Experimental Evaluation of a Stiffness-Fault-Tolerant Control Strategy on an Elastic Actuator for Wearable Robotics
Rodrigo J. Velasco-Guillen, Victor Grosu, Bram Vanderborght, Josep M. Font-Llagunes, Philipp Beckerle
- Year
- 2022
- Citations
- 3
Abstract
Elastic actuator design is primarily motivated by safety in human-robot interaction and energy efficiency. However, the increased complexity when compared to direct drives, makes the system prone to technical faults in elastic and kinematic elements. On the other hand, Variable Stiffness Actuators (VSA) often introduce non-linear torque-deformation relationships. We propose a stiffness-fault-tolerant control strategy for elastic actuators with non-linear compliant characteristics, capable of adapting to stiffness changes. The control strategy is demonstrated experimentally in a MACCEPA-based elastic actuator designed to power a knee exoskeleton. We develop the general model-based control scheme based on impedance control and adapt it to the experimental actuation system. Experiential evaluations under oscillatory and sigmoid trajectories are considered to analyze performance. Results demonstrate that the control system is capable of accurately tracking a reference trajectory under stiffness variations and interaction disturbance.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002