정보처리학회논문지:소프트웨어 및 데이터공학 (KIPS Transactions on Software and Data Engineering)

  • 제12권5호
  • /
  • Pages.237-242
  • /
  • 2023
  • /
  • 2287-5905(pISSN)
  • /
  • 2734-0503(eISSN)
한국정보처리학회 (Korea Information Processing Society)
권호 표지
DOI QR코드

DOI QR Code

딥러닝을 이용한 객체검출과 비평탄 지형 보행을 위한 4족 로봇

Quadruped Robot for Walking on the Uneven Terrain and Object Detection using Deep Learning

  • 박명숙 (한경대학교 전기전자제어공학과) ;
  • 한성민 (한경대학교 전기전자제어공학과) ;
  • 김상훈 (한경대학교 ICT로봇기계공학부)
  • 투고 : 2022.12.21
  • 심사 : 2023.02.03
  • 발행 : 2023.05.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

고성능의 보행 로봇에 관한 연구가 활발하게 이루어지고 있으며 4족 보행 로봇은 비평탄 지형에서 이동성과 적응력이 뛰어나 많은 관심을 받고 있지만 높은 비용으로 도입과 활용성에 어려움이 있다. 본 논문에서는 저비용의 4족 로봇에 지능적 기능을 적용하여 활용도를 높이기 위해 임베디드 보드에 IMU와 강화학습을 탑재하여 비평탄 지형 극복능력을 개선하고 카메라와 딥러닝을 이용하여 객체를 자동으로 검출하는 방법을 제시한다. 로봇은 4족 포유류 동물의 다리 형태로 구성되고 각 다리는 3 자유도를 가진다. 설계된 3D 모델로 시뮬레이션 환경에서 복잡한 지형을 학습시키고 실제 로봇에 적용한다. 본 연구방법의 적용을 통해 평탄 지형과 비평탄 지형의 보행 능력에 크게 차이가 나지 않음을 확인하였으며 제한된 실험조건에서 실시간으로 사람 검출을 수행하는 동작을 확인하였다.

Research on high-performance walking robots is being actively conducted, and quadruped walking robots are receiving a lot of attention due to their excellent mobility and adaptability on uneven terrain, but they are difficult to introduce and utilize due to high cost. In this paper, to increase utilization by applying intelligent functions to a low-cost quadruped robot, we present a method of improving uneven terrain overcoming ability by mounting IMU and reinforcement learning on embedded board and automatically detecting objects using camera and deep learning. The robot consists of the legs of a quadruped mammal, and each leg has three degrees of freedom. We train complex terrain in simulation environments with designed 3D model and apply it to real robot. Through the application of this research method, it was confirmed that there was no significant difference in walking ability between flat and non-flat terrain, and the behavior of performing person detection in real time under limited experimental conditions was confirmed.

키워드

과제정보

이 논문은 2020년도 정부(교육부)의 재원으로 한국연구재단 기초연구사업의 지원을 받아 수행된 연구임(No.2020R1F1A1067496).

참고문헌

  1. T. M. Wang, Y. Tao, and H. Liu, "Current researches and future development trend of intelligent robot: A review," International Journal of Automation and Computing, Vol.15, No.5, pp.525-546, 2018. https://doi.org/10.1007/s11633-018-1115-1
  2. K. I. Kim, "ASTI market insight 32: Service robot," Korea Institute of Science and Technology Information, 2022.
  3. P. Biswal and P. K. Mohanty, "Development of quadruped walking robots: A review," Ain Shams Engineering Journal, Vol.12, No.2, pp.2017-2031, 2021. https://doi.org/10.1016/j.asej.2020.11.005
  4. H. Chai et al., "A survey of the development of quadruped robots: Joint configuration, dynamic locomotion control method and mobile manipulation approach," Biomimetic Intelligence and Robotics, Vol.2, No.1, 100029, 2022.
  5. C. Semini, N. G. Tsagarakis, E. Guglielmino, M. Focchi, F. Cannella, and D. G. Caldwell, "Design of HyQ-a hydraulically and electrically actuated quadruped robot," Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering, Vol.225, No.6, pp.831-849, 2011. https://doi.org/10.1177/0954411911401783
  6. M. Hutter et al., "Anymal-toward legged robots for harsh environments," Advanced Robotics, Vol.31, No.17, pp.918- 931, 2017. https://doi.org/10.1080/01691864.2017.1378591
  7. T. M. Wang, Y. Tao, and H. Liu, "MIT cheetah 3: Design and control of a robust, dynamic quadruped robot," In 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp.2245-2252, 2018.
  8. Y. H. Lee and H. R. Choi, "Development of quadruped walking robot AiDIN Series," Robot and Human, Vol.17, No.2, pp.3-10, 2020.
  9. J. S. Cho, J. H. Seo, and S. D. Park, "Development of quadrupedal robot technology for field application: JINPOONG," Robot and Human, Vol.17, No.2, pp.24-31, 2020.
  10. G. H. Jeong and D. G. Kim, "Hyundai E&C, the current status and future of digital transformation in the construction industry," Construction Engineering and Management, Vol.23, No.1, pp.25-31, 2022.
  11. SpotMicroAI [Internet], https://spotmicroai.readthedocs.io/en/latest/ (accessed on 3 February 2023)
  12. M. Rahme, I. Abraham, M. L. Elwin, and T. D. Murphey, "Linear policies are sufficient to enable low-cost quadrupedal robots to traverse rough terrain," In 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp.8469-8476, 2021.
  13. S. M. Han, S. J. Yu, G. Lee, M. S. Pak, and S. H. Kim, "Real-time human detection method based on quadrupedal walking robot," Proceedings of the Korea Information Processing Society Conference, Vol.29, No.2, pp.468-470, 2022.
  14. D. J. Hyun, S. Seok, J. Lee, and S. Kim, "High speed trotrunning: Implementation of a hierarchical controller using proprioceptive impedance control on the MIT Cheetah," The International Journal of Robotics Research, Vol.33, No.11, pp.1417-1445, 2014. https://doi.org/10.1177/0278364914532150
  15. M. A. Sen, V. Bakircioglu and M. Kalyoncu, "Inverse kinematic analysis of a quadruped robot," International journal of Scientific & Technology Research, Vol.6, No.9, pp.285- 289, 2017.
  16. J. Redmon and A. Farhadi, "Yolov3: An incremental improvement," arXiv preprint arXiv:1804.02767, 2018.
  17. A. Bochkovskiy, C. Y. Wang, and H. Y. M. Liao, "Yolov4: Optimal speed and accuracy of object detection," arXiv preprint arXiv:2004.10934, 2020.
  18. W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C. Y. Fu, and A. C. Berg, "Ssd: Single shot multibox detector," In European Conference on Computer Vision, pp.21-37, Springer, Cham, 2016.
  19. C. Y. Wang, A. Bochkovskiy, and H. Y. M. Liao, "YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors," arXiv preprint arXiv:2207.02696, 2022.
  20. M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L. C. Chen, "Mobilenetv2: Inverted residuals and linear bottlenecks," In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.4510-4520, 2018.
  21. T. Y. Lin et al., "Microsoft coco: Common objects in context," In Computer Vision-ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13, Springer International Publishing, pp.740-755, 2014.
  22. H. Vanholder, "Efficient inference with tensorrt," In GPU Technology Conference, Vol.1, 2016.
  23. tensorrt_demos [Internet], https://github.com/jkjung-avt/tensorrt_demos (accessed on 3 February 2023)