高动态响应的四足机器人全动力学建模与规划控制

足式机器人由于其天然的环境适应性而被期望广泛应用于各种非结构化场景。其中四足机器人既具备一定的承载能力，又能够高速灵活的运行，不管是学术研究还是工业应用都将其视作未来的发展方向和研究热点。

要想让四足机器人能够在不同非结构化场景下稳定健壮的行走，主要需要做到两点：一是提高机器人运动学和动力学模型的计算精度和速度，提升控制算法的鲁棒性，让机器人拥有一个反应迅速且聪明的“大脑”；二是攻克机器人实时驱动控制系统硬件技术，设计结构紧凑、反应灵敏的机器人样机，让机器人有一个健壮的“身体”。本文主要从这两个方面进行四足机器人样机设计及其运动控制算法研究。

目前四足机器人的模型主要采用SLIP模型或者质点模型，不能够精确地表达机器人实际的运行状态。本文将四足机器人看作一个由13个刚体和12个关节组成的系统。基于旋量理论建立了四足机器人全身运动学和动力学模型，将机器人在不同步态和不同行走阶段下的状态看作不同的拓扑机构，分别建立动力学模型，找到机器人关节和连杆之间的约束力矩阵关系，进一步求解机器人动力学问题。并基于上述模型提出了四足机器人步态规划和轨迹规划算法。

为验证算法和模型的正确性和可行性，先后迭代了两代四足机器人样机。其中控制器采用Ubuntu16.04操作系统并添加了Xenomai实时内核使之成为实时操作系统。控制器与驱动器之前使用Ethercat工业以太网总线协议，可实现1kHz的实时控制。机器人电机采用外转子直流无刷电机，配合1:10减速比的行星减速机，可以达到36 N·M额定扭矩和108N·M峰值扭矩。

最终在Adams仿真环境下验证了四足机器人walk步态和trot步态下的行走算法，使用实物样机测试了机器人控制的实时性，准确性，同时也验证了运动学模型的正确性和可行性，能够控制实物样机完成身体转动和基本的行走。

Leg robot are expected to be widely used in various unstructured scenarios due to their natural adaptability to the environment. Among them, quadruped robot not only has certain carrying capacity, but also can run flexibly at high speed. Both academic research and industrial application regard quadruped robot as the future development direction and research hotspot.

In order to enable quadruped robots to walk stably and robustly in different unstructured scenes, there are two main requirements: one is to improve the computational accuracy and speed of robot kinematics and dynamics model, improve the robustness of control algorithm, and make the robot have a quick and intelligent brain; Second, the robot real-time drive control system hardware technology to design a compact and responsive robot prototype, so that the robot has a robust body. In this paper, the prototype design and motion control algorithm of quadruped robot are studied from these two aspects.

At present, the models of quadruped robots mainly use SLIP model or particle model, which cannot accurately express the actual running state of the robot. In this paper, a quadruped robot is regarded as a system consisting of 13 rigid bodies and 12 joints. Based on the screw theory to establish the kinematics and dynamics models of quadruped robot body, will be a robot in different phases of gait and walking under state as different topology structure, respectively, to establish dynamics model to establish the binding force of matrix relationship between robot joints and connecting rod, solve the robot dynamics problems further. Based on the model, gait planning and trajectory planning algorithms for quadruped robots are proposed.

In order to verify the correctness and feasibility of the algorithm and model, two generations of prototype quadruped robots were built successively. The controller adopted Ubuntu16.04 operating system and added Xenomai real-time kernel to make it a real-time operating system. Controllers and drivers previously used the Ethercat industrial Ethernet bus protocol to achieve 1kHz real-time control. The robot motor adopts brushless DC motor with external rotor, which can reach 36 N·M rated torque and 108N·M peak torque combined with planetary reducer with 1:10 reduction ratio.

Finally, the walking algorithm of quadruped robot under walk gait and trot gait is verified in ADAMS simulation environment. The real time and accuracy of robot control are tested by physical prototype, and the correctness and feasibility of kinematics model are also verified. It can control physical prototype to complete body rotation and basic walking.

[1] 周济. 智能制造——“中国制造2025”的主攻方向 [J]. 中国机械工程, 2015, 26(17): 2273-84.
[2] 高峰, 郭为忠. 中国机器人的发展战略思考 [J]. 机械工程学报, 2016, 52(07): 1-5.
[3] 武雨飞. 探究智能移动机器人的现状及展望 [J]. 中国战略新兴产业, 2018, (12): 148.
[4] 徐国保, 尹怡欣, 周美娟. 智能移动机器人技术现状及展望 [J]. 机器人技术与应用, 2007, (02): 29-34.
[5] ZHONG Y, WANG R, FENG H, et al. Analysis and research of quadruped robot’s legs: A comprehensive review [J]. International Journal of Advanced Robotic Systems, 2019, 16(3).
[6] MOSHER R. Test and evaluation of a versatile walking truck; proceedings of the Proceedings of Off-Road Mobility Research Symposium, Washington DC, 1968, F, 1968 [C].
[7] HIROSE S. Some considerations on a feasible walking mechanism as a terrain vehicle; proceedings of the 3rd CISM-IFToMM Int Symp on Theory and Practice of Robots and Manipulators, F, 1978 [C].
[8] HIROSE S. Adaptive gait control of a quadruped walking vehicle; proceedings of the Robotics Research: 1st Int Symp on Robotics Research (ISRR1), F, 1983 [C].
[9] ARIKAWA K, HIROSE S. Development of quadruped walking robot TITAN-VIII; proceedings of the Proc of International Conference on Intelligent Robots & Systems, F, 1996 [C].
[10] HODOSHIMA R, DOI T, FUKUDA Y, et al. Development of TITAN XI: a quadruped walking robot to work on slopes; proceedings of the 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)(IEEE Cat No 04CH37566), F, 2004 [C]. IEEE.
[11] KATO K, HIROSE S. Development of the quadruped walking robot, TITAN-IX — mechanical design concept and application for the humanitarian de-mining robot [J]. Adv Robot, 2012, 15(2): 191-204.
[12] RAIBERT M H, BROWN H B, CHEPPONIS M. Experiments in Balance with a 3d One-Legged Hopping Machine [J]. Int J Robot Res, 1984, 3(2): 75-92.
[13] RAIBERT M, CHEPPONIS M, BROWN H. Running on four legs as though they were one [J]. IEEE Journal on Robotics and Automation, 1986, 2(2): 70-82.
[14] RAIBERT M H. Legged robots that balance [M]. MIT press, 1986.
[15] NELSON G, BLANKESPOOR K, RAIBERT M. Walking BigDog: Insights and challenges from legged robotics [J]. Journal of biomechanics, 2006, (39): S360.
[16] RAIBERT M, BLANKESPOOR K, NELSON G, et al. BigDog, the Rough-Terrain Quadruped Robot [J]. IFAC Proceedings Volumes, 2008, 41(2): 10822-5.
[17] DYNAMICS B. LS3-Legged Squad Support System [Z]. https://www.youtube.com/watch?v=R7ezXBEBE6U. 2012
[18] DYNAMICS B. Introducing WildCat [Z]. https://www.youtube.com/watch?v=wE3fmFTtP9g&t=62s. 2013
[19] DYNAMICS B. Introducing Spot (previously SpotMini) [Z]. https://www.youtube.com/watch?v=tf7IEVTDjng. 2016
[20] DYNAMICS B. Spot Launch [Z]. https://www.youtube.com/watch?v=wlkCQXHEgjA. 2019
[21] HABERLAND M, KARSSEN J D, KIM S, et al. The effect of swing leg retraction on running energy efficiency; proceedings of the 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems, F, 2011 [C]. IEEE.
[22] KARSSEN J D, HABERLAND M, WISSE M, et al. The optimal swing-leg retraction rate for running; proceedings of the 2011 IEEE International Conference on Robotics and Automation, F, 2011 [C]. IEEE.
[23] SEOK S, WANG A, OTTEN D, et al. Actuator Design for High Force Proprioceptive Control in Fast Legged Locomotion [J]. 2012 Ieee/Rsj International Conference on Intelligent Robots and Systems (Iros), 2012: 1963-8.
[24] VALENZUELA A K, KIM S. Optimally scaled hip-force planning: A control approach for quadrupedal running; proceedings of the 2012 IEEE International Conference on Robotics and Automation, F, 2012 [C]. IEEE.
[25] PARK H W, PARK S, KIM S. Variable-speed Quadrupedal Bounding Using Impulse Planning: Untethered High-speed 3D Running of MIT Cheetah 2 [J]. Ieee Int Conf Robot, 2015: 5163-70.
[26] 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): 167-92.
[27] BLEDT G, POWELL M J, KATZ B, et al. MIT Cheetah 3: Design and Control of a Robust, Dynamic Quadruped Robot [Z]. Ieee Int C Int Robot. 2018: 2245-52.10.1109/iros.2018.8593885
[28] KATZ B G. A low cost modular actuator for dynamic robots [D]; Massachusetts Institute of Technology, 2018.
[29] HUTTER M, REMY C D, HöPFLINGER M A, et al. SLIP running with an articulated robotic leg; proceedings of the 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems, F 18-22 Oct. 2010, 2010 [C].
[30] HUTTER M, REMY C D, HOEPFLINGER M A, et al. Efficient and Versatile Locomotion With Highly Compliant Legs [J]. IEEE/ASME Transactions on Mechatronics, 2013, 18(2): 449-58.
[31] HUTTER M, GEHRING C, JUD D, et al. ANYmal - A Highly Mobile and Dynamic Quadrupedal Robot [J]. 2016 Ieee/Rsj International Conference on Intelligent Robots and Systems (Iros 2016), 2016: 38-44.
[32] GEHRING C, BELLICOSO C D, COROS S, et al. Dynamic Trotting on Slopes for Quadrupedal Robots [J]. 2015 Ieee/Rsj International Conference on Intelligent Robots and Systems (Iros), 2015: 5129-35.
[33] FANKHAUSER P, BELLICOSO C D, GEHRING C, et al. Free Gait - An Architecture for the Versatile Control of Legged Robots [J]. Ieee-Ras Int C Human, 2016: 1052-8.
[34] GEHRING C, COROS S, HUTTER M, et al. Practice Makes Perfect: An Optimization-Based Approach to Controlling Agile Motions for a Quadruped Robot [J]. IEEE Robotics & Automation Magazine, 2016, 23(1): 34-43.
[35] BELLICOSO C D, JENELTEN F, FANKHAUSER P, et al. Dynamic Locomotion and Whole-Body Control for Quadrupedal Robots [J]. 2017 Ieee/Rsj International Conference on Intelligent Robots and Systems (Iros), 2017: 3359-65.
[36] 张秀丽. 四足机器人节律运动及环境适应性的生物控制研究 [D]; 清华大学, 2004.
[37] ZHANG X L, HAOJUN Z J, LIU P, et al. Designing a quadrupedal robot mimicking cat locomotion [J]. Ieee Sys Man Cybern, 2006: 4471-+.
[38] ZHANG X L, ZHENG H J, GUAN X, et al. A biological inspired quadruped robot: Structure and control [J]. 2005 Ieee International Conference on Robotics and Biomimetics, 2006: 387-+.
[39] ZHANG X, ZHENG H, CHEN L. Gait transition for a quadrupedal robot by replacing the gait matrix of a central pattern generator model [J]. Adv Robot, 2012, 20(7): 849-66.
[40] 何冬青, 马培荪. 四足机器人动态步行仿真及步行稳定性分析 [J]. 计算机仿真, 2005, 22(2): 146-9.
[41] 李彬. 视觉地形分类和四足机器人步态规划方法研究与应用 [D]; 山东大学, 2012.
[42] 李贻斌, 李彬, 荣学文, et al. 液压驱动四足仿生机器人的结构设计和步态规划 [J]. 山东大学学报(工学版), 2011, 41(05): 32-6+45.
[43] 柴汇, 孟健, 荣学文, et al. 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(04): 385-91.
[44] 柴汇. 液压驱动四足机器人柔顺及力控制方法的研究与实现 [D]; 山东大学, 2016.
[45] 刘斌. 四足机器人落地过程中缓冲策略的研究 [D]; 山东大学, 2017.
[46] UNITREE. Robot [Z]. https://www.unitree.com/cn/. 2020
[47] 王兴兴. 新型电驱动式四足机器人研制与测试 [D]; 上海大学, 2016.
[48] 朱秋国. 浅谈四足机器人的发展历史、现状与未来 [J]. 杭州科技, 2017, (02): 47-50.
[49] 朱秋国. “绝影”机器人助力智慧安防 [J]. 中国测绘, 2019, (03): 31-3.
[50] 周坤, 李川, 李超, et al. 面向未知复杂地形的四足机器人运动规划方法 [J]. 机械工程学报, 2020, 56(02): 210-9.
[51] 云深处科技. 绝影 [Z]. https://www.deeprobotics.cn/products.html. 2020
[52] 蔡润斌. 四足机器人运动规划及协调控制 [D]; 国防科学技术大学, 2013.
[53] MCGHEE R B. Some finite state aspects of legged locomotion [J]. Mathematical Biosciences, 1968, 2(1-2): 67-84.
[54] BLEDT G, WENSING P M, KIM S. Policy-regularized model predictive control to stabilize diverse quadrupedal gaits for the MIT cheetah; proceedings of the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), F, 2017 [C]. IEEE.
[55] BLEDT G, WENSING P M, INGERSOLL S, et al. Contact model fusion for event-based locomotion in unstructured terrains; proceedings of the 2018 IEEE International Conference on Robotics and Automation (ICRA), F, 2018 [C]. IEEE.
[56] 熊蓉. 仿生腿足式机器人的发展——浙江大学控制学院机器人实验室熊蓉教授谈国内外腿足式机器人研究情况 [J]. 机器人技术与应用, 2017, 000(002): 29-36.
[57] ARM P, ZENKL R, BARTON P, et al. Spacebok: A dynamic legged robot for space exploration; proceedings of the 2019 international conference on robotics and automation (ICRA), F, 2019 [C]. IEEE.
[58] KAU N, SCHULTZ A, FERRANTE N, et al. Stanford doggo: An open-source, quasi-direct-drive quadruped; proceedings of the 2019 International conference on robotics and automation (ICRA), F, 2019 [C]. IEEE.
[59] SEOK S, WANG A, CHUAH M Y, et al. Design Principles for Highly Efficient Quadrupeds and Implementation on the MIT Cheetah Robot [J]. 2013 Ieee International Conference on Robotics and Automation (Icra), 2013: 3307-12.
[60] GALLARDO-ALVARADO J, RODRíGUEZ-CASTRO R, DELOSSANTOS-LARA P J. Kinematics and dynamics of a 4- P RUR Schönflies parallel manipulator by means of screw theory and the principle of virtual work [J]. Mechanism and Machine Theory, 2018, 122: 347-60.
[61] DE JONG J J, VAN DIJK J, HERDER J L. A screw based methodology for instantaneous dynamic balance [J]. Mechanism and Machine Theory, 2019, 141: 267-82.
[62] DU W, FNADI M, BENAMAR F. Rolling based locomotion on rough terrain for a wheeled quadruped using centroidal dynamics [J]. Mechanism and Machine Theory, 2020, 153.
[63] FEATHERSTONE R. Rigid Body Dynamics Algorithms [J]. Springer US, Boston, MA,, 2008.
[64] KIMURA H, FUKUOKA Y, COHEN A H. Biologically inspired adaptive walking of a quadruped robot [J]. Philos Trans A Math Phys Eng Sci, 2007, 365(1850): 153-70.
[65] POLET D T, BERTRAM J E A. An inelastic quadrupedal model discovers four-beat walking, two-beat running, and pseudo-elastic actuation as energetically optimal [J]. Plos Comput Biol, 2019, 15(11).
[66] GARCIA E, DE SANTOS P G. On the improvement of walking performance in natural environments by a compliant adaptive gait [J]. Ieee T Robot, 2006, 22(6): 1240-53.
[67] VUKOBRATOVIĆ M, BOROVAC B. Zero-moment point—thirty five years of its life [J]. International journal of humanoid robotics, 2004, 1(01): 157-73.