Emergency Response Person Localization and Vital Sign Estimation Using a Semi-Autonomous Robot Mounted SFCW Radar

TL;DR

This paper tackles emergency-response localization of multiple people and breathing-rate estimation using a robot-mounted SFCW radar. It introduces a complete, unsupervised signal-processing chain that builds a virtual MIMO array, performs 2D-MUSIC localization with spatial smoothing, applies clutter rejection, and extracts vital signs via a spatial filter whose output phase encodes chest motion; breathing rates are then estimated from spectral peaks. A comprehensive open-access dataset (62 radar measurements, 59 ground-truth vital-sign records across 62 scenarios with up to five persons) is provided to benchmark multi-person detection, localization, and vital-sign estimation in free-space, through-wall, and door-occluded settings. Results show sub-15 cm localization errors in many free-space cases, with performance degrading under wall conditions, highlighting practical challenges and opportunities for robust through-wall sensing and tracker integration in disaster-response robotics. The work offers a solid baseline, quantitative benchmarks, and openly available code to accelerate development of radar-based rescue technologies with real-world impact.

Abstract

The large number and scale of natural and man-made disasters have led to an urgent demand for technologies that enhance the safety and efficiency of search and rescue teams. Semi-autonomous rescue robots are beneficial, especially when searching inaccessible terrains, or dangerous environments, such as collapsed infrastructures. For search and rescue missions in degraded visual conditions or non-line of sight scenarios, radar-based approaches may contribute to acquire valuable, and otherwise unavailable information. This article presents a complete signal processing chain for radar-based multi-person detection, 2D-MUSIC localization and breathing frequency estimation. The proposed method shows promising results on a challenging emergency response dataset that we collected using a semi-autonomous robot equipped with a commercially available through-wall radar system. The dataset is composed of 62 scenarios of various difficulty levels with up to five persons captured in different postures, angles and ranges including wooden and stone obstacles that block the radar line of sight. Ground truth data for reference locations, respiration, electrocardiogram, and acceleration signals are included. The full emergency response benchmark data set as well as all codes to reproduce our results, are publicly available at https://doi.org/10.21227/4bzd-jm32.

Figures (21)

• Figure 1: emergenCITYrobot 'Scout'
• Figure 2: Single transmit waveform of an SFCW radar.
• Figure 3: MIMO setup with two transmit antennas and $M_{r}$ physical receive antennas, which can be extended to $M$ virtual antennas.
• Figure 4: Visualization of the model order estimation procedure. Only the first 30 indices are shown.
• Figure 5: An illustration of the spatial filtering procedure for Person 1 in M16. The right plot, which shows the output of the spatial filter clearly demonstrates the extraction of the breathing signal.