A Physics-Informed Digital Twin Framework for Calibrated Sim-to-Real FMCW Radar Occupancy Estimation

TL;DR

This paper introduces a physics-informed sim2real framework for FMCW radar occupancy detection and people counting, achieving high accuracy with minimal real data by aligning simulated and real noise characteristics.

Contribution

It presents a novel calibrated domain randomization method that improves simulation-to-real transfer for radar perception tasks using minimal real calibration data.

Findings

Achieves 97% accuracy in occupancy detection

Achieves 72% accuracy in people counting

Outperforms baseline simulation methods

Abstract

Learning robust radar perception models directly from real measurements is costly due to the need for controlled experiments, repeated calibration, and extensive annotation. This paper proposes a lightweight simulation-to-real (sim2real) framework that enables reliable Frequency Modulated Continuous Wave (FMCW) radar occupancy detection and people counting using only a physics-informed geometric simulator and a small unlabeled real calibration set. We introduce calibrated domain randomization (CDR) to align the global noise-floor statistics of simulated range-Doppler (RD) maps with those observed in real environments while preserving discriminative micro-Doppler structure. Across real-world evaluations, ResNet18 models trained purely on CDR-adjusted simulation achieve 97 percent accuracy for occupancy detection and 72 percent accuracy for people counting, outperforming ray-tracing baseline simulation and conventional random domain randomization baselines.