The Journal of Korea Institute of Information, Electronics, and Communication Technology (한국정보전자통신기술학회논문지)

Korea Information Electronic Communication Technology (한국정보전자통신기술학회)
권호 표지
DOI QR코드

DOI QR Code

Prevention of Collision with Other Vessels Using Camera Sensors with Kalman Filter

칼만 필터가 적용된 카메라 센서를 이용한 타 선박과의 충돌 예방

  • Dae-il Sung (Department of Electronics and Control Engineering, Republic of Korea Naval Academy) ;
  • Sung-Joo Kim (Department of Electronics and Control Engineering, Republic of Korea Naval Academy) ;
  • Young-Min Kim (Department of Electronics and Control Engineering, Republic of Korea Naval Academy) ;
  • Yun-Sung Jung (Department of Electronics and Control Engineering, Republic of Korea Naval Academy) ;
  • Min-Seok Han (Department of Electronics and Control Engineering, Republic of Korea Naval Academy)
  • 성대일 ;
  • 김성주 ;
  • 김영민 ;
  • 정윤성 ;
  • 한민석
  • Received : 2024.05.30
  • Accepted : 2024.06.12
  • Published : 2024.06.29
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we present a method of applying the kalman filter to control and correct errors in camera sensor recognition depending on the sea state environment. First, the specifications of the ship were described and the degree of error due to rolling was measured. After presenting the distance from the surface of the water to the sidelight required for simulation through PKMR-211, the ship selected as the model, error correction was performed using the camera error value as a variable in the feedback control system. In the experiment, the degree of rolling of the ship was expressed as variables 𝛼 and 𝛽, expressed in angles, and the angle change according to distance was compared. When comparing the error before and after applying the kalman filter in sea state 4, it decreased from +1.5556° to -1.1544° in red light regardless of distance, and the same result was confirmed in green light. Through this, calculations were performed considering the movement of the ship according to the maritime environment, and the future maneuverability of the ship was presented after error correction.

본 논문에서는 해상상태 환경에 따른 카메라 센서 인식의 오차를 제어하고 보정 하기 위해 칼만 필터를 적용한 방법을 제시한다. 먼저, 함정의 제원을 기술하여 롤링에 따른 오차 정도를 측정하였다. 모델로 선택한 함정인 PKMR-211을 통해 시뮬레이션에 필요한 수면으로부터 현등까지 거리를 제시한 후, 피드백 제어 시스템에서 카메라 오차값을 변수로 삼아 오차 보정을 수행하였다. 실험에서는 함정의 롤링 정도를 각도로 나타내는 변수인 𝛼, 𝛽로 표기하고, 거리에 따른 각도 변화를 비교하였다. 해상 상태 3의 상황에서 칼만 필터를 적용하기 전과 후의 오차를 비교했을 때, 거리와 무관하게 적색등에서는 +1.5556°에서 -1.1544°까지 줄어들었고, 녹색등에서도 동일한 결과를 확인하였다. 이를 통해, 해상 환경에 따른 함정의 움직임을 고려하여 계산을 수행하고, 오차 보정 이후 함정의 향후 기동 안정성을 제시하였다.

Keywords

Acknowledgement

This Paper was supported by Research Fund of Republic of Korea Naval Academy in 2024.

References

  1. Boo Sung Youn, "A Study on the Conceptual Design of an Unmanned Surface Vehicle(USV) for the Korean Navy", Journal of the KIMST, Vol.7, No.3, pp. 59-60, 2004 
  2. Soo Ho Choi, "A Study on the Inter-Korean Conflict Over the Northern Limit Line", Law Journal, Vol 32, pp. 567-569, 2010 
  3. Ha Seok Wun, Quiroz Paul, Yong Ho Moon "Improvement of UAV Attitude Information Estimation Performance Using Image Processing and Kalman Filter", Journal of Convergence for Information Technology, Vol.8, No.6, pp. 136-141, 2017 Ha Seok Wun, Quiroz Paul, Yong Ho Moon 
  4. Bokyung Seo, Janghee Lee, Suk I.Yoo, "Robust Hand Tracking Using Kalman Filter and Feature Point", Korean Institute of Information Scientists and Engineers, Vol 37. No.1, pp 520, 2010 
  5. Ministry of Oceans and Fisheries(Maritime Safety Police Division), "Maritime Safety Act", 2023. 1. 5, https://www.law.go.kr/%EB%B2%95%EB%A0%B9/%ED%95%B4%EC%82%AC%EC%95%88%EC%A0%84%EB%B2%95/%2818702,20220104%29 
  6. HJ Shipbuilding & Construction, https://www.hjsc.co.kr/ship/lph_highspeed.asp, 2024. 5.12. 
  7. Edward V. Lewis, "Principles of Naval Architecture (Second Revision): Volume III . Motions in Waves and Controllability", Jersey City, NJ, The Society of Naval Architects and Marine Engineers, pp. 129-176, 1989. 
  8. Kwang-Hoon Lee, Dong-Wook Kim, Yong-Moo Kwon, Eun-Young Chang, Sung-Kyu Kim "Analysis on the cause inducing an uncorrected disparity and distorted depth information by the image distance in stereo camera system", The Journal of the KICS, Vol.34, No.11b, pp. 1321, 2009 
  9. Nam Hyeok Kim, Chi Ho Park, Chung Hee Lee, Young Chul Lim, Jong Hwan Kim, Method for estimating location based on object recognition using kalman filter, Patent Number 10-1364047, 2012.11. 5., 2014. 2.11.