Chenyang Zhao
ABSTRACT
Tankendo which is a form of martial arts that originated in Japan is suitable for even aged persons or women and is good for promotion of health since a short and easy to maneuver shinai (bamboo sword) is used during matches as well as during practice. However, since the population of persons active in Tankendo is small, it is not easy to gain experience of actual play against other human players. In this research, the authors propose a method of control of a Tankendo robot that employs a trolley robot that uses a mecanum wheel as the movement mechanism. Although this robot cannot start suddenly or stop suddenly like humans, since it has seven pivots, it is possible to finely control the direction or posture of the vehicle as well as its speed. The control commands are of the three types of “speed”, “direction of travel”, and “rotation ratio (curvature of the curve)”, and, in addition, trapezoidal speed control is being carried out in order to prevent loss of synchronization. In other words, there is a time delay until the target speed of the control command is reached. Using a trolley robot with these characteristics, the inputs are the "robot state" (position, posture, etc.) and the "opponent state" (direction of the shinai, movement speed, distance from the opponent, etc.) at the current instant of time, and the "robot control commands" from several instants of time earlier. For the output, an optimal reference table using "speed," "direction of travel," and " rotation ratio" is machine-learned using reinforcement learning. The authors propose a robot control method that can defend against a human attack by moving to an appropriate position at an appropriate time, just before the attack becomes effective strike.
- Do Tanken, Enjoy Tankendo. https://youtu.be/50jKcqMsBsoGoogle Scholar
- M. Nakamura and K. Hoshino. 2018. Tankendo motion estimation system with robustness against differences in color and size between users’ clothes using 4-color markers with elastic belts,” Proc. Intl. Conf. on Biomedical and Bioinformatics Engineering, pp.110-117. http://dx.doi.org/10.1145/3301879.3301890Google ScholarDigital Library
- Y. Ozone, C. Zhao, and K. Hoshino. 2020. “Introducing Swordsman Robot Defensing,” Proc. 3rd Intl. Conf. on Digital Medicine and Image Processing, 14 pages, https://doi.org/10.1145/3441369.3441375Google ScholarDigital Library
- K. Hoshino and C. Cheng. 2016. "Development of a swordsmanship machine enabling the inner and outer muscles to be safely trained while having fun," Proc. 3rd Intl. Conf. on Biomedical and Bioinformatics Engineering, pp.59-62. http://dx.doi.org/10.1145/3022702.3022711Google ScholarDigital Library
- K. Hoshino, M. Nakamura, Y. Nakai, Y. Ozone, S. Shimanoe, and Y. Noguchi. 2019. “A World Camera for Recording the Game Tactics in Martial Arts Using Bamboo Swords,” Intl. Conf. on Biomedical and Bioinformatics Engineering, 6, pp.1-6. https://dl.acm.org/doi/10.1145/3375923. 3375955Google Scholar
- K. Hoshino, Y. Noguchi, Y. Nakai, Y. Ozone, and M. Nakamura. 2019. “Predicting the Types of Striking and Thrusting Motions by Using Deep Learning,” Intl. Conf. on Biomedical and Bioinformatics Engineering, 6, pp.1-4. https://dl.acm.org/doi/10.1145/3375923.3375954Google Scholar
- K. Hoshino, Y. Ozone, and C. Zhao. 2021. “A Tankendo Robot that Defends Itself from Opponent Attacks by Moving Immediately before a Valid Strike or Thrust,” Proc. Intl. Conf. on Image, Video and Signal Processing Proceedings (IVSP 2021), 10 pages. https://doi.org/10.1145/3459212.3459230Google ScholarDigital Library
- J Ibarz, J Tan, C Finn, M Kalakrishnan, P Pastor, S Levine: “How to train your robot with deep reinforcement learning: lessons we have learned,” The International Journal of Robotics Research, 40(4-5): 698-721, 2021.Google ScholarDigital Library
- A R Mahmood, D Korenkevych, B J Komer, J Bergstra. 2018. “Setting up a reinforcement learning task with a real-world robot,” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 4635-4640, 2018.Google Scholar
- M Abreu, N Lau, A Sousa, L P Reis. 2019. “Learning low level skills from scratch for humanoid robot soccer using deep reinforcement learning,” 2019 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC). IEEE, 1-8.Google ScholarCross Ref
- C R Gil, H Calvo, H Sossa. 2019. “Learning an efficient gait cycle of a biped robot based on reinforcement learning and artificial neural networks,” Applied Sciences, 9(3): 502.Google ScholarCross Ref
- H Oliff, Y Liu, M Kumar, M Williams, M Ryan. 2020. “Reinforcement learning for facilitating human-robot-interaction in manufacturing,” Journal of Manufacturing Systems, 56: 326-340.Google ScholarCross Ref
- Y Wen, J Si , A Brandt, X Gao, H Huang. 2019. “Online reinforcement learning control for the personalization of a robotic knee prosthesis,” IEEE transactions on cybernetics, 50(6): 2346-2356.Google Scholar
- Y Cheng, W Zhang. 2016. “Concise deep reinforcement learning obstacle avoidance for underactuated unmanned marine vessels.” Neurocomputing, 272: 63-73, 2018.Google ScholarDigital Library
- M Duguleana, G Mogan: “Neural networks based reinforcement learning for mobile robots obstacle avoidance[J]. Expert Systems with Applications”, 62: 104-115.Google Scholar
- J. Schulman, F. Wolski, P. Dhariwal, A. Radford, O. 2017. Klimov: “Proximal Policy Optimization Algorithms,” arXiv:1707.06347v2 [cs.LG], 12 pages.Google Scholar
Recommendations
A transformable wheel-legged mobile robot
This paper proposes a new type of transformable wheel-legged mobile robot that could be applied on both flat and rugged terrains. It integrates stability and maneuverability of wheeled robot and obstacle climbing capability of legged robot by means of a ...
A biologically inspired method for robot navigation in a cluttered environment
The problem of wheeled mobile robot (WMR) navigation toward an unknown target in a cluttered environment has been considered. The biologically inspired navigation algorithm is the equiangular navigation guidance (ENG) law combined with a local obstacle ...
Movement Performance Analysis of Mecanum Wheeled Omnidirectional Mobile Robot
2019 IEEE International Conference on Mechatronics and Automation (ICMA)At present, omnidirectional mobile robot is widely used for its flexible movement. Omnidirectional movement realization can be divided into two forms: omnidirectional wheel with special structure and ordinary wheel with steering mechanism. The ...
Comments