DOI QR코드

DOI QR Code

드론과 지상로봇 간의 협업을 위한 광학흐름 기반 마커 추적방법

Optical Flow-Based Marker Tracking Algorithm for Collaboration Between Drone and Ground Vehicle

  • 백종환 (한경대학교 전기전자제어공학과) ;
  • 김상훈 (한경대학교 전기전자제어공학과)
  • 투고 : 2017.12.21
  • 심사 : 2018.01.28
  • 발행 : 2018.03.31

초록

본 논문에서는 드론과 지상 로봇 간 효과적인 협업을 위하여 광학 흐름 기술 기반의 특징점 추적 알고리즘을 제안하였다. 드론의 비행 중 빠른 움직임에 의하여 많은 문제점이 발생하여 지상물체를 성공적으로 인식하기 위해 직관적이면서도 식별자를 가지고 있는 마커를 사용했다. 특징점 추출이 우수한 FAST알고리즘과 움직임 감지가 우수한 루카스-카나데 광학흐름 알고리즘의 장점들을 혼합하여 기존 특징점-특징량 기반 객체 추적 방법보다 개선된 속도의 실험결과를 보여준다. 또한 제안한 마커의 검출방법에 적절한 이진화 방법을 제안하여 주어진 마커에서의 검출 정확도를 개선하였으며, 추적속도는 유사한 환경의 기존연구보다 40% 이상 개선됨을 확인하였다. 또한 비행드론의 경량화와 속도개선에 문제가 없도록 최소형 고성능의 임베디드 환경을 선택하였으며, 제한된 개발환경에서도 물체검출과 추적 등 복잡한 연산이 가능하도록 동작환경에 대하여 연구하였다. 향후에는 다른 환경에서 빠르게 움직이는 두 로봇 간의 협업의 정확도를 향상시키기 위해 지능적 비전기능에 대해 추가할 예정이다.

In this paper, optical flow based keypoint detection and tracking technique is proposed for the collaboration between flying drone with vision system and ground robots. There are many challenging problems in target detection research using moving vision system, so we combined the improved FAST algorithm and Lucas-Kanade method for adopting the better techniques in each feature detection and optical flow motion tracking, which results in 40% higher in processing speed than previous works. Also, proposed image binarization method which is appropriate for the given marker helped to improve the marker detection accuracy. We also studied how to optimize the embedded system which is operating complex computations for intelligent functions in a very limited resources while maintaining the drone's present weight and moving speed. In a future works, we are aiming to develop collaborating smarter robots by using the techniques of learning and recognizing targets even in a complex background.

키워드

참고문헌

  1. H. Oh et al., "Trends on policy and technology of S&T and ICT," in Korea Institute of Science and Technology Evaluation and Planning, No.66, pp.24-37, 2016.
  2. F. Koohifar, A. Kumbhar and I. Guvenc, "Receding Horizon Multi-UAV Cooperative Tracking of Moving RF Source," in IEEE Communications Letters., Vol.21, No.6, pp.1433-1436, 2017. https://doi.org/10.1109/LCOMM.2016.2603977
  3. K. A. Ghamry, Y. Dong, M. A. Kamel, and Y. Zhang, "Real-time autonomous take-off, tracking and landing of UAV on a moving UGV platform," in Control and Automation, 24th Mediterranean Conference (MED), pp.1236-1241, 2016.
  4. T. Mcgee, Z. Kim, and X. Xiao, "Vision-based road-following using a small autonomous aircraft," in Proceedings of IEEE Conf. Aerospace, Montana, pp.3006-3015, 2004.
  5. J. Beck et al., "Sensor Fusion Docking System of Drone and Ground Vehicles Using Image Object Detection," in Software and Data Engineering, KIPS Transactions, Vol.6, No.4, pp. 217-222, 2016.
  6. B. Dzodzo, L. Han, X. Chen, H. Qian, and Y. Xu, "Realtime 2D code based localization for indoor robot navigation." in IEEE International Conference on Robotics and Biomimetics (ROBIO), Shenzhen, pp.486-492. 2013.
  7. H. Kobayashi, "A personal localization system using selfcontained 2D barcode landmarks" IEEE 24th International Symposium on Industrial Electronics (ISIE), Buzios, pp.882-887. 2015.
  8. S. Lee, "Embedded vision based SLAM for consumer robots," Ph. D. dissertation, Sungkyunkwan University, Seoul, Korea, 2012.
  9. C. Harris and M. Stephens, "A combined corner and edge detector," in Alvey Vision Conference, Vol.15, No.50, pp. 10-5244. 1988.
  10. N. Luo, Q Sun, Q. Chen, Z. Ji, and D. Xia. "A novel tracking algorithm via feature points matching," in PloS One, Vol.10, No.1, 2015.
  11. E. Tola, V. Lepetit and P. Fua, "Daisy: An efficient dense descriptor applied to wide-baseline stereo," in IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.32, No.6, pp.815-830. 2010. https://doi.org/10.1109/TPAMI.2009.77
  12. B. Okumura, M. Kanbara, and N. Yokoya, "Augmented reality based on estimation of defocusing and motion blurring from captured images," in Mixed and Augmented Reality (ISMAR), IEEE/ACM International Symposium, pp.219-225. 2006.
  13. E. Rosten and T. Drummond, "Fusing points and lines for high performance tracking," in Tenth IEEE International Conference (ICCV), Vol.2, pp.1508-1515. 2005.
  14. D. G. Lowe, "Distinctive image features from scale-invariant keypoints," in International Journal of Computer Vision, Vol. 60, No.2, pp.91-110. 2004. https://doi.org/10.1023/B:VISI.0000029664.99615.94
  15. E. Rosten and T. Drummond. "Machine learning for highspeed corner detection," in Computer Vision (ECCV), pp. 430-443, 2006.
  16. B. D. Lucas and T. Kanade, "An iterative image registration technique with an application to stereo vision," pp.674-679, 1981.