TY - GEN
T1 - Packages delivery based on marker detection for UAVs
AU - Feng, Kai
AU - Li, Weixing
AU - Ge, Shengyang
AU - Pan, Feng
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/8
Y1 - 2020/8
N2 - Intelligent UAVs have great application prospects in the express delivery industry. However, efficient environmental awareness and self-control capabilities are must-have for UAVs. In order to solve this, this paper design an efficient delivery system combined with marker detection, which improve the delivery and landing accuracy. In this system, the embedded platform as upper-computer carries out the task of delivery marker detection and trajectory planning, the underlying flight controller completes the UAV attitude control, motion control and data interaction through communication protocol. Besides, the embedded platform simultaneously accomplishes the control commands to UAVs based on Robot Operating System (ROS) distributed architecture, which can be quickly transplanted and deployed. Aiming at the system's visual perception problem, this paper proposes a color detection algorithm based on color channel ratio to detect the markers' color quickly. The experiment used different colors of "mailbox" as the receiving device on the ground to guide UAV to package delivery. The UAVs carry the packages and search for the mailbox independently, and finally control the delivery servo to release package. After the delivery completed, the UAVs return to the take-off location and start the visual landing in conjunction with the detection of the landmark. Experimental results show that the proposed package delivery system can perform tasks quickly and efficiently.
AB - Intelligent UAVs have great application prospects in the express delivery industry. However, efficient environmental awareness and self-control capabilities are must-have for UAVs. In order to solve this, this paper design an efficient delivery system combined with marker detection, which improve the delivery and landing accuracy. In this system, the embedded platform as upper-computer carries out the task of delivery marker detection and trajectory planning, the underlying flight controller completes the UAV attitude control, motion control and data interaction through communication protocol. Besides, the embedded platform simultaneously accomplishes the control commands to UAVs based on Robot Operating System (ROS) distributed architecture, which can be quickly transplanted and deployed. Aiming at the system's visual perception problem, this paper proposes a color detection algorithm based on color channel ratio to detect the markers' color quickly. The experiment used different colors of "mailbox" as the receiving device on the ground to guide UAV to package delivery. The UAVs carry the packages and search for the mailbox independently, and finally control the delivery servo to release package. After the delivery completed, the UAVs return to the take-off location and start the visual landing in conjunction with the detection of the landmark. Experimental results show that the proposed package delivery system can perform tasks quickly and efficiently.
KW - Object Detection
KW - Packages delivery
KW - Robot Operating System (ROS)
KW - UAVs
UR - http://www.scopus.com/inward/record.url?scp=85091592242&partnerID=8YFLogxK
U2 - 10.1109/CCDC49329.2020.9164677
DO - 10.1109/CCDC49329.2020.9164677
M3 - Conference contribution
AN - SCOPUS:85091592242
T3 - Proceedings of the 32nd Chinese Control and Decision Conference, CCDC 2020
SP - 2094
EP - 2099
BT - Proceedings of the 32nd Chinese Control and Decision Conference, CCDC 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 32nd Chinese Control and Decision Conference, CCDC 2020
Y2 - 22 August 2020 through 24 August 2020
ER -