计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (22): 231-238.DOI: 10.3778/j.issn.1002-8331.1807-0271

• 工程与应用 • 上一篇    下一篇

多标记室内小型无人机定位与姿态估计方法

周克旻,周蓉,滕婧,陈亦奇   

  1. 华北电力大学 控制与计算机工程学院,北京 102200
  • 出版日期:2019-11-15 发布日期:2019-11-13

Method of Location and Attitude Estimation for Small Indoor UAV with Multiple Markers

ZHOU Kemin,  ZHOU Rong, TENG Jing, CHEN Yiqi   

  1. School of Control and Computer Engineering, North China Electric Power University, Beijing 102200, China
  • Online:2019-11-15 Published:2019-11-13

摘要: 随着无人机(Unmanned Aerial Vehicle,UAV)小型化、轻便化的发展,因其价格低廉,以及在娱乐和服务领域的广泛使用的特点,使得如何实现一个便捷且易实现的自主飞行跟踪系统成为关注点。由于无人机在室内GPS信号弱,使得跟踪与姿态获取成为进一步室内无人机自主控制的重点与难点。与动辄几十万美元搭建的无人机跟踪系统相比,采用低成本单目摄像机的无人机跟踪系统具有更高的科研价值和更广泛的应用前景。针对目前流行的基于增强现实(Augmented Reality,AR)技术的ArUco标记算法和颜色空间域标记算法,设计了一种多标记的无人机跟踪系统。在无人机目标跟踪过程中比较两种方法,验证了两种方法非接触式深度传感器无人机跟踪和姿态估计的效果,并比较了两种方法对空间亮度与空间颜色复杂度的鲁棒性,以及不同跟踪距离下视频中无人机检出率与跟踪精度。实验结果表明,基于深度摄像机获得的无人机位置和姿态数据,无人机可以进行自主的PID控制飞行,且AR标记在复杂环境下无人机的检出率、跟踪实时性、姿态估计精度以及鲁棒性都优于颜色标记,为之后室内无人机在非接触式传感的控制、路径规划、自主规避等进一步实验研究提供了无人机的位置和姿态数据。

关键词: ArUco, 室内无人机, 跟踪, 姿态估计

Abstract: With the development of Unmanned Aerial Vehicle(UAV) of miniaturization and portability, and because of the traits of low price and being widely used in entertainment and service fields, it becomes a focus how to create an autonomous flight tracking system which is convenient and easy to implement. Because of the weak GPS signal in the indoor environment, indoor tracking and gesture obtaining of UAV becomes much difficult and hard tasks of the autonomous control of indoor UAV. Compared with frequently hundreds of thousands of dollars of UAV tracking system, using a low-cost monocular camera UAV tracking system has higher research value and wider application prospect. Aiming at the popular ArUco labeling algorithm based on AR technology and the color space domain labeling algorithm, this paper proposes a method of location and attitude estimation for small indoor UAV with multiple markers. Then, this paper compares and verifies the effectiveness of the two methods for UAV tracking and attitude estimation based on non-contact depth sensor. At the same time, this paper compares the robustness of the two methods of spatial brightness and spatial color complexity, as well as the detection rate and tracking precision of UAV in video at different tracking distances. The experimental results show that the detection rate, real-time tracking, attitude estimation accuracy and robustness of AR tag are better than color tag in complex environment. The position and attitude data of indoor UAV are provided for further experimental research on non-contact sensor control, path planning and autonomous avoidance.

Key words: ArUco, indoor UAV, tracking, attitude estimation