Vision-Based Safe Human-Robot Collaboration with Uncertainty Guarantees

arXiv cs.RO / 4/17/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper presents a vision-based framework for human pose estimation and motion prediction aimed at enabling certifiably safe human-robot collaboration.
It combines aleatoric uncertainty estimation with out-of-distribution (OOD) detection to produce high probabilistic confidence in the predictions.
The authors propose conformal prediction sets for human motion forecasts, designed to provide valid uncertainty guarantees that can be used inside certifiable safety frameworks.
Evaluation includes both recorded human motion datasets and a real-world human-robot collaboration experiment to demonstrate the pipeline’s effectiveness.
The work targets “safety with uncertainty bounds” by structuring model outputs so they come with formal confidence guarantees rather than heuristics.

Abstract

We propose a framework for vision-based human pose estimation and motion prediction that gives conformal prediction guarantees for certifiably safe human-robot collaboration. Our framework combines aleatoric uncertainty estimation with OOD detection for high probabilistic confidence. To integrate our pipeline in certifiable safety frameworks, we propose conformal prediction sets for human motion predictions with high, valid confidence. We evaluate our pipeline on recorded human motion data and a real-world human-robot collaboration setting.