DOI QR코드

DOI QR Code

Dynamic Obstacle Avoidance and Optimal Path Finding Algorithm for Mobile Robot Using Q-learning

Q-learning을 이용한 모바일 로봇의 동적 장애물 회피 및 최적 경로 탐색 알고리즘

  • Received : 2017.08.14
  • Accepted : 2017.09.03
  • Published : 2017.09.30

Abstract

After the recent Go competition of Alphgo versus Sedol Lee, people have been interested in machine learning. When Alphago learned Go, Alphago used reinforcement learning that is a part of machine learning. Reinforcement learning is a learning algorithm that chooses the action based on the environment without the exact system model and has been successfully applied to scheduling and board game such as chess. Also, it is used in the field of robot control but the existing researches confirms the performance through the simulation not suggesting the solutions when there are dynamic obstacles. Therefore, in this paper we implemented an algorithm for a mobile robot control system that can find the optimal path by learning the surrounding environment through Q-learning and avoid dynamic obstacles. Experiments have shown that the mobile robot with this algorithm can successfully avoid obstacles.

최근 알파고와 이세돌 9단의 바둑 경기를 통해 머신러닝에 대한 관심이 크게 증폭되고 있다. 알파고가 바둑을 배울 때 사용한 알고리즘은 머신러닝의 한 분야인 강화 학습(Reinforcement Learning)이다. 강화 학습은 정확한 시스템의 모델 없이 환경에 따라 수행할 행동을 선택하는 학습방법이며, 주로 스케줄링과 체스와 같은 게임에 성공적으로 적용됐다. 또한 로봇 제어 분야에서도 사용되고 있으나, 기존의 연구들은 대부분 모의실험을 통해 그 성능을 확인할 뿐 동적 장애물이 존재할 시에 대한 해답을 제시하지 않고 있다. 따라서 본 논문에서는 Q-learning을 통해 주변 환경을 학습하고 난 후 동적 장애물이 존재할 때 이를 회피할 수 있는 알고리즘을 구현하였으며, 실험을 통해 성공적으로 장애물을 회피해 가는 모습을 확인하였다.

Keywords

References

  1. K. J. Baik and B. J. Jang, "Hand Gesture Classification Using Multiple Doppler Radar and Machine Learning", Journal of Korean Institute of Electromagnetic Engineering and Science, Vol. 28, No. 1, pp. 33-41, Jan. 2017. https://doi.org/10.5515/KJKIEES.2017.28.1.33
  2. J. K. Kim, B. D. Kim, D. W. Yoon, and J. W. Choi, "Deep Neural Network-based Automatic Modulation Classification Technique" Journal of KIIT, Vol. 14, No. 12, pp. 107-115, Dec. 2016.
  3. M. S. An and D. S. Kang, "Development of Image Analysis System Using Object Classifier basd on Deep Convolutional Neural Network", Journal of KIIT, Vol. 14, No. 5, pp. 67-73, May 2016.
  4. S. C. Park, M. E. Lee, S. H. Kim, I. S. Na, and Y. J. Chen, "Machine Learning for Medical Image Analysis", Journal of KISS : Software and Applications, Vol. 39, No. 3, pp. 163-174, Mar. 2012.
  5. J. S. Han and K. C. Kwak, "Image Classification Using Convolutional Neural Network and Extreme Learning Machine Classifier Based on ReLU Function", Journal of KIIT, Vol. 15, No. 2, pp. 15-23, Feb. 2017. https://doi.org/10.14801/jkiit.2017.15.2.15
  6. D. H. Min, K. W. Jung, K. Y. Kwon, and J. Y. Park, "Mobile Robot Control Based on A Recent Reinforcement Learning Method", Journal of Korean Institute of Intelligent Systems, Vol. 21, No. 1, pp. 67-70, Apr. 2011.
  7. S. W. Seo, H. D. Kim, and K. B. Sim, "Reinforcement Learning Based Evolution and Learning Algorithm for Cooperative Behavior of Swarm Robot System", Journal of Korean Institute of Intelligent Systems, Vol. 17, No. 5, pp. 591-597, Oct. 2007. https://doi.org/10.5391/JKIIS.2007.17.5.591
  8. K. B. Sim and D. W. Lee, "Behavior Learning and Evolution of Individual Robot for Cooperative Behavior of Swarm Robot System", Journal of Korean Institute of Intelligent Systems, Vol. 16, No. 2, pp. 131-137, Apr. 2006. https://doi.org/10.5391/JKIIS.2006.16.2.131
  9. J. Huang, B. Yang, and D. Y. Liu, "A Distributed Q-Learning Algorithm for Multi-Agent Team Coordination", Proc of the Fourth International Conference on Machine Learning and Cybernetics, Vol. 1, pp. 108-113, Aug. 2005.
  10. T. Zhou, B. R. Hong, C. X. Shi, and H. Y. Zhou, "Cooperative Behavior Acquisition Based Modular Q Learning in Multi-Agent System", Proc of the Fourth International Conference on Machine Learning and Cybernetics, Vol. 1, pp. 205-210, Aug. 2005.
  11. An Open-Source Software Library for Machine Intelligence, https://www.tensorflow.org/. [Accessed: Jul. 13, 2017]
  12. R. G. Brown and P. Y. C. Hwang, "Introduction to Random Signals and Applied Kalman Filtering", John Wiley & Sons, May 1998.

Cited by

  1. 강화학습을 기반으로 하는 열사용자 기계실 설비의 열효율 향상에 대한 연구 vol.27, pp.2, 2017, https://doi.org/10.5855/energy.2018.27.2.026
  2. Multiobjective model-free learning for robot pathfinding with environmental disturbances vol.16, pp.6, 2017, https://doi.org/10.1177/1729881419885703
  3. VFH+ based 3D Spatial Map using Depth Camera vol.19, pp.7, 2021, https://doi.org/10.14801/jkiit.2021.19.7.35