Start Date

8-5-2024 11:55 AM

End Date

8-5-2024 11:40 AM

Document Type

Full Paper

Keywords

UGV-UAV collaboration, Leader-follower Navigation, Object Detection, Path Planning

Description

In this study, a proof-of-concept leader-follower relationship between an unmanned aerial vehicle (UAV) and an unmanned ground vehicle (UGV) as a method of navigation is developed and tested. A JetHexa hexapod robot was tasked with following a CoDrone EDU quadrotor with no communication between the two platforms. To implement the UAV-following behavior, a You Only Look Once version 5 (YOLOv5) object detection model was developed and employed as part of a Robotic Operating System (ROS) package. According to the results, in controlled testing environments, the UGV is able to follow the UAV over multiple waypoints, following at distances as close as ten centimeters. With future optimization improvements, this leader-follower dynamic may provide a simple means for offloading environmental traversability analysis and path planning to platforms with more advantageous vantage points.

DOI

https://doi.org/10.5038/VLGL1231

Download

Share

COinS
 
May 8th, 11:55 AM May 8th, 11:40 AM

A UGV-UAV Formation Using Leader-Follower-Based Collaborative Navigation

In this study, a proof-of-concept leader-follower relationship between an unmanned aerial vehicle (UAV) and an unmanned ground vehicle (UGV) as a method of navigation is developed and tested. A JetHexa hexapod robot was tasked with following a CoDrone EDU quadrotor with no communication between the two platforms. To implement the UAV-following behavior, a You Only Look Once version 5 (YOLOv5) object detection model was developed and employed as part of a Robotic Operating System (ROS) package. According to the results, in controlled testing environments, the UGV is able to follow the UAV over multiple waypoints, following at distances as close as ten centimeters. With future optimization improvements, this leader-follower dynamic may provide a simple means for offloading environmental traversability analysis and path planning to platforms with more advantageous vantage points.