Variable Admittance Control Method Based on Human-Robot Collaboration State, and Apparatus and Computer Program for Executing the Same

Robot Admittance Control Technology with Real-Time Parameter Variation Based on Collaborative State

This technology distinguishes between intentional and unintentional human movements through frequency analysis of multi-degree-of-freedom force signals and dynamically adjusts admittance model parameters, such as inertia and damping, based on the resulting collaborative state.

Fixed admittance gains create a trade-off between sensitive control and safety, while existing DFT-based observers suffer from slow real-time response speeds, making immediate adjustments difficult.

By using low-pass and high-pass filters for frequency analysis, this technology reduces computational load to increase response speed and enables real-time variable control of parameters based on the collaborative state. It can be applied to collaborative robots and physical human-robot interaction systems, providing control performance that ensures both sensitivity and safety tailored to the work environment.

‍

‍

Key Features:

• Acquiring multi-degree-of-freedom force signals to distinguish between intentional and unintentional human movements
• Obtaining a collaborative state value based on multi-degree-of-freedom force signals using low-pass and high-pass filters
• Adjusting admittance parameters using a reference collaborative state value corresponding to a preset reference frequency
• Applying parameters configured for sensitive control when the collaborative state value is below the reference collaborative state value