Abstract
This paper presents an automatic collision avoidance algorithm for multiple ships using reinforcement learning (RL). Obstacle zone by target (OZT) is used to grasp multiple ships’ dynamic information in the form of 2-dimensional areas. OZT shows a dangerous area where collisions may happen. Then a new method using a virtual sensor which is separated in a grid is proposed to detect multiple ships simultaneously. The sensor detects OZTs efficiently and provides information about where OZTs expands as a part of a state vector with a fixed dimension as inputs for a RL algorithm. I applied a deep RL algorithm. An agent of deep RL learned manoeuvre using a set of ship encounter situations called Imazu problem. The learned model can avoid all encounter situations of up to three target ships in simulations. The proposed approach can learn manoeuvre to manage both waypoint navigation and collision avoidance.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
IMO takes first steps to address autonomous ships (2018). http://www.imo.org
Varas, J.M., Hirdaris, S., Smith, R., Scialla, P., Caharija, W., Bhuiyan, Z., Mills, T., Naeem, W., Hu, L., Renton, I., Motson, D., Rajabally, E.: MAXCMAS project: autonomous COLREGs Compliant Ship Navigation. In: Proceedings of the 16th Conference on Computer Applications and Information Technology in the Maritime Industries 2017, pp. 454-464 (2017)
Imazu, H.: Computation of OZT by using collision course. Navigation 188, 78–81 (2014)
Imazu, H.: Evaluation method of collision risk by using true motion. Int. J. Mar. Navig. Saf. Sea Transp. (TransNav) 11(1), 65–70 (2017)
Schulman, J., Levine, S., Abbeel, P., Jordan, M., Moritz, P.: Proximal policy optimization algorithms, arXiv preprint arXiv:1707.06347 (2017)
Mnih, V., Badia, A.P., Mirza, M., Graves, A., Lillicrap, T., Harley, T., Silver, D., Kavukcuoglu, K.: Asynchronous methods for deep reinforcement learning. In: Proceedings of the 33rd International Conference on Machine Learning, PMLR, vol. 48, pp. 1928-1937 (2016)
Imazu, H., Koyama, T.: The optimization of the criterion for collision avoidance action-II. J. Jpn. Inst. Navig. 72, 23–30 (1985)
Imazu, H., Koyama, T.: The optimization of the criterion for collision avoidance action-III. J. Jpn. Inst. Navig. 73, 19–26 (1985)
Kouzuki, A., Hasegawa, K.: Automatic collision avoidance system for ships using fuzzy control. J. Kansai Soc. Nav. Arch. Jpn. 205, 1–10 (1987)
Cai, Y., Hasegawa, K.: Evaluating of marine traffic simulation system through imazu problem. In: The Proceedings of Japan Society of Naval Architecture and Ocean Engineering, vol. 17, pp. 191–194 (2013)
Imazu, H.: Research on collision avoidance manoeuvre (in Japanese). Ph.D. thesis. University of Tokyo, Japan (1987)
Hu, L., Naeem, W., Rajabally, E., Watson, G., Mills, T., Bhuiyan, Z., Salter, I.: COLREGs-compliant path planning for autonomous surface vehicles: a multi-objective optimization approach. In: 20th IFAC World Congress, vol. 50, pp. 13662–13667 (2017)
Nagasawa, A., Hara, K., Inoue, K.: The subjective difficulties of the situation of collision avoidance-I: toward the rating by simulation. J. Jpn. Inst. Navig. 79, 91–100 (1988)
Nagasawa, A., Hara, K., Inoue, K., Kose, K.: The subjective difficulties of the situation of collision avoidance-II: toward the rating by simulation. J. Jpn. Inst. Navig. 88, 137–144 (1993)
Taniguchi, Y., Matsuda, A., Sera, W., Terada, D., Hashimoto, H.: Validation of a ship collision avoidance algorithm in congested sea area by means of model experiment. In: The Proceedings of Japan Society of Naval Architecture and Ocean Engineering, vol. 23, pp. 627–632 (2016)
Kuwata, Y., Wolf, M.T., Zarzhitsky, D., Huntsberger, T.L.: Safe maritime autonomous navigation with COLREGs, using velocity obstacles. IEEE J. Ocean. Eng. 39, 110–119 (2014)
Mitsubori, K., Kamio, T., Tanaka, T.: Finding the course and collision avoidance based on reinforcement learning algorithm. Navigation 170, 26–31 (2009)
Shen, H., Hashimoto, H., Matsuda, A., Taniguchi, Y., Terada, D., Guo, C.: Automatic collision avoidance of multiple ships based on deep Q-learning. Appl. Ocean Res. 86, 268–288 (2019)
Rachman, A.S.A.: 3D-LIDAR multi object tracking for autonomous driving: multi-target detection and tracking under urban road uncertainties, Master thesis, Delft University of Technology, Netherlands (2017)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: Proceedings of the 3rd International Conference on Learning Representations, International Conference on Learning Representations (ICLR) (2015)
DeepX, Inc.: Machina a library for real-world deep reinforcement learning (2019). https://machina-rl.org/
Paszke, A., Gross, S., Chintala, S., Chanan, G., Yang, E., DeVito, Z., Lin, Z., Desmaison, A., Antiga, L., Lerer, A.: Automatic Differentiation in PyTorch, NIPS Autodiff Workshop (2017)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
A Appendix
A Appendix
The configuration of the networks and the hyper-parameters of PPO in this paper are shown in Table 3. Machina supports only Ubuntu and doesn’t supports Windows officially. Therefore, some implementations in the machina code such as the use of multiprocessing in Pytorch and Python and the log output format have been modified for Windows.
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Sawada, R. (2020). Automatic Collision Avoidance Using Deep Reinforcement Learning with Grid Sensor. In: Sato, H., Iwanaga, S., Ishii, A. (eds) Proceedings of the 23rd Asia Pacific Symposium on Intelligent and Evolutionary Systems. IES 2019. Proceedings in Adaptation, Learning and Optimization, vol 12. Springer, Cham. https://doi.org/10.1007/978-3-030-37442-6_3
Download citation
DOI: https://doi.org/10.1007/978-3-030-37442-6_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-37441-9
Online ISBN: 978-3-030-37442-6
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)