多臂空间机器人操作大型目标的全身接触柔顺控制研究
|
摘要
多臂空间机器人操作大型目标时,由于受到来自目标把手的较大拖拽力,机器人无法对目标实施有效的操作和运动约束,这些无法预知的外部拖拽可能导致机器人支撑臂的损坏或者任务失败。为了应对上述挑战,提出了一种基于全身阻抗控制和独立导纳控制的组合式接触柔顺全身控制策略,分别在空间机器人多刚体系统质心层面构建机械阻抗特性和在每条机械臂建立导纳特性,实现对外部拖拽的有效管理。基于二次规划(QP)方法构建了系统控制器,将机器人全身运动行为统一起来,设计了包括全身柔顺控制和关节力矩优化在内的QP目标函数,用来实现接触柔顺控制。通过仿真和降维度气浮实验方法综合验证了上述方法的有效性。
A spatial multi-arm robot almost cannot perform effective manipulation constraint on the target due to the large drag force and motion from the multi-contact handles of target when it manipulates a large target on orbit. The unpredictable external drag may result in the damage to the arms or the failure of capture task. A contact compliance control strategy is proposed based on a combination control law of wholebody impedance control at the center of mass( CoM) level and the independent admittance control at every supporting arm level. The robot can effectively manage the external drag force by constructing the mechanical impedance at the CoM of multi-rigid body system of spatial robot and the admittance at every arm,respectively. A quadratic programming( QP) method is used to build a system controller for the contact compliance control,which can unify the robot's whole-body motion behavior. The effectiveness of the proposed method is verified by simulation and the dimensionality reduction experiment on air-bearing bed.
A spatial multi-arm robot almost cannot perform effective manipulation constraint on the target due to the large drag force and motion from the multi-contact handles of target when it manipulates a large target on orbit. The unpredictable external drag may result in the damage to the arms or the failure of capture task. A contact compliance control strategy is proposed based on a combination control law of wholebody impedance control at the center of mass( CoM) level and the independent admittance control at every supporting arm level. The robot can effectively manage the external drag force by constructing the mechanical impedance at the CoM of multi-rigid body system of spatial robot and the admittance at every arm,respectively. A quadratic programming( QP) method is used to build a system controller for the contact compliance control,which can unify the robot's whole-body motion behavior. The effectiveness of the proposed method is verified by simulation and the dimensionality reduction experiment on air-bearing bed.
引文
[1] YOSHIDA K,HASHIZUME K,ABIKO S. Zero reaction maneu-ver:flight validation with ETS-VII space robot and extension to ki-nematically redundant arm[C]∥Proceedings of IEEE Internation-al Conference on Robotics and Automation. Seoul,South Korea:IEEE,2001:441-446.
[2] FRIEND R B. Orbital express program summary and mission over-view[J]. Proceedings of the SPIE,2008,6958:695803-1-695803-11.
[3] BERND S. Automation and robotics in the German space program-unmanned on-orbit servicing(OOS)&the TECSAS mission[J].Cerebrovascular Diseases,2003,16(3):272-279.
[4] REINTSEMA D,THAETER J,RATHKE A,et al. DEOS-theGerman robotics approach to secure and de-orbit malfunctionedsatellites from low earth orbits[C]∥Proceedings of the 10th Inter-national Symposium on Artificial Intelligence,Robotics and Auto-mation in Space. Sapporo,Japan:Springer-Verlag,2010,
[5] SMITH P H,TAMPPARI L,ARVIDSON R E,et al. Introductionto special section on the Phoenix Mission:landing site character-ization experiments,mission overviews,and expected science[J].Journal of Geophysical Research,2008,113(3):5146-5163.
[6] OKI T,ABIKO S,NAKANISHI H,et al. Time-optimal detum-bling maneuver along an arbitrary arm motion during the capture ofa target satellite[C]∥Proceedings of International Conference onIntelligent Robots and Systems. Hamburg, Germany:IEEE,2015:625-630.
[7] UMETANI Y,YOSHIDA K. Resolved motion rate control of spacemanipulators with generalized Jacobian matrix[J]. IEEE Transac-tions on Robotics&Automation,1989,5(3):303-314.
[8] CAVENAGO F,VOLI L,MASSARI M. Adaptive hybrid systemframework for unified impedance and admittance control[J]. Jour-nal of Intelligent&Robotic Systems,2017,78(3/4):359-375.
[9] AGHILI F. A prediction and motion-planning scheme for visuallyguided robotic capturing of free-floating tumbling objects with un-certain dynamics[J]. IEEE Transactions on Robotics,2012,28(3):634-649.
[10] HOGAN N. Impedance control-an approach to manipulation. I-theory. II-implementation. III-applications[J]. Journal of Dyna-mic Systems Measurement and Control,1985,107(4):481-489.
[11] RASTEGARI R,MOOSAVIAN S A A. Multiple impedance con-trol of space free-flying robots via virtual linkages[J]. Acta As-tronautica,2010,66(5):748-759.
[12] STOLFI A,GASBARRI P,SABATINI M. A combined impe-dance-PD approach for controlling a dual-arm space manipulatorin the capture of a non-cooperative target[J]. Acta Astronautica,2017,139:243-253.
[13] BARATA J,HUSSEIN M. The Moore-Penrose pseudoinverse:atutorial review of the theory[J]. Brazilian Journal of Physics,2012,42(1/2):146-165.
[14] KUINDERSMA S,PERMENTER F,TEDRAKE R. An efficient-ly solvable quadratic program for stabilizing dynamic locomotion[C]∥Proceedings of IEEE International Conference on Roboticsand Automation. Hong Kong,China:IEEE,2014:2589-2594.
[15] FENG S Y,WHITMAN E C,XINJILEFU X,et al. Optimiza-tion-based full body control for the DARPA robotics challenge[J]. Journal of Field Robotics,2015,32(2):293-312.
[16] DE LASA M,MORDATCH I,HERTZMANN A. Feature-basedlocomotion controllers[J]. ACM Transactions on Graphics,2010,29(4):131:1-131:10.
[17]李世敬,王解法,冯祖仁,等.基于动态观测器的不确定机器人鲁棒控制研究[J].兵工学报,2005,26(2):263-266.LI S J,WANG J F,FENG Z R,et al. A study on the uncertainrobust control of robot based on dynamic observer[J]. Acta Ar-mamentarii,2005,26(2):263-266.(in Chinese)
[18] ESCANDE A,MANSARD N,WIEBER P B. Hierarchical quad-ratic programming:fast online humanoid-robot motion generation[J]. International Journal of Robotics Research,2014,33(7):1006-1028.
[19]李晖,伍清河.空间关联系统的最优控制[J].兵工学报,2010,31(6):685-691.LI H,WU Q H. Optimal control of spatial interconnected systems[J]. Acta Armamentarii,2010. 31(6):685-691.(in Chinese)
[20] HUANG Q,NAKAMURA Y. Sensory reflex control for humanoidwalking[J]. IEEE Transactions on Robotics,2005,21(5):977-984.
[21] GRANT M,BOYD S. CVX:Matlab software for disciplined con-vex programming,version 1. 21[M]. Palo Alto,CA,US:Stan-ford University,2008:155-210.
[2] FRIEND R B. Orbital express program summary and mission over-view[J]. Proceedings of the SPIE,2008,6958:695803-1-695803-11.
[3] BERND S. Automation and robotics in the German space program-unmanned on-orbit servicing(OOS)&the TECSAS mission[J].Cerebrovascular Diseases,2003,16(3):272-279.
[4] REINTSEMA D,THAETER J,RATHKE A,et al. DEOS-theGerman robotics approach to secure and de-orbit malfunctionedsatellites from low earth orbits[C]∥Proceedings of the 10th Inter-national Symposium on Artificial Intelligence,Robotics and Auto-mation in Space. Sapporo,Japan:Springer-Verlag,2010,
[5] SMITH P H,TAMPPARI L,ARVIDSON R E,et al. Introductionto special section on the Phoenix Mission:landing site character-ization experiments,mission overviews,and expected science[J].Journal of Geophysical Research,2008,113(3):5146-5163.
[6] OKI T,ABIKO S,NAKANISHI H,et al. Time-optimal detum-bling maneuver along an arbitrary arm motion during the capture ofa target satellite[C]∥Proceedings of International Conference onIntelligent Robots and Systems. Hamburg, Germany:IEEE,2015:625-630.
[7] UMETANI Y,YOSHIDA K. Resolved motion rate control of spacemanipulators with generalized Jacobian matrix[J]. IEEE Transac-tions on Robotics&Automation,1989,5(3):303-314.
[8] CAVENAGO F,VOLI L,MASSARI M. Adaptive hybrid systemframework for unified impedance and admittance control[J]. Jour-nal of Intelligent&Robotic Systems,2017,78(3/4):359-375.
[9] AGHILI F. A prediction and motion-planning scheme for visuallyguided robotic capturing of free-floating tumbling objects with un-certain dynamics[J]. IEEE Transactions on Robotics,2012,28(3):634-649.
[10] HOGAN N. Impedance control-an approach to manipulation. I-theory. II-implementation. III-applications[J]. Journal of Dyna-mic Systems Measurement and Control,1985,107(4):481-489.
[11] RASTEGARI R,MOOSAVIAN S A A. Multiple impedance con-trol of space free-flying robots via virtual linkages[J]. Acta As-tronautica,2010,66(5):748-759.
[12] STOLFI A,GASBARRI P,SABATINI M. A combined impe-dance-PD approach for controlling a dual-arm space manipulatorin the capture of a non-cooperative target[J]. Acta Astronautica,2017,139:243-253.
[13] BARATA J,HUSSEIN M. The Moore-Penrose pseudoinverse:atutorial review of the theory[J]. Brazilian Journal of Physics,2012,42(1/2):146-165.
[14] KUINDERSMA S,PERMENTER F,TEDRAKE R. An efficient-ly solvable quadratic program for stabilizing dynamic locomotion[C]∥Proceedings of IEEE International Conference on Roboticsand Automation. Hong Kong,China:IEEE,2014:2589-2594.
[15] FENG S Y,WHITMAN E C,XINJILEFU X,et al. Optimiza-tion-based full body control for the DARPA robotics challenge[J]. Journal of Field Robotics,2015,32(2):293-312.
[16] DE LASA M,MORDATCH I,HERTZMANN A. Feature-basedlocomotion controllers[J]. ACM Transactions on Graphics,2010,29(4):131:1-131:10.
[17]李世敬,王解法,冯祖仁,等.基于动态观测器的不确定机器人鲁棒控制研究[J].兵工学报,2005,26(2):263-266.LI S J,WANG J F,FENG Z R,et al. A study on the uncertainrobust control of robot based on dynamic observer[J]. Acta Ar-mamentarii,2005,26(2):263-266.(in Chinese)
[18] ESCANDE A,MANSARD N,WIEBER P B. Hierarchical quad-ratic programming:fast online humanoid-robot motion generation[J]. International Journal of Robotics Research,2014,33(7):1006-1028.
[19]李晖,伍清河.空间关联系统的最优控制[J].兵工学报,2010,31(6):685-691.LI H,WU Q H. Optimal control of spatial interconnected systems[J]. Acta Armamentarii,2010. 31(6):685-691.(in Chinese)
[20] HUANG Q,NAKAMURA Y. Sensory reflex control for humanoidwalking[J]. IEEE Transactions on Robotics,2005,21(5):977-984.
[21] GRANT M,BOYD S. CVX:Matlab software for disciplined con-vex programming,version 1. 21[M]. Palo Alto,CA,US:Stan-ford University,2008:155-210.