FRESHR-GSI: A Generalized Safety Model and Evaluation Framework for Mobile Robots in Multi-Human Environments
Pranav Kumar Pandey, Ramviyas Parasuraman, Prashant Doshi
AI summary
Problem
Existing human safety metrics target industrial manipulators or single-human interactions, failing to accurately assess safety for mobile robots in dynamic, multi-human shared spaces where averaging masks individual risks.
Approach
The authors develop FRESHR, a real-time RGB-D vision pipeline that detects humans and extracts distance, relative velocity, and bearing to compute a directional, bounded Generalized Safety Index (GSI) using proxemics zones and a smooth minimum aggregation for multiple humans.
Key results
- Introduced a bounded, proxemics-guided GSI metric (0–1 scale) for directional human safety assessment
- Developed FRESHR, a real-time RGB-D vision pipeline integrating YOLO-v7 and depth data for multi-human detection and tracking
- Validated GSI's accuracy and superiority over existing safety scales in real-world multi-human robot experiments
- Demonstrated effective crowd-robot safety analysis and motion planning evaluation using LogSumExp aggregation
Why it matters
Provides mobile robot developers and HRI researchers with a scalable, accurate safety evaluation tool that prevents risk masking in crowded environments, improving human-robot coexistence.
Abstract
Human safety is critical in applications involving close human-robot interactions (HRI) and is a key aspect of physical compatibility between humans and robots. While measures of human safety in HRI exist, these mainly target industrial settings involving robotic manipulators. Less attention has been paid to settings where mobile robots and humans share the space. This paper introduces a new robot-centered directional framework of human safety. It is particularly useful for evaluating mobile robots as they operate in environments populated by multiple humans. The framework integrates several key metrics, such as each human’s relative distance, speed, and orientation. The core novelty lies in the framework’s flexibility to accommodate different application requirements while allowing for both the robot-centered and external observer points of view. We instantiate the framework by using RGB-D based vision integrated with a deep learning based human detection pipeline, termed FRESHR (Framework for RGB-D based Evaluation of Safety of Humans for mobile Robots), to yield a proxemics- guided generalized safety index (GSI) that instantaneously assesses human safety. We extensively validate GSI’s capability of producing appropriate and fine-grained safety measures in real-world experimental scenarios and demonstrate its superior efficacy against extant safety models.