Safety Interaction Mechanisms and Robust Control Methods for Human-Robot Collaborative Systems in Dynamic Environments

Abstract

Traditional human-robot collaboration methods primarily rely on rule-based control and physical human-robot interaction, suitable for industrial scenarios with clear rules and relative stability. However, with the advancement of robotics technology, complex environments challenge the rapid adaptability of traditional approaches, limiting system adaptability. We propose a safety interaction framework based on multimodal perception, adaptive safety constraint generation and dynamic adjustment, and collaborative decision-making strategies to address risks arising from dynamic uncertainty. At the robust control level, we integrate adaptive control with intelligent optimization strategies to handle unpredictable human behavior and dynamic environmental changes. This research achieves dynamic obstacle avoidance and conflict resolution through adaptive safety constraints, significantly enhancing the safe operation of human-robot collaborative systems across diverse scenarios.