Abstract
Evolutionary Computation (EC) has strengths in terms of computation for gait optimization. However, conventional evolutionary algorithms use typical gait parameters such as step length and swing height, which limit the trajectory deformation for optimization of the foot trajectory. Furthermore, the quantitative index of fitness convergence is insufficient. In this paper, we perform gait optimization of a quadruped robot using foot placement perturbation based on EC. The proposed algorithm has an atypical solution search range, which is generated by independent manipulation of each placement that forms the foot trajectory. A convergence index is also introduced to prevent premature cessation of learning. The conventional algorithm and the proposed algorithm are applied to a quadruped robot; walking performances are then compared by gait simulation. Although the two algorithms exhibit similar computation rates, the proposed algorithm shows better fitness and a wider search range. The evolutionary tendency of the walking trajectory is analyzed using the optimized results, and the findings provide insight into reliable leg trajectory design.
Similar content being viewed by others
References
Hutter M, Gehring C, Höpflinger M A, Blösch M, Siegwart R. Toward combining speed, efficiency, versatility, and robustness in an autonomous quadruped. IEEE Transactions on Robotics, 2014, 30, 1427–1440.
Kolter J Z, Rodgers M P, Ng A Y. A control architecture for quadruped locomotion over rough terrain. IEEE International Conference on Robotics and Automation (ICRA), Pasadena, CA, USA, 2008, 811–818.
Ugurlu B, Havoutis I, Semini C, Caldwell D G. Dynamic trot-walking with the hydraulic quadruped robot-hyQ: Analytical trajectory generation and active compliance control. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Tokyo, Japan, 2013, 6044–6051.
Sakakibara Y, Kan K, Hosoda Y, Hattori M, Fujie M. Foot trajectory for a quadruped walking machine. IEEE International Workshop on Intelligent Robots and Systems, Towards a New Frontier of Applications, Ibaraki, Japan, 1990, 315–322.
Sun L, Meng M Q H, Chen W M, Liang H W, Mei T. Design of quadruped robot based neural network. In Advances in Neural Networks, Liu D, Fei S, Hou Z G, Zhang H, Sun C, eds., Springer, Berlin, Germany, 2007, 843–851.
Kim K Y, Park J H. Ellipse-based leg-trajectory generation for galloping quadruped robots. Journal of mechanical science and technology, 2008, 22, 2099–2106.
Righetti L, Ijspeert A J. Pattern generators with sensory feedback for the control of quadruped locomotion. IEEE International Conference on Robotics and Automation (ICRA), Pasadena, USA, 2008, 819–824.
Santos C P, Matos V. Gait transition and modulation in a quadruped robot: A brainstem-like modulation approach. Robotics and Autonomous Systems, 2011, 59, 620–634.
Sprowitz A T, Ajallooeian M, Tuleu A, Ijspeert A J. Kinematic primitives for walking and trotting gaits of a quadruped robot with compliant legs. Frontiers in Computational Neuroscience, 2014, 8, 27.
Wu X, Shao X S, Wang W. Stable quadruped walking with the adjustment of the center of gravity. IEEE International Conference on Mechatronics and Automation (ICMA), Takamatsu, Kagawa, Japan, 2013, 1123–1128.
Wang B X, Wan Z W, Zhou C L, Wu J, Qiu Y F, Gao Z J. A multi-module controller for walking quadruped robots. Journal of Bionic Engineering, 2019, 16, 253–263.
Zhang J Q, Gao F, Han X L, Chen X B, Han X Y. Trot gait design and CPG method for a quadruped robot. Journal of Bionic Engineering, 2014, 11, 18–25.
Kim M S, Uther W. Automatic gait optimisation for quadruped robots. Australasian Conference on Robotics and Automation, Brisbane, Australia, 2003, 1–3.
Kohl N, Stone P. Machine learning for fast quadrupedal locomotion. AAAI, 2004, 4, 611–616.
Kohl N, Stone P. Policy gradient reinforcement learning for fast quadrupedal locomotion. IEEE International Conference on Robotics and Automation (ICRA), New Orleans, LA, USA, 2004, 3, 2619–2624.
Şen M A, Kalyoncu M. Optimal tuning of PID controller using grey wolf optimizer algorithm for quadruped robot. Balkan Journal of Electrical and Computer Engineering, 2018, 6, 29–35.
Izumi K, Yamaguchi T, Watanabe K. Free-gait of quadruped robots using neural networks for determining the order of swing leg. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Sendai, Japan, 2004, 4, 3400–3405.
Tan J, Zhang T, Coumans E, Iscen A, Bai Y, Hafner D, Vanhoucke V. Sim-to-real: Learning agile locomotion for quadruped robots. arXiv, 2018.
Hu B, Shao S B, Cao Z C, Xiao Q, Li Q Z, Ma C. Learning a faster locomotion gait for a quadruped robot with model-free deep reinforcement learning. IEEE International Conference on Robotics and Biomimetics (ROBIO), Dali, China, 2019, 1097–1102.
SONY Corporation of America: Sony-aibo. [2020-10-01], http://us.aibo.com/, 2018.
Chernova S, Veloso M. An evolutionary approach to gait learning for four-legged robots. IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Sendai, Japan, 2004, 3, 2562–2567.
Hornby G S, Takamura S, Yamamoto T, Fujita M. Autonomous evolution of dynamic gaits with two quadruped robots. IEEE Transactions on Robotics, 2005, 21, 402–410.
Chae K G, Park J H. Trajectory optimization with GA and control for quadruped robots. Journal of Mechanical Science and Technology, 2009, 23, 114–123.
Golubovic D, Hu H S. Parameter optimisation of an evolutionary algorithm for on-line gait generation of quadruped robots. IEEE International Conference on Industrial Technology, Maribor, Slovenia, 2003, 1, 221–226.
Röfer T. Evolutionary gait-optimization using a fitness function based on proprioception. Robot Soccer World Cup, 2004, 310–322.
Maleki S, Parsa A, Ahmadabadi M N. Modeling, control and gait design of a quadruped robot with active spine towards energy efficiency. 3rd RSI International Conference on Robotics and Mechatronics (ICROM), Tehran, Iran, 2015, 271–276.
Kočo E, Kovačić Z. Multiobjective locomotion optimization of quadruped robot with different 2DOF configurations of actuated spine. 24th Mediterranean Conference on Control and Automation (MED), Athens, Greece, 2016, 504–511.
Gong D X, Yan J, Zuo G Y. A review of gait optimization based on evolutionary computation. Applied Computational Intelligence and Soft Computing, 2010.
Such F P, Madhavan V, Conti E, Lehman J, Stanley K O, Clune J. Deep neuroevolution: Genetic algorithms are a competitive alternative for training deep neural networks for reinforcement learning. arXiv, 2017.
Siddique M N H, Tokhi M O. Training neural networks: backpropagation vs genetic algorithms. International Joint Conference on Neural Networks, Washington, DC, USA, 2001, 4, 2673–2678.
Fan X M, Sayers W, Zhang S J, Han Z W, Ren L Q, Chizari H. Review and classification of bio-inspired algorithms and their applications. Journal of Bionic Engineering, 2020, 17, 611–631.
Day L M, Jayne B C. Interspecific scaling of the morphology and posture of the limbs during the locomotion of cats (felidae). Journal of Experimental Biology, 2007, 210, 642–654.
Pike A V L, Alexander R M. The relationship between limb-segment proportions and joint kinematics for the hind limbs of quadrupedal mammals. Journal of Zoology, 2002, 258, 427–433.
Coppelia Robotics: V-rep, [2021-03-06], https://www.coppeliarobotics.com
Acknowledgment
This work was supported in part by the National Research Foundation of Korea (NRF) Grant funded by the Korean Government (MSIT) (No.NRF-2019R1A2 C2084677), and in part by the 2021 Research Fund (1.210052.01) of UNIST (Ulsan National Institute of Science and Technology).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Kim, J., Ba, D.X., Yeom, H. et al. Gait Optimization of a Quadruped Robot Using Evolutionary Computation. J Bionic Eng 18, 306–318 (2021). https://doi.org/10.1007/s42235-021-0026-y
Published:
Issue Date:
DOI: https://doi.org/10.1007/s42235-021-0026-y