Answers are grounded in this paper and cite their sources.
In this paper, we present a multi-modal control framework to provide compliant behaviors on industrial robots, even when the robots are position commanded. This is obtained by fusing multi-modal sensor signals from robot skin with different control approaches. These compliant behaviors allow to teach robots safely. The presented framework is able to bridge kinesthetically demonstrated activities with low-level robot commands using a state-of-the-art teaching by demonstration method based on a semantic engine. We validate our framework in a real wheeled robot for an industrial scenario, where our presented framework enables a stiff robotic system to be compliant, flexible, and adaptable to different working conditions, e.g. different end-effectors with multiple command interfaces (position/velocity and torque interfaces).
No takes yet. Share an insight, caveat, or question.
Dean‐Leon et al. (2019) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: