Journal of Computer Applications ›› 2019, Vol. 39 ›› Issue (4): 1001-1005.DOI: 10.11772/j.issn.1001-9081.2018091952

Previous Articles     Next Articles

Monocular vision obstacle avoidance method for quadcopter based on deep learning

ZHANG Wuyang1,2, ZHANG Wei1, SONG Fang1, LONG Lin2   

  1. 1. Laboratory of Intelligent Control and Robotics, Shanghai University of Engineering Science, Shanghai 201620, China;
    2. College of Mechanical and Automobile Engineering, Shanghai University of Engineering Science, Shanghai 201620, China
  • Received:2018-09-20 Revised:2018-11-15 Online:2019-04-10 Published:2019-04-10
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (51505273).


张午阳1,2, 章伟1, 宋芳1, 龙林2   

  1. 1. 上海工程技术大学 机器人智能控制实验室, 上海 201620;
    2. 上海工程技术大学 机械与汽车工程学院, 上海 201620
  • 通讯作者: 宋芳
  • 作者简介:张午阳(1994-),男,安徽淮北人,硕士研究生,主要研究方向:无人机避障、计算机视觉;章伟(1977-),男,安徽桐城人,副教授,博士,主要研究方向:非线性控制与观测器、鲁棒控制、无人机控制;宋芳(1980-),女,黑龙江大庆人,副教授,博士,主要研究方向:机器人、运动控制;龙林(1976-),男,安徽桐城人,硕士,主要研究方向:智能化控制、物联网。
  • 基金资助:

Abstract: A monocular vision obstacle avoidance method for quadrotor based on deep learning was proposed to help quadrotors to avoid obstacles. Firstly, the position of object in the image was obtained by object detection, and by calculating the height of the object box in the image, the distance between quadcopter and obstacle was estimated. Then, whether performing obstacle avoidance was determined by synergetic computer. Finally, experiments were conducted on a flight test platform based on Pixhawk flight control board. The results show that the proposed method can be applied to quadcoptor obstacle avoidance with low speed. Compared with traditional active sensor methods, the proposed method greatly reduces the occupied volume with only one monocular camera as sensor. This method is robust and can identify people with different postures as obstacles.

Key words: deep learning, object detection, monocular vision, quadcopter obstacle avoidance

摘要: 针对无人机避障问题,提出一种基于深度学习的四旋翼无人机单目视觉避障方法。首先通过目标检测框选出目标在图像中的位置,并通过计算目标选框上下边距的长度,以此来估量出障碍物到无人机之间的距离;然后通过协同计算机判断是否执行避障动作;最后使用基于Pixhawk搭建的飞行实验平台进行实验。实验结果表明,该方法可用于无人机低速飞行条件下避障。该方法所用到的传感器只有一块单目摄像头,而且相对于传统的主动式传感器避障方法,所占用无人机的体积大幅减小。该方法鲁棒性较好,能够准确识别不同姿态的人,实现对人避障。

关键词: 深度学习, 目标检测, 单目视觉, 无人机避障

CLC Number: