Impedance control of redundant manipulators for human-robot interaction
Problem to solve
- Developing impedance control techniques for redundant manipulators that enable human-like robot operations and fosters symbiotic human-robot interaction
Figure 1. Application of impedance control
Technical challenges
- Predicting human intent or evaluating human responsiveness in real-time
- Achieving hybrid control of force and position during contact with a human or object
- Leveraging the null space of the redundancy in redundant manipulators to endow human-like behavior and enhance task performance
Specific research topics
- Developing a reinforcement learning (RL)-based variable impedance control approach for enabling symbiotic human-robot interaction
- Designing an RL framework for optimizing the null-space behavior of redundant manipulators to achieve human-like behavior
Experimental setup
Figure 2. Experimental setup
Control and AI techniques used in research
- Impedance control techniques
- Reinforcement learning (RL) algorithms
- Numerical optimization algorithms
Expected results
- The developed impedance control techniques are expected to showcase the potential for robot manipulators to be utilized in a wider range of applications within future mobility scenarios.
Relevant references
- J. H. Kim, K. Choi, and I. G. Jang*, “Model predictive control-based time-optimal trajectory planning of the distributed actuation mechanism augmented by the maximum performance evaluation,” Applied Sciences, vol. 11, No. 16, pp. 7513, 2021.
Relevant research projects or grants
- 초실감 보병 전투 기술 훈련 기술(국방기술진흥연구소, 2022-2027)