Teach-Repeat-Replan: A Complete and Robust System for Aggressive Flight in Complex Environments

被引:110
作者
Gao, Fei [1 ]
Wang, Luqi [2 ]
Zhou, Boyu [2 ]
Zhou, Xin [1 ]
Pan, Jie [2 ]
Shen, Shaojie [2 ]
机构
[1] Zhejiang Univ, State Key Lab Ind Control & Technol, Hangzhou 310007, Peoples R China
[2] Hong Kong Univ Sci & Technol, Dept Elect & Comp Engn, Hong Kong, Peoples R China
关键词
Trajectory; Drones; Education; Robustness; Robots; Planning; Optimization; Aerial systems; applications; motion; and path planning; autonomous vehicle navigation; collision avoidance; TRAJECTORY GENERATION;
D O I
10.1109/TRO.2020.2993215
中图分类号
TP24 [机器人技术];
学科分类号
080202 ; 1405 ;
摘要
In this article, we propose a complete and robust system for the aggressive flight of autonomous quadrotors. The proposed system is built upon on the classical teach-and-repeat framework, which is widely adopted in infrastructure inspection, aerial transportation, and search-and-rescue. For these applications, a human's intention is essential for deciding the topological structure of the flight trajectory of the drone. However, poor teaching trajectories and changing environments prevent a simple teach-and-repeat system from being applied flexibly and robustly. In this article, instead of commanding the drone to precisely follow a teaching trajectory, we propose a method to automatically convert a human-piloted trajectory, which can be arbitrarily jerky, to a topologically equivalent one. The generated trajectory is guaranteed to be smooth, safe, and dynamically feasible, with a human preferable aggressiveness. Also, to avoid unmapped or moving obstacles during flights, a fast local perception method and a sliding-windowed replanning method are integrated into our system, to generate safe and dynamically feasible local trajectories onboard. We name our system as teach-repeat-replan. It can capture users' intention of a flight mission, convert an arbitrarily jerky teaching path to a smooth repeating trajectory, and generate safe local replans to avoid unexpected collisions. The proposed planning system is integrated into a complete autonomous quadrotor with global and local perception and localization submodules. Our system is validated by performing aggressive flights in challenging indoor/outdoor environments. We release all components in our quadrotor system as open-source ros packages.
引用
收藏
页码:1526 / 1545
页数:20
相关论文
共 46 条
[1]  
[Anonymous], 2018, ARXIV180309650
[2]   The Quickhull algorithm for convex hulls [J].
Barber, CB ;
Dobkin, DP ;
Huhdanpaa, H .
ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 1996, 22 (04) :469-483
[3]  
Berczi LP, 2016, 2016 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS 2016), P3973, DOI 10.1109/IROS.2016.7759585
[4]  
Blöchliger F, 2018, IEEE INT CONF ROBOT, P3818
[5]  
C. deBoor, 1971, LA4728 LOS AL SCI LA
[6]  
Choset H. M., 2005, Principles of Robot Motion: Theory, Algorithms, and Implementations
[7]   Computing Large Convex Regions of Obstacle-Free Space Through Semidefinite Programming [J].
Deits, Robin ;
Tedrake, Russ .
ALGORITHMIC FOUNDATIONS OF ROBOTICS XI, 2015, 107 :109-124
[8]  
Felzenszwalb P.F., 2012, Theory Comput., V8, P415, DOI [10.4086/toc.2012.v008a019, DOI 10.4086/TOC.2012.V008A019]
[9]   Visual Teach and Repeat for Long-Range Rover Autonomy [J].
Furgale, Paul ;
Barfoot, Timothy D. .
JOURNAL OF FIELD ROBOTICS, 2010, 27 (05) :534-560
[10]   Flying on point clouds: Online trajectory generation and autonomous navigation for quadrotors in cluttered environments [J].
Gao, Fei ;
Wu, William ;
Gao, Wenliang ;
Shen, Shaojie .
JOURNAL OF FIELD ROBOTICS, 2019, 36 (04) :710-733