Shared Autonomy via Variable Impedance Control and Virtual Potential Fields for Encoding Human Demonstration

TL;DR

The paper tackles encoding complex manipulation tasks from human demonstrations and enabling safe, cooperative execution in shared autonomy settings. It introduces a time-invariant motion generator based on virtual potential fields to encode trajectories and wrench profiles with obstacle avoidance, and couples it with a dynamic authority arbitration strategy. The variable impedance control and force-control scheme use an energy-tank based passivity guarantee, with the human authority parameter $\alpha_h(t)$ driven by interaction forces to smoothly switch control between human and robot. Validation on a Franka Emika Research 3 robot and simulations demonstrates accurate force production, responsive adaptation to perturbations, and the ability to reproduce complex, closed-loop trajectories while avoiding obstacles.

Abstract

This article introduces a framework for complex human-robot collaboration tasks, such as the co-manufacturing of furniture. For these tasks, it is essential to encode tasks from human demonstration and reproduce these skills in a compliant and safe manner. Therefore, two key components are addressed in this work: motion generation and shared autonomy. We propose a motion generator based on a time-invariant potential field, capable of encoding wrench profiles, complex and closed-loop trajectories, and additionally incorporates obstacle avoidance. Additionally, the paper addresses shared autonomy (SA) which enables synergetic collaboration between human operators and robots by dynamically allocating authority. Variable impedance control (VIC) and force control are employed, where impedance and wrench are adapted based on the human-robot autonomy factor derived from interaction forces. System passivity is ensured by an energy-tank based task passivation strategy. The framework's efficacy is validated through simulations and an experimental study employing a Franka Emika Research 3 robot. More information can be found on the project website https://shailjadav.github.io/SALADS/

Figures (9)

• Figure 1: Experimental setups illustrating shared autonomy in human-robot collaboration for furniture co-manufacturing tasks (left) and button-pressing tasks (right).
• Figure 2: Overview of proposed framework consisting of a motion generator, variable impedance control, and an authority arbitrator.
• Figure 3: Parameter influence on $\hat{\alpha}_{\text{h}}$ calculation with $b = 2$ setting the deadband for $\hat{\alpha}_{\text{h}}$, and $a$ setting the slope of the $\hat{\alpha}_{\text{h}}$ transition to the maximum of 1.
• Figure 4: Comparison of the Proposed Algorithm with SEDS and LPV-DS on the LASA handwriting dataset.
• Figure 5: Demonstration of obstacle avoidance using the proposed approach, following the desired guidance direction.