딥러닝을 이용한 객체검출과 비평탄 지형 보행을 위한 4족 로봇
Quadruped Robot for Walking on the Uneven Terrain and Object Detection using Deep Learning
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
고성능의 보행 로봇에 관한 연구가 활발하게 이루어지고 있으며 4족 보행 로봇은 비평탄 지형에서 이동성과 적응력이 뛰어나 많은 관심을받고 있지만 높은 비용으로 도입과 활용성에 어려움이 있다. 본 논문에서는 저비용의 4족 로봇에 지능적 기능을 적용하여 활용도를 높이기 위해임베디드 보드에 IMU와 강화학습을 탑재하여 비평탄 지형 극복능력을 개선하고 카메라와 딥러닝을 이용하여 객체를 자동으로 검출하는 방법을제시한다. 로봇은 4족 포유류 동물의 다리 형태로 구성되고 각 다리는 3 자유도를 가진다. 설계된 3D 모델로 시뮬레이션 환경에서 복잡한 지형을학습시키고 실제 로봇에 적용한다. 본 연구방법의 적용을 통해 평탄 지형과 비평탄 지형의 보행 능력에 크게 차이가 나지 않음을 확인하였으며제한된 실험조건에서 실시간으로 사람 검출을 수행하는 동작을 확인하였다.
Abstract
Research on high-performance walking robots is being actively conducted, and quadruped walking robots are receiving a lot of attentiondue to their excellent mobility and adaptability on uneven terrain, but they are difficult to introduce and utilize due to high cost. Inthis paper, to increase utilization by applying intelligent functions to a low-cost quadruped robot, we present a method of improvinguneven terrain overcoming ability by mounting IMU and reinforcement learning on embedded board and automatically detecting objectsusing 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 ofthis research method, it was confirmed that there was no significant difference in walking ability between flat and non-flat terrain, andthe behavior of performing person detection in real time under limited experimental conditions was confirmed.
참고문헌 (23)
- T. M. Wang, "Current researches and future development trend of intelligent robot : A review", International Journal of Automation and Computing, 15(5), pp.525~546 (2018)
- ASTI market insight 32 : Service robot, "K. I. Kim", Korea Institute of Science and Technology Information, ~ (2022)
- P. Biswal, "Development of quadruped walking robots : A review", Ain Shams Engineering Journal, 12(2), pp.2017~2031 (2021)
- H. Chai, "A survey of the development of quadruped robots : Joint configuration, dynamic locomotion control method and mobile manipulation approach", Biomimetic Intelligence and Robotics, 2(1), pp.100029~ (2022)
- C. Semini, "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, 225(6), pp.831~849 (2011)
- M. Hutter, "Anymal-toward legged robots for harsh environments", Advanced Robotics, 31(17), pp.918~931 (2017)
- T. M. Wang, "MIT cheetah 3 : Design and control of a robust, dynamic quadruped robot", 2018 IEEE, 2018, 2245~2252
- Y. H. Lee, "Development of quadruped walking robot AiDIN Series", Robot and Human, 17(2), pp.3~10 (2020)
- J. S. Cho, "Development of quadrupedal robot technology for field application : JINPOONG", Robot and Human, 17(2), pp.24~31 (2020)
- 강현욱, "활주로 건설공사의 위험사건에 따른 손실비용 및 위험예비비 분석", 한국건설관리학회 논문집, 23(3), pp.25~35 (2022), 한국건설관리학회
- "SpotMicroAI"[URL]
- M. Rahme, "Linear policies are sufficient to enable low-cost quadrupedal robots to traverse rough terrain", 2021 IEEE, 2021, 8469~8476
- S. M. Han, "Real-time human detection method based on quadrupedal walking robot", Proceedings of the Korea Information Processing Society Conference, 29(2), pp.468~470 (2022)
- D. J. Hyun, "High speed trotrunning : Implementation of a hierarchical controller using proprioceptive impedance control on the MIT Cheetah", The International Journal of Robotics Research, 33(11), pp.1417~1445 (2014)
- M. A. Sen, "Inverse kinematic analysis of a quadruped robot", International journal of Scientific & Technology Research, 6(9), pp.285~289 (2017)
- J. Redmon, "Yolov3 : An incremental improvement", arXiv preprint arXiv : 1804. 02767, ~ (2018)
- A. Bochkovskiy, "Yolov4 : Optimal speed and accuracy of object detection", arXiv preprint arXiv : 2004. 10934, ~ (2020)
- W. Liu, "Ssd : Single shot multibox detector", European Conference on Computer Vision, pp.21~37 (2016), Springer
- C. Y. Wang, "YOLOv7 : Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors", arXiv preprint arXiv : 2207. 02696, ~ (2022)
- M. Sandler, "Mobilenetv2 : Inverted residuals and linear bottlenecks", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.4510~4520 (2018)
- T. Y. Lin, "Microsoft coco : Common objects in context", Computer Vision–ECCV 2014: 13th European Conference, (Part V 13), pp.740~755 (2014), Springer International Publishing
- H. Vanholder, "Efficient inference with tensorrt", GPU Technology Conference, 1 (2016)
- "tensorrt_demos"[URL]