Coordinated control of two robotic manipulators for physical interactions with astronauts
Year: 2010
Type: dissertation
Abstract: The following thesis presents the complete development, simulation, and experimental work towards the coordinated control of a dual-manipulator system for physical human-robot interactions. The project aims at pushing robot autonomy towards new boundaries. In the context of space exploration, the potential gain of increased autonomy is enormous and has been a major area of research. Enabling cooperation between an astronaut and a robot has the potential to increase the overall productivity of extra-vehicular activities without the hazard and cost of an additional astronaut. The concept of physical human-robot interaction is motivated by a simple, reliable and robust language between humans and robots, id est, the physical world. In this work, the scope of the problem is limited to the control of two robotic manipulators to manipulate a single object while allowing human input via external forces as part of mission objectives. The robot used here is the WorkPartner, a centaur-like robot whose upper body consists of a two degrees-of-freedom torso and two manipulators with ve degrees-of-freedom each. This thesis presents revisited solutions to the underlying problems of dynamics compensation and external force estimation, integrated in a modern implementation using the virtual model control concept, kinetostatic transmission elements and a behaviour-based strategy. Classic solutions are reviewed and a control strategy is adopted which incorporates widely applicable solutions, and thus, forms a modern basis for further developments. Some simple scenarios are validated for basic stability and performance through simulations, rst with a simple pendulum system and then, with the WorkPartner 's upper-body. Finally, experimental results further reinforces the validity and performance arguments for the proposed algorithms and control architecture.
Author: Sven Mikael Persson
Language: English
Subfield: Mechanical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2887742018
