Control of Nameplate Pasting Robot for Sand Mold Based on Deep Reinforcement Learning
In order to solve the problem of low-efficiency in the manual operation process of nameplate pasting for sand mold, an intelligent simulation system based on visual sensing and industrial robot is designed to paste nameplate on sand molds, and a deep reinforcement learning control method is proposed. The simulation system including the robot, visual sensor and sand mold is established in ROS combined with the physical simulation engine Gazebo. Then the task of nameplate pasting for sand molds is expressed as a markov process and the robot is trained by DQN method to learn a strategy to complete the task of pasting the nameplate of sand mold. A multi-level reward function algorithm based on multi-distances and collision information is proposed to improve the train success rate. Finally, the method is verified in the simulation system. The results show that the nameplate can be quickly attached to the sand mold cavity by the industrial robot.
KeywordsSand mold Nameplate pasting Industrial robot DQN
This work is partially supported by the National Science Foundation for Young Scientists of China (Grant No. 51805071), the Fundamental Research Funds for the Central Universities (Grant No. DUT18RC(3)073) and Changjiang Scholar Program of Chinese Ministry of Education (No. T2017030).
- 3.Bo, T., Zeyu, G., Han, D.: Survey on uncalibrated robot visual servoing control. Chin. J. Theor. Appl. Mech. 48(4), 767–783 (2016)Google Scholar
- 4.Jia, B., Liu, S., Zhang, K., Chen, J.: Survey on robot visual servo control: vision system and control strategies. AAS 41(5), 861–873 (2015)Google Scholar
- 6.Malis, E., Rives, P.: Robustness of image-based visual servoing with respect to depth distribution errors. In: Proceedings of the 2003 IEEE International Conference on Robotics and Automation, Taipei, China, pp. 1056–1061. IEEE (2003)Google Scholar
- 8.Mnih, V., et al.: Playing atari with deep reinforcement learning. arXiv preprint arXiv:1312.5602 (2013)
- 9.Zhang, F., Leitner, J., Milford, M., et al.: Towards vision-based deep reinforcement learning for robotic motion control. Comput. Sci. (2015)Google Scholar
- 10.Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. 2nd edition in progress. London, England (2017)Google Scholar
- 11.Xian, G.: The study of robotic arm control policy based on DQN. Beijing Jiaotong University, pp. 41–45 (2018)Google Scholar
- 12.James, S., Johns, E.: 3D simulation for robot arm control with deep Q-learning. arXiv preprint (2016)Google Scholar