IAMU Research Project FY2021 Final Report

Evaluation of Sharp-Lookout Skills Using Performance Assessment toward Developing Automatic Target Avoiding System

Publication Year 2022
Research Coordinator Jun Kayano
Research Partners Momoko Kitada; Samuel R. Pecota
Publisher International Association of Maritime Universities (IAMU)
DOI https://doi.org/10.68436/r20210103

Download PDF

Abstract

In this study, we aimed to develop an algorithm of new target avoiding system based on the characteristics of professional lookout skills. Firstly, we constructed an algorithm for searching collision avoidance route by using PSO with OZT as collision risk index. The simulation experiments were carried out and the effectiveness of the proposed algorithm was verified. However, in other simulation experiment for the proposed algorithm applying to all the ships, ships collided or closed to collide to each other because this algorithm calculates the collision avoidance route only where the other ships would not change their behaviors. To improve the algorithm, we tried to identify the characteristics of experienced navigators’ lookout skill. Before that, we administered an online questionnaire among experienced deck officers and less experienced ones to understand sociotechnical barriers to the newly proposed system to assist decision-making of deck officers on watch. Next, to visualize the characteristics of experienced navigators’ lookout skills, based on the results of simulation experiment by using Eye-tracker, we compared the line-of-sight data of skilled ship navigators with the ship maneuvering behavior and considered quantitative data on the µpoint-of-sight’ and µtime-of-sight’ before and after acting for each ship maneuvering. We also measured low frequency/high frequency (LF/HF) values by the heart rate variability to show mental workload. Finally, we focused on the experienced navigators’ skills of predicting future behavior of the other ships. In consideration of uncertainty of the other ships’ future behaviors, we calculated not only OZT but also potential OZT. SVM was also applied to predict the other ship’s behavior as well as to search for collision avoidance route. Though the simulation experiments, the effectiveness of the proposed algorithm was verified.

Keywords

Automatic Collision Avoidance, Lookout Skill, Obstacle Zone by Target

References

  1. Hayama Imazu. "Evaluation method of collision risk by using true motion." TransNav: International Journal on Marine Navigation and Safety of Sea Transportation 11.1 (2017).
  2. Clifford Baker and Denise McCafferty, "Accident database review of human element concerns: What do the results mean for classification." Proc. Int Conf.‘Human Factors in Ship Design and Operation, RINA Feb. 2005.
  3. Hayama Imazu, Junji Fukuto, and Masayoshi Numano. "Obstacle zone by target and its expression." The Journal of Japan Institute of Navigation 107 (2002): 191-197.
  4. Jun Kayano, Hayama Imazu, Hitoi Tamaru and Takahiko Fujisaka, “"On the Prediction of Ship’s Motion for OZT by Using Kalman Filter." The Journal of Japan Institute of Navigation 123 (2010).
  5. Ryohei Sawada, Keiji Sato, and Takahiro Majima. "Automatic ship collision avoidance using deep reinforcement learning with LSTM in continuous action spaces." Journal of Marine Science and Technology 26.2 (2021): 509-524.
  6. Paolo Fiorini, and Zvi Shiller. "Motion planning in dynamic environments using velocity obstacles." The international journal of robotics research 17.7 (1998): 760-772.
  7. Yoshiaki Kuwata, Michael T. Wolf, Dimitri Zarzhitsky, and Terrance L. Huntsberger, "Safe maritime autonomous navigation with COLREGS, using velocity obstacles." IEEE Journal of Oceanic Engineering 39.1 (2013): 110-119.
  8. Huang, Yamin, P. H. A. J. M. Van Gelder, and Yuanqiao Wen. "Velocity obstacle algorithms for collision prevention at sea." Ocean Engineering 151 (2018): 308-321.
  9. Wenjin Zhangi, Chengyong Yan, Hongguang Lyu, Pingling Wang, Zongyao Xue, Zehua Li, and Bai Xiao "COLREGS-based path planning for ships at sea using velocity obstacles." Ieee Access 9 (2021): 32613-32626.
  10. Kazuhiko Hasegawa, Junji Fukuto, Rina Miyake, and Masahiro Yamazaki, "An intelligent ship handling simulator with automatic collision avoidance function of target ships." Proc. of INSLC 17 (2012).
  11. James Kennedy and Russell Eberhart, “Particle Swarm Optimization”, In Proceedings of ICNN'95-international conference on neural networks , (1995) Vol. 4, pp. 1942-1948.
  12. Hayama Imazu and Takeo Koyama, “The Optimization of the Criterion for Collision Avoidance Action-II”, The Journal of Japan Institute of Navigation, Vol.72, (1985), pp.23-30
  13. Kensaku Nomoto and Katashi Taguchi, “On Steering Qualities of Ships (2)” , Journal of the Society of Naval Architects of Japan, Vol. 101, (1957), pp. 57-66
  14. Hayama Imazu, Research on collision avoidance maneuvere., Ph.D. thesis, The University of Tokyo (1987)
  15. WMU, “Transport 2040, Automation, Technology, Employment: The Future of Work”, WMU, (2019).
  16. Christine Chauvin, Jean-Pierre Clostermann, and Jean-Michel Hoc, “Situation Awareness and the Decision-Making Process in a Dynamic Situation: Avoiding Collisions at Sea”, Journal of Cognitive Engineering and Decision Making, Vol. 2, (2008), 1-23.
  17. Yemao Man, Margareta Ltzh|ft, Nicole A. Costa, Monica Lundh, and Scott N. Mackinnon, “Gaps Between Users and Designers: A Usability Study About a Tablet-Based Application Used on Ship Bridges”, Advances in Human Aspects of Transportation, (2018), pp 213-224.
  18. Masao Furusho, Masaaki Tomonaga, Tadahiko Fukuda and Shigetoshi Kai, “A Lookout of the Ship's Operator from the Standpoint of Sight-line Displacement”, The Journal of Japan Institute of Navigation, Vol.97, (1997), pp.231-239
  19. Masao Furusho, “Visual Perception of the Targets for a Good Lookout at Sea”, The Journal of Japan Institute of Navigation, Vol.87, (1992), pp.59-66
  20. Convention on the International Regulations for Preventing Collisions at Sea
  21. Ryoko Fukuda, Minoru Sakuma, Etsuo Nakamura, and Tadahiko Fukuda, “Experimental consideration on the definition of a fixation point”, The Japanese Journal of Ergonomics, Vol.32, No.4, (1996), pp.197-204
  22. nac Image Technology inc. HP, https://www.nacinc.jp/ (2022.5.26)
  23. Koji Murai, Chihiro Nishizaki, Jun Kayano, Hitoi Tamaru amd Ruri Shoji,“Quantitative Evaluation of Good Seamanship using Physiological Data to realize Maritime Autonomous Surface Ships,” Proceedings of 2021 Asia Navigation Conference, S4-A-2 (2021)
  24. Hiroaki Kobayashi and Shoichi Senda, “A Study on the Measurement of Human Mental Work-load in Ship Handling Using SNS Value”, The Journal of Japan Institute of Navigation, Vol. 98, (1998), pp.247-255
  25. B. Scholkopf, C.J.C. Burges, A.J. Smola, Advances in Kernel Methods-Support Vector Learning, The MIT Press (1999), B.Scholkopf and A.Smola, Learning with kernels, MIT Press (2002)
  26. Japan, Russian Federation and United Arab Emirates, Outline of the draft guidelines for MASS operations, presented at Maritime Safety Committee 105th session, IMO (2022)