Adaptive Impedance Control for Physical Human-Robot Interaction with a Tethered UAV
OPEN ACCESS
Loading...
Author / Producer
Date
2023-05
Publication Type
Master Thesis
ETH Bibliography
yes
Citations
Altmetric
OPEN ACCESS
Data
Rights / License
Abstract
Research into Aerial Physical Interaction has created the opportunity for aerial robots to physically interact with humans. This thesis explores such Physical Aerial Human-Robot Interaction through a system in which a human is physically connected to an aerial vehicle via a tether. This tethered interaction is exploited as a force-based communication channel which guides the human along a pre-defined path. Prior work has established methods for safe human guidance; however, this guidance is purely reactive, and the force-based communication channel is slow to converge to a desired force. Further, while past work has shown that incorporating human odometry into the guidance controller improves path following accuracy, no method has been implemented to acquire this odometry in real-time, onboard the UAV. This thesis improves the path-guidance framework by addressing these issues. First, modern path-following methods are used to guide the human which accounts for both path geometry and human movement. Second, a method of remapping the guidance law to polar coordinates is proposed to enable closed-loop control of the force orientation. Third, a method of real-time human odometry estimation is implemented onboard the aircraft. These methods are validated in simulation before being further evaluated in real-world flight tests.
The final method is shown to enable improve the accuracy of guidance and force-based communication during all phases of testing while maintaining human comfort and safety. These contributions enable more useful human path guidance through tethered interaction with an unmanned aerial vehicle.
Permanent link
Publication status
published
External links
Editor
Contributors
Book title
Journal / series
Volume
Pages / Article No.
Publisher
ETH Zurich
Event
Edition / version
Methods
Software
Geographic location
Date collected
Date created
Subject
Robotics; Control systems; physical human-robot interaction (pHRI); Aerial robots
Organisational unit
03737 - Siegwart, Roland Y. / Siegwart, Roland Y.