JRM Vol.27 No.1 pp. 57-63
doi: 10.20965/jrm.2015.p0057


Self-Generation of Reward by Moderate-Based Index for Senor Inputsvspace

Kentarou Kurashige and Kaoru Nikaido

Department of Information and Electronic Engineering, Muroran Institute of Technology
27-1 Mizumoto-cho, Muroran, Hokkaido 050-8585, Japan

August 17, 2014
December 19, 2014
February 20, 2015
reward generation, reinforcement learning, pleasure and pain, robot-human interaction, inborn index and immunity evaluation
Moderate-based reward generator

In conventional reinforcement learning, a reward function influences the learning results, and therefore, the reward function is very important. To design this function considering a task, knowledge of reinforcement learning is required. In addition to this, a reward function must be designed for each task. These requirements make the design of a reward function unfeasible. We focus on this problem and aim at realizing a method to generate a reward without the design of a special reward function. In this paper, we propose a universal evaluation for sensor inputs, which is independent of a task and is modeled on the basis of the indicator of pleasure and pain in biological organisms. This evaluation estimates the trend of sensor inputs based on the ease of input prediction. Instead of the design of a reward function, our approach assists a human being in learning how to interact with an agent and teaching it his/her demand. We recruited a research participant and attempted to solve the path planning problem. The results show that a participant can teach an agent his/her demand by interacting with the agent and the agent can generate an adaptive route by interacting with the participant and the environment.

Cite this article as:
K. Kurashige and K. Nikaido, “Self-Generation of Reward by Moderate-Based Index for Senor Inputsvspace,” J. Robot. Mechatron., Vol.27, No.1, pp. 57-63, 2015.
Data files:
  1. [1] R. S. Sutton and A. G. Barto, “Reinforcement Learning,” The MIT Press, 1998.
  2. [2] M. Riedmiller, T. Gabel, R. Hafner, and S. Lange, “Reinforcement learning for robot soccer,” Autonomous Robots, Vol.27, No.1 pp. 57-73, 2009.
  3. [3] R. Yamashina, M. Kuroda, and T. Yabuta, “Caterpillar Robot Locomotion Based on Q-Learning using Objective/Subjective Reward,” Proc. of IEEE/SICE Int. Symposium on System Integration (SII 2011), pp. 1311-1316, 2011.
  4. [4] M. Hara, N. Kawabe, J. Huang, and T. Yabuta, “Acquisition of a Gymnast-Like Robotic Giant-Swing Motion by Q-Learning and Improvement of the Repeatability,” J. of Robotics and Mechatronics, Vol.23, No.1, pp.126-136, 2011.
  5. [5] K. Inoue, T. Arai, and J. Ota, “Acceleration of Reinforcement Learning by a Mobile Robot Using Generalized Inhibition Rules,” J. of Robotics and Mechatronics, Vol.22, No.1, pp. 122-133, 2010. Vol.22, No.1, 2010.
  6. [6] S. Aoyagi and K. Hiraoka, “Path Searching of Robot Manipulator Using Reinforcement Learning -- Reduction of Searched Configuration Space Using SOM and Multistage Learning --,” J. of Robotics and Mechatronics, Vol.22, No.4, pp. 532-541, 2010.
  7. [7] K. Yamada, “Expression of Continuous State and Action Spaces for Q-Learning Using Neural Networks and CMAC,” J. of Robotics and Mechatronics, Vol.24, No.2, pp. 330-339, 2012.
  8. [8] P. Weng, R. Busa-Fekete, and E. Hüllermeier, “Interactive Q-Learning with Ordinal Rewards and Unreliable Tutor,” ECML/PKDD Workshop Reinforcement Learning with Generalized Feedback, 2013.
  9. [9] S. Whiteson, “Evolutionary Computation for Reinforcement Learning” in M. Wiering and M. van Otterlo (Eds.), Reinforcement Learning: State of the Art, pp. 325-358, Springer, 2012.
  10. [10] K. Kurashige and Y. Onoue, “The robot learning by using “sense of pain”,” Proc. of Int. Symposium on Humanized Systems 2007, pp. 1-4, 2007.
  11. [11] J. A. Starzyk, “Motivation in Embodied Intelligence,” in Frontiers in Robotics, Automation and Control, I-Tech Education and Publishing, pp. 83-110, Oct. 2008.
  12. [12] J. A. Starzyk, “Motivated Learning for Computational Intelligence,” in B. Igelnik (Ed.), Computational Modeling and Simulation of Intellect: Current State and Future Perspectives, IGI Publishing, ch.11, pp. 265-292, 2011.
  13. [13] S. Sugimoto, “The Effect of Prolonged Lack of Sensory Stimulation upon Human Behavior,” Philosophy, Vol.50, pp. 361-374, 1967.
  14. [14] S. Sugimoto, “Human Mental Processes under Sensory Restriction Environment,” The Japanese Society of Social Psychology, Vol.1, No.2, pp. 27-34, 1986.
  15. [15] N. Matsunaga, A. T. Zengin, H. Okajima, and S. Kawaji, “Emulation of Fast and Slow Pain Using Multi-Layered Sensor Modeled the Layered Structure of Human Skin,” J. of Robotics and Mechatronics, Vol.23, No.1, pp. 173-179, 2011.
  16. [16] J. Zhen, H. Aoki, E. Sato-Shimokawara, and T. Yamaguchi, “Obtaining Objects Information from a Human Robot Interaction using Gesture and Voice Recognition,” IWACIII 2011 Proc., 101_GS1_1, 2011.
  17. [17] S. Hashimoto, A. Ishida, M. Inami, and T. Igarashi, “TouchMe: An Augmented Reality Interface for Remote Robot Control,” J. of Robotics and Mechatronics, Vol.25, No.3, pp. 529-537, 2013.
  18. [18] N. Kubota and Y. Urushizaki, “Communication Interface for Human-Robot Partnership,” J. of Robotics and Mechatronics, Vol.16, No.5, pp. 526-534, 2004.
  19. [19] M. Quigley, B. Gerkey, K. Conley, J. Faust, T. Foote, J. Leibs, E. Berger, R. Wheeler, and A. Ng, “ROS: An open-source Robot Operating System,” ICRA Workshop on Open Source Software, 2009.

*This site is desgined based on HTML5 and CSS3 for modern browsers, e.g. Chrome, Firefox, Safari, Edge, IE9,10,11, Opera.

Last updated on Nov. 15, 2018