Contactless Respiratory Monitoring on Heterogeneous Mobile Robots: A Multimodal Edge-Computing Framework

2026-06-16 · Source: Computer Vision and Pattern Recognition · Field: Technology & Digital — Robotics & Autonomous Systems, Artificial Intelligence & Machine Learning, Computer Vision & Pattern Recognition · Depth: Expert, quick

Summary

A new modality-adaptive contactless respiratory-rate (RR) monitoring framework has been developed for heterogeneous mobile robots equipped with onboard edge computing. This system addresses challenges in field deployment for emergency response and disaster recovery by minimizing physical contact and improving safety. It integrates brightness-adaptive sensor selection across RGB, thermal, near-infrared (NIR), and low-light cameras, alongside keypoint-guided chest region-of-interest (ROI) extraction for posture-robust monitoring, and a signal-quality-index (SQI)-based filtering mechanism for reliable respiratory estimation. Evaluated on three distinct robotic platforms, including quadruped and wheeled locomotion, and various edge-computing architectures, the framework demonstrates generalization without per-platform algorithmic retuning. Performance varies by modality: RGB offers coverage up to 8m, NIR is effective up to 6m, thermal is reliable at short range, and low-light sensing supports monitoring in complete darkness up to 8m. This framework supports autonomous triage and victim assessment in hazardous search-and-rescue settings.

Key takeaway

For Robotics Engineers deploying mobile robots for remote triage or victim assessment in hazardous environments, you should consider integrating multimodal contactless respiratory monitoring. This framework allows your robots to adapt to varying conditions, using RGB for up to 8m range, NIR for 6m, and low-light cameras for complete darkness, without needing platform-specific retuning. This capability enhances operational safety and enables autonomous assessment in critical search-and-rescue scenarios.

Key insights

A multimodal, edge-computing framework enables robust contactless respiratory monitoring on diverse mobile robots for hazardous environments.

Principles

• Adaptive sensor selection improves robustness.
• Keypoint-guided ROI enhances posture resilience.
• SQI filtering ensures reliable data.

Method

The framework combines brightness-adaptive sensor selection, keypoint-guided chest ROI extraction, and signal-quality-index (SQI)-based filtering to estimate respiratory rates on mobile robots.

In practice

• Deploy RGB for broad 8m coverage.
• Use low-light sensors for dark environments.
• Integrate NIR for 6m range in varied light.

Topics

• Contactless Monitoring
• Mobile Robotics
• Edge Computing
• Multimodal Sensing
• Search and Rescue
• Computer Vision

Best for: Computer Vision Engineer, AI Scientist, Robotics Engineer, AI Engineer, Research Scientist

Related on AIssential

• Illumination-Robust Camera-Based Heart-Rate Estimation for Physiological Sensing in Robots — Robust remote photoplethysmography (rPPG) for robots is achievable despite illumination variations.
• Divide, Deliberate, Decide: A Multi-Agent Framework for Fine-Grained Egocentric Action Recognition — Multi-agent VLM deliberation with heterogeneous specialists improves fine-grained egocentric action recognition by decorrelating model priors.
• Stereo Vision-Based Fall Prediction and Detection using Human Pose Estimation on the AMD Kria K26 SOM — Privacy-preserving, real-time fall detection is feasible on edge devices using stereo vision and quantized human pose estimation.
• Energy-Aware NECO for Single-Pass Pixel-wise Out-of-Distribution Detection in Semantic Segmentation — A hybrid single-pass OOD detector for semantic segmentation improves accuracy and efficiency for mobile robots.
• Efficient Sensor Fusion for Gesture Recognition on Resource-Constrained Devices — Sensor fusion of low-resolution ToF and IR enables efficient, private gesture recognition on resource-constrained devices.
• An Integrated Roadside Sensing and Communication Framework for Vulnerable Road User Safety at Signalized Intersections — Multi-modal sensing and edge analytics significantly improve Vulnerable Road User safety at signalized intersections.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Computer Vision and Pattern Recognition.