Robot assistance primitives with force-field guidance for shared task collaboration

Abstract

This paper proposes a novel framework for human-robot collaboration (HRC) that addresses the critical need for robots to effectively collaborate with humans on shared tasks within unstructured and dynamic environments. While prior research focused on safety-related aspects, such as collision avoidance in shared workspaces, the task-oriented aspects of human-robot collaboration remain largely underexplored. To address this gap, our framework introduces Robot Assistance Primitives (RAPs). These low-level robot actions integrate both safety and task-related behaviours, enabling the robot to function as a collaborative "third hand", and provide assistance across the full spectrum of both physical and contactless interactions. A key component of our approach is an extension of impedance control with virtual force fields, which unifies task-related interactions and safety-related aspects within a single control scheme. The framework leverages a state-of-the-art visual perception pipeline that constructs and tracks real-time 3D digital representations of the workspace and the human operator. Additionally, an Augmented Reality Head-Mounted Display (AR-HMD) facilitates multimodal task programming through user gaze, gestures, and speech, as well as providing visual feedback to foster trust during interactions. We validate the feasibility of the proposed framework and conduct a user study to further evaluate user interactions in a collaborative soldering and assembly task. This research not only addresses limitations of current HRC frameworks but also paves the way for exploring novel collaborative applications.