|
|
基于深度视觉的多旋翼无人机障碍物检测研究
|
Abstract:
随着多旋翼无人机的智能化发展，在无人机执行任务过程中，无论是对动态还是静态障碍物都要进行感知和规避，而且感知系统作为无人机自主飞行过程中的第一步，也是极其重要的部分。针对这个问题，本文主要研究了无人机对动、静态障碍物的检测。首先，使用目标检测算法实时地检测图像中障碍物，并确定其在像素级别上的位置，结合立体匹配算法估计障碍物的深度信息，从而得到障碍物的三维坐标；实现了基于深度视觉的障碍物检测。本文的研究内容可以为无人机及其它机器人在复杂、动态的环境中安全运行提供重要的方法论指导，具有重要的工程实践意义。
With the intelligent development of multi-rotor UAV, both dynamic and static obstacles need to be perceived and avoided during the task execution process. Moreover, the perception system, as the first step in the autonomous flight process of unmanned aerial vehicles, is also an extremely important part. In response to this issue, this article mainly studies the detection of dynamic and static obstacles by drones. Firstly, real time detection of obstacles in images using object detection algorithms and determining their position at the pixel level, combined with stereo matching algorithms to estimate the depth information of obstacles, thus obtaining the three-dimensional coordinates of obstacles; Implemented obstacle detection based on depth vision. The research content of this article can provide important methodological guidance for the safe operation of drones and other robots in complex and dynamic environments, and has important engineering practical significance.
| [1] | Ren, S., He, K., Girshick, R., et al. (2017) Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis & Machine Intelligence, 39, 1137-1149. https://doi.org/10.1109/TPAMI.2016.2577031 |
| [2] | Redmon, J., Divvala, S., Girshick, R., et al. (2016) You Only Look Once: Unified, Real-Time Object Detection. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 27-30 June 2016, 779-788. https://doi.org/10.1109/CVPR.2016.91 |
| [3] | 张志佳, 范莹莹, 邵一鸣, 等. 基于改进YOLO v3模型的多类交通标识检测[J]. 沈阳工业大学学报, 2023, 45(1): 66-70. |
| [4] | Redmon, J. and Farhadi, A. (2018) YOLOv3: An Incremental Improvement. arXiv: 1804.02767. https://doi.org/10.48550/arXiv.1804.02767 |
| [5] | 王卫星, 刘泽乾, 高鹏, 等. 基于改进YOLO v4的荔枝病虫害检测模型[J]. 农业机械学报, 2023, 54(5): 227-235. |
| [6] | Bochkovskiy, A., Wang, C.Y. and Liao, H. (2020) YOLOv4: Optimal Speed and Accuracy of Object Detection. arXiv: 2004.10934. https://doi.org/10.48550/arXiv.2004.10934 |
| [7] | Wang, L., Zhao, J., Xia, X., et al. (2015) Bridge Crack Measurement System Based on Binocular Stereo Vision Technology. Journal of Computer Applications, 35, 901-904. |
| [8] | Nedevschi, S., Danescu, R., Frentiu, D., et al. (2004) High Accuracy Stereo Vision System for Far Distance Obstacle Detection. IEEE Intelligent Vehicles Symposium, Parma, 14-17 June 2004, 292-297. https://doi.org/10.1109/IVS.2004.1336397 |
| [9] | 张凤静, 周建江, 夏伟杰. 基于双目立体视觉的汽车安全车距测量方法[J]. 智能系统学报, 2011, 6(1): 79-84. |
| [10] | 马智. 基于双目视觉的汽车油箱盖识别与检测技术研究[D]: [硕士学位论文]. 长春: 吉林大学, 2022. |
| [11] | Deng, J., Jun, D., Xiaojing, X., et al. (2020) A Review of Research on Object Detection Based on Deep Learning. Journal of Physics: Conference Series, 1684, Article 012028. https://doi.org/10.1088/1742-6596/1684/1/012028 |
| [12] | 涂铭, 金智勇. 深度学习与目标检测工具、原理与算法[M]. 北京: 机械工业出版社, 2021: 131-147. |
| [13] | 李德鑫. 基于YOLOv5s的河道漂浮垃圾检测研究与应用[D]: [硕士学位论文]. 徐州: 中国矿业大学, 2021. |
| [14] | 张展. 基于双目视觉的三维重建关键技术研究[D]: [博士学位论文]. 北京: 中国科学院大学(中国科学院沈阳计算技术研究所), 2019. |
| [15] | Xing, M., Xun, S., Zhou, M., et al. (2011) On Building an Accurate Stereo Matching System on Graphics Hardware. 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops), Barcelona, 6-13 November 2011, 467-474. https://doi.org/10.1109/ICCVW.2011.6130280 |
