Learning Personalized Safety Interventions for Haptic Human-Robot Shared Control

2026-07-21

AI Summary

This work addresses the limitation of existing haptic guidance systems, which rely on predefined intervention policies and struggle to accommodate users’ personalized safety preferences. To overcome this, the paper proposes a novel framework that learns personalized haptic safety intervention policies directly from sparse user demonstrations, eliminating the need for manual parameter tuning. The approach integrates a differentiable Control Barrier Function (CBF) optimization layer with a Learning-from-Haptics (LfH) mechanism, enabling intuitive and adaptive shared control between human and robot. Both simulation and real-world hardware experiments demonstrate that the method significantly reduces the mismatch between haptic feedback and user expectations using only a small number of user inputs, thereby substantially improving the performance of personalized safety interventions.

Abstract

Haptic feedback provides an implicit channel for communicating safety intentions during human-robot shared control. Existing haptic guidance systems typically employ predefined intervention strategies that cannot accommodate the diverse safety preferences of individual users or application scenarios. To address this limitation, we propose a Learning from Haptics (LfH) framework that learns user-preferred safety interventions from sparse demonstrations, eliminating the need for manual trial-and-error design. Our framework is built on a differentiable Control Barrier Function (CBF)-based optimization layer that automatically adjusts the underlying safety parameters to match the demonstrated haptic responses. Instead of tuning controller parameters directly, users teach the system how they expect it to intervene during teleoperation. The resulting haptic guidance reflects the demonstrated intervention preferences while preserving the intuitive interaction of haptic shared control. Simulation and hardware experiments demonstrate that the proposed framework can learn personalized safety interventions from sparse user input and reduce the mismatch between the generated haptic feedback and the demonstrated preferences.

Problem

• personalized safety interventions
• haptic human-robot shared control
• safety preferences
• haptic feedback
• user adaptation

Innovation

• Learning from Haptics
• Control Barrier Function
• personalized safety intervention
• haptic shared control
• differentiable optimization