanipulators[J]. IEEE Transactions on Systems,M an,and Cybernetics(Part B:Cybernetics),2004,34(5):2126-2132.
[17] ZHANG Y,WU H,ZHANG Z,et al. Acceleration-levelrepetitive motion planning of redundant planar robotssolved by a simplified LVI-based primal-dual neural net-w ork[J]. Robotics and Computer-Integrated M anufactur-ing,2013,29(2):328-343.
[18]付根平,杨宜民,陈建平,等.基于ZMP误差校正的仿人机器人步行控制[J].机器人,2013,35(1):39-44.FU G-P,YANG Y-M,CHEN J-P,et al. Walking controlfor humanoid robot based on ZM P error correction[J].Robot,2013,35(1):39-44.(in Chinese)
[19]杨亮,傅瑜,何怀文.基于腿部关节控制的仿人机器人偏摆力矩控制方法[J].控制与决策,2016,31(1):79-83.YANG L,FU Y,HE H-W. A yaw moment control methodfor humanoid robot based on leg joints control[J]. Controland Decision,2016,31(1):78-83.(in Chinese)
[20]CHEN C,WEN C,LIU Z,et al. Adaptive asymptotic con-trol of multivariable systems based on a one-parameter es-timation approach[J]. Automatica,2017,83:124-132.
[21]YANG C,JIANG Y,HE W,et al. Adaptive parameter es-timation and control design for robot manipulators w ith fi-nite-time convergence[J]. IEEE Transactions on IndustrialElectronics,2018,65(10):8112-8123.
[22]YANG L,LIU Z,ZHANG Y. Online walking control sys-tem for biped robot w ith optimized learning mechanism:an experimental study[J]. Nonlinear Dynamics,2016,86(3):2035-2047.
[23]KANEHIRA N,KAWASAKI T U,OHTA S,et al. Designand experiments of advanced leg module(HRP-2L)for hu-manoid robot(HRP-2)development[A]. Proceeding ofthe 2002 IEEE International Conference on Intelligent Ro-bots and Systems,Sw itzerland[C]. US:IEEE,2002. 2455-2460.
[24] HU K,OTT C,LEE D. Learning and generalization ofcompensative zero-moment point trajectory for bipedw alking[J]. IEEE Transactions on Robotics,2016,32(3):717-725.