全向过约束四足机器人的设计科学与运动控制

    相比于轮式移动机器人，模仿腿足动物的四足机器人具有丰富的步态和能适应各种非结构化环境，能在军事侦察、工业巡检和抢险救援中发挥重要作用。作为四足机器人非常重要的组成单元，机器人腿的设计直接影响着机器人的运动能力。机器人腿的设计需要在运动敏捷性和系统鲁棒性之间进行权衡。当前四足机器人多采用基于平面机构的三自由度腿，虽然能满足基本的运动需求，但是存在关节负载不均衡和腿部惯量大等问题。因此，我们利用机构设计在动力传输和运动生成方面的机械优势来提升机器人腿的运动性能。
    本文创新性地采用过约束Bennett 机构，设计一种负载均衡、低腿部惯量和高动态的机器人腿；针对闭环机构，介绍了一种等效生成树模型的简化建模方法；开展了过约束Bennett 机器人腿的正逆运动学、微分运动学和逆动力学建模；定量表征了机构的速度特性、承载特性和各向同性；介绍了一种通用且易于实施和扩展的四足机器人全向运动控制框架，包括行为模式与步态的调度、状态估计、单腿柔顺控制和模型预测控制等。
     为验证过约束四足机器人设计的可行性，搭建了八自由度和十二自由度过约束四足机器人样机。开展了八自由度四足机器人的前向行走、侧向行走和原地转向实验，验证了其全向运动潜力，这是基于平面机构设计的四足机器人难以做到的；另外，在PyBullet 仿真环境中实现了十二自由度四足机器人全向运动、多地形行走和侧向扰动恢复，初步验证了控制算法的有效性和鲁棒性；最后，开展了单腿样机轨迹跟踪实验，表征了其动态性能和本体感知力控性能。

    Compared with wheeled mobile robots, quadruped robots that mimic legged animals have rich gaits and can adapt to various unstructured environments. They are expected to play an important role in military reconnaissance, industrial inspection, and emergency rescue. As one of the most critical parts of the legged robots, the design of robotic legs directly affects their locomotion ability. The design of robotic legs requires a trade-off between agility and system robustness. At present, quadrupeds mostly utilize 3-DoFs legs based on planar mechanisms. Although it can meet basic motion requirements, there are problems such as uneven joint payload and high leg inertia. Thus, we try to improve the robotic leg performance through the mechanical advantages of mechanism design in terms of power transmission and motion generation.
    The thesis proposed a novel robotic leg design with a balanced joint payload, low inertia, and highly dynamic based on the overconstrained Bennett mechanism. To simplify the modeling of the closed-loop mechanisms, an equivalent spanning-tree model is presented. The modeling of the Bennett leg is introduced, including kinematics, differential kinematics, and inverse dynamics. Its velocity performance, payload performance, and isotropy are also analyzed quantitatively. A general and easily expandable control framework of quadruped for omni-directional locomotion is carried out, including scheduling of behavioral patterns and gaits, state estimation, leg compliance control, and model predictive control.
    To validate the feasibility of the quadruped design, 8-DoFs, and 12-DoFs overconstrained quadruped prototypes are developed. The potential of omni-directional locomotion with only 8-DoFs is verified through experiments of forward walking, lateral walking, and turning on the spot, which is challenging for an 8-DoFs quadruped with planar leg design. Additionally, omni-directional locomotion, multi-terrain locomotion, and push recovery with 12-DoFs are achieved in PyBullet simulation, which preliminarily verifies the effectiveness and robustness of the proposed controllers. Finally, the dynamic performance and proprioceptive force control performance are evaluated by leg experiments.

[1] RAIBERT M, BLANKESPOOR K, NELSON G, et al. Bigdog, the rough-terrain quadruped robot[J]. IFAC Proceedings Volumes, 2008, 41(2): 10822-10825.
[2] DI CARLO J, WENSING P M, KATZ B, et al. Dynamic locomotion in the mit cheetah 3 through convex model-predictive control[C]//2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2018: 1-9.
[3] HUTTER M, GEHRING C, JUD D, et al. Anymal-a highly mobile and dynamic quadrupedal robot[C]//2016 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, 2016: 38-44.
[4] LEE J, HWANGBO J, WELLHAUSEN L, et al. Learning quadrupedal locomotion over challenging terrain[J]. Science robotics, 2020, 5(47): eabc5986.
[5] The Air Force Just Tested ”Robot Dogs” For Use In Base Security[EB/OL]. 2023. https://www.thedrive.com/the-war-zone/36229/the-air-force-just-tested-robot-dogs-for-use-in-base-security.
[6] Robotic Industrial Inspection[EB/OL]. 2023. https://www.anybotics.com/robotic-industrial-inspection/.
[7] RÖNNAU A, HEPPNER G, NOWICKI M, et al. LAURON V: A versatile six-legged walking robot with advanced maneuverability[C]//2014 IEEE/ASME International Conference on Advanced Intelligent Mechatronics. IEEE, 2014: 82-87.
[8] XIAO A, TONG W, YANG L, et al. Robotic guide dog: Leading a human with leash-guided hybrid physical interaction[C]//2021 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2021: 11470-11476.
[9] ARM P, ZENKL R, BARTON P, et al. Spacebok: A dynamic legged robot for space exploration[C]//2019 international conference on robotics and automation (ICRA). IEEE, 2019: 6288-6294.
[10] HYUN D J, SEOK S, LEE J, et al. High speed trot-running: Implementation of a hierarchical controller using proprioceptive impedance control on the MIT Cheetah[J]. The International Journal of Robotics Research, 2014, 33(11): 1417-1445.
[11] KATZ B G. A low cost modular actuator for dynamic robots[D]. Massachusetts Institute of Technology, 2018.
[12] Legacy Robots[EB/OL]. 2023. https://www.bostondynamics.com/legacy.
[13] Spot[EB/OL]. 2023. https://www.bostondynamics.com/products/spot.
[14] HUTTER M. StarlETH & Co.: Design and control of legged robots with compliant actuation[D]. ETH Zurich, 2013.
[15] FANKHAUSER P, BELLICOSO C D, GEHRING C, et al. Free gait—an architecture for the versatile control of legged robots[C]//2016 IEEE-RAS 16th International Conference on Humanoid Robots (Humanoids). IEEE, 2016: 1052-1058.
[16] KOLVENBACH H, HAMPP E, BARTON P, et al. Towards jumping locomotion for quadruped robots on the moon[C]//2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2019: 5459-5466.
[17] SEOK S, WANG A, CHUAH M Y, et al. Design principles for energy-efficient legged locomotion and implementation on the MIT cheetah robot[J]. Ieee/asme transactions on mechatronics, 2014, 20(3): 1117-1129.
[18] ANANTHANARAYANAN A, AZADI M, KIM S. Towards a bio-inspired leg design for highspeed running[J]. Bioinspiration & biomimetics, 2012, 7(4): 046005.
[19] PARK H W, WENSING P M, KIM S. High-speed bounding with the MIT Cheetah 2: Control design and experiments[J]. The International Journal of Robotics Research, 2017, 36(2): 167-192.
[20] PARK H W, PARK S, KIM S. Variable-speed quadrupedal bounding using impulse planning: Untethered high-speed 3d running of mit cheetah 2[C]//2015 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2015: 5163-5170.
[21] BLEDT G, POWELL M J, KATZ B, et al. MIT Cheetah 3: Design and control of a robust, dynamic quadruped robot[C]//2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2018: 2245-2252.
[22] KIM D, DI CARLO J, KATZ B, et al. Highly dynamic quadruped locomotion via whole-body impulse control and model predictive control[A]. 2019.
[23] SEMINI C, BARASUOL V, BOAVENTURA T, et al. Towards versatile legged robots through active impedance control[J]. The International Journal of Robotics Research, 2015, 34(7): 1003-1020.
[24] WINKLER A, HAVOUTIS I, BAZEILLE S, et al. Path planning with force-based foothold adaptation and virtual model control for torque controlled quadruped robots[C]//2014 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2014: 6476-6482.
[25] SEMINI C, BARASUOL V, GOLDSMITH J, et al. Design of the hydraulically actuated, torquecontrolled quadruped robot HyQ2Max[J]. IEEE/Asme Transactions on Mechatronics, 2016, 22(2): 635-646.
[26] HyQReal[EB/OL]. 2023. https://www.iit.it/web/dynamic-legged-systems/hyqreal.
[27] 荣学文. SCalf 液压驱动四足机器人的机构设计与运动分析[D]. 济南: 山东大学, 2013.
[28] 柴汇, 孟健, 荣学文, 等. 高性能液压驱动四足机器人SCalf 的设计与实现[J]. 机器人, 2014, 36(4): 7.
[29] 杨琨. 液压驱动四足机器人能耗分析、优化及动力系统研究[D]. 山东大学, 2020.
[30] 蒋振宇. 基于SLIP 模型的四足机器人对角小跑步态控制研究[D]. 哈尔滨工业大学, 2014.
[31] 李满天, 蒋振宇, 郭伟, 等. 四足仿生机器人单腿系统[J]. 机器人, 2014, 36(1): 8.
[32] HU N, LI S, GAO F. Multi-objective hierarchical optimal control for quadruped rescue robot [J]. International Journal of Control, Automation and Systems, 2018, 16(4): 1866-1877.
[33] 邓黎明. 四足小象机器人实时控制系统的设计与研究[D]. 上海交通大学, 2014.
[34] 田兴华, 高峰, 陈先宝, 等. 四足仿生机器人混联腿构型设计及比较[J]. 机械工程学报,2013, 49(6): 8.
[35] Vision60[EB/OL]. 2023. https://www.ghostrobotics.io/latest.
[36] Jueying X20[EB/OL]. 2023. https://www.deeprobotics.cn/en/products_jy_303.html.
[37] Unitree B1[EB/OL]. 2023. https://www.unitree.com/cn/products/B1.
[38] SARANLI U, BUEHLER M, KODITSCHEK D E. RHex: A simple and highly mobile hexapod robot[J]. The International Journal of Robotics Research, 2001, 20(7): 616-631.
[39] POULAKAKIS I, PAPADOPOULOS E, BUEHLER M. On the stability of the passive dynamics of quadrupedal running with a bounding gait[J]. The International Journal of Robotics Research, 2006, 25(7): 669-687.
[40] BLACKMAN D J, NICHOLSON J V, ORDONEZ C, et al. Gait development on minitaur, a direct drive quadrupedal robot[C]//Unmanned Systems Technology XVIII: volume 9837. SPIE, 2016: 141-155.
[41] KALOUCHE S. Design for 3d agility and virtual compliance using proprioceptive force control in dynamic legged robots[D]. School of Computer Science., Carnegie Mellon Univ. Pittsburgh, PA, 2016.
[42] PRATT G A, WILLIAMSON M M. Series elastic actuators[C]//Proceedings 1995 IEEE/RSJ International Conference on Intelligent Robots and Systems. Human Robot Interaction and Cooperative Robots: volume 1. IEEE, 1995: 399-406.
[43] HUTTER M, GEHRING C, HÖPFLINGER M A, et al. Toward combining speed, efficiency, versatility, and robustness in an autonomous quadruped[J]. IEEE Transactions on Robotics, 2014, 30(6): 1427-1440.
[44] KENNEALLY G, DE A, KODITSCHEK D E. Design principles for a family of direct-drive legged robots[J]. IEEE Robotics and Automation Letters, 2016, 1(2): 900-907.
[45] KAU N, SCHULTZ A, FERRANTE N, et al. Stanford doggo: An open-source, quasi-directdrive quadruped[C]//2019 International conference on robotics and automation (ICRA). IEEE, 2019: 6309-6315.
[46] RAIBERT M H. Legged robots that balance[M]. MIT press, 1986.
[47] DE A, KODITSCHEK D E. Vertical hopper compositions for preflexive and feedback-stabilized quadrupedal bounding, pacing, pronking, and trotting[J]. The International Journal of Robotics Research, 2018, 37(7): 743-778.
[48] PRATT J, DILWORTH P, PRATT G. Virtual model control of a bipedal walking robot[C]//Proceedings of International Conference on Robotics and Automation: volume 1. IEEE, 1997:193-198.
[49] WANG Y, BOYD S. Fast model predictive control using online optimization[J]. IEEE Transactions on control systems technology, 2009, 18(2): 267-278.
[50] CHEN H, HONG Z, YANG S, et al. Quadruped Capturability and Push Recovery via a Switched-Systems Characterization of Dynamic Balance[A]. 2022.
[51] NAKAMURA Y, HANAFUSA H, YOSHIKAWA T. Task-priority based redundancy control of robot manipulators[J]. The International Journal of Robotics Research, 1987, 6(2): 3-15.
[52] MISTRY M, BUCHLI J, SCHAAL S. Inverse dynamics control of floating base systems using orthogonal decomposition[C]//2010 IEEE international conference on robotics and automation. IEEE, 2010: 3406-3412.
[53] NEUNERT M, STÄUBLE M, GIFTTHALER M, et al. Whole-body nonlinear model predictive control through contacts for quadrupeds[J]. IEEE Robotics and Automation Letters, 2018, 3(3): 1458-1465.
[54] MASTALLI C, HAVOUTIS I, FOCCHI M, et al. Motion planning for quadrupedal locomotion: Coupled planning, terrain mapping, and whole-body control[J]. IEEE Transactions on Robotics, 2020, 36(6): 1635-1648.
[55] FARSHIDIAN F, JELAVIĆ E, WINKLER A W, et al. Robust whole-body motion control of legged robots[C]//2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2017: 4589-4596.
[56] CAVAGNA G, HEGLUND N, TAYLOR C. Walking, running and galloping: mechanical similarities between different animals[M]. Academic Press, 1976.
[57] FOCCHI M, DEL PRETE A, HAVOUTIS I, et al. High-slope terrain locomotion for torquecontrolled quadruped robots[J]. Autonomous Robots, 2017, 41: 259-272.
[58] DING Y, PANDALA A, PARK H W. Real-time model predictive control for versatile dynamic motions in quadrupedal robots[C]//2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019: 8484-8490.
[59] Overconstrained mechanism[EB/OL]. 2023. https://en.wikipedia.org/wiki/Overconstrained_mechanism.
[60] BENNETT G T. A New Mechanism[J]. Engineering, 1903, 76: 777-778.
[61] Chebychev–Grübler–Kutzbach criterion[EB/OL]. 2023. https://en.wikipedia.org/wiki/Chebychev%E2%80%93Gr%C3%BCbler%E2%80%93Kutzbach_criterion.
[62] SONG C, CHEN Y, CHEN I M. A 6R linkage reconfigurable between the line-symmetric Bricard linkage and the Bennett linkage[J]. Mechanism and Machine Theory, 2013, 70: 278-292.
[63] LYNCH K M, PARK F C. Modern robotics[M]. Cambridge University Press, 2017.
[64] Quartic function[EB/OL]. 2023. https://en.wikipedia.org/wiki/Quartic_function.
[65] FEATHERSTONE R. Rigid body dynamics algorithms[M]. Springer, 2014.
[66] FELIS M L. RBDL: an efficient rigid-body dynamics library using recursive algorithms[J/OL]. Autonomous Robots, 2016: 1-17. http://dx.doi.org/10.1007/s10514-016-9574-0.
[67] CARPENTIER J, VALENZA F, MANSARD N, et al. Pinocchio: fast forward and inverse dynamics for poly-articulated systems[Z]. 2015–2021.
[68] 潘阳. P-P 结构六足机器人性能设计与控制实验研究[D]. 上海交通大学, 2014.
[69] OLDS K C. Global indices for kinematic and force transmission performance in parallel robots [J]. IEEE Transactions on Robotics, 2015, 31(2): 494-500.
[70] KITANO S, HIROSE S, HORIGOME A, et al. TITAN-XIII: sprawling-type quadruped robot with ability of fast and energy-efficient walking[J]. Robomech Journal, 2016, 3: 1-16.
[71] BLEDT G, WENSING P M, INGERSOLL S, et al. Contact model fusion for event-based locomotion in unstructured terrains[C]//2018 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2018: 4399-4406.
[72] CARIGNAN C R, CLEARY K R. Closed-loop force control for haptic simulation of virtual environments[M]. Haptics-e, The electronic journal of haptics research, 2000.
[73] BLOESCH M, HUTTER M, HOEPFLINGER M A, et al. State estimation for legged robotsconsistent fusion of leg kinematics and IMU[J]. Robotics, 2013, 17: 17-24.
[74] GEHRING C, BELLICOSO C D, COROS S, et al. Dynamic trotting on slopes for quadrupedal robots[C]//2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2015: 5129-5135.
[75] Moving average[EB/OL]. 2023. https://en.wikipedia.org/wiki/Moving_average.
[76] Kahan summation algorithm[EB/OL]. 2023. https://en.wikipedia.org/wiki/Kahan_summation_algorithm#Further_enhancements.
[77] PRATT J, CARFF J, DRAKUNOV S, et al. Capture point: A step toward humanoid push recovery[C]//2006 6th IEEE-RAS international conference on humanoid robots. IEEE, 2006: 200-207.
[78] Derivatives of a Bézier Curve[EB/OL]. 2023. https://pages.mtu.edu/~shene/COURSES/cs3621/NOTES/spline/Bezier/bezier-der.html.
[79] KALAKRISHNAN M, BUCHLI J, PASTOR P, et al. Fast, robust quadruped locomotion over challenging terrain[C]//2010 IEEE International Conference on Robotics and Automation. IEEE, 2010: 2665-2670.
[80] CHEN L, YE S, SUN C, et al. Optimized foothold planning and posture searching for energyefficient quadruped locomotion over challenging terrains[C]//2020 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2020: 399-405.
[81] NGUYEN Q, POWELL M J, KATZ B, et al. Optimized jumping on the mit cheetah 3 robot[C]//2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019: 7448-7454.
[82] STELLATO B, BANJAC G, GOULART P, et al. OSQP: an operator splitting solver for quadratic programs[J/OL]. Mathematical Programming Computation, 2020, 12(4): 637-672. https://doi.org/10.1007/s12532-020-00179-2.
[83] FERREAU H J, KIRCHES C, POTSCHKA A, et al. qpOASES: A parametric active-set algorithm for quadratic programming[J]. Mathematical Programming Computation, 2014, 6: 327-363.
[84] WALTER R M, CARRIER D R. Ground forces applied by galloping dogs[J]. Journal of Experimental Biology, 2007, 210(2): 208-216.
[85] ROHMER E, SINGH S P, FREESE M. V-REP: A versatile and scalable robot simulation framework[C]//2013 IEEE/RSJ international conference on intelligent robots and systems. IEEE, 2013: 1321-1326.
[86] GARRIDO-JURADO S, MUÑOZ-SALINAS R, MADRID-CUEVAS F J, et al. Automatic generation and detection of highly reliable fiducial markers under occlusion[J]. Pattern Recognition, 2014, 47(6): 2280-2292.
[87] COUMANS E, BAI Y. PyBullet, a Python module for physics simulation for games, robotics and machine learning[EB/OL]. 2016–2021. http://pybullet.org.
[88] BHATIA A, JOHNSON A M, MASON M T. Direct drive hands: Force-motion transparency in gripper design[C]//Robotics: science and systems. 2019.