SynthSoM-Twin: A Multi-Modal Sensing-Communication Digital-Twin Dataset for Sim2Real Transfer via Synesthesia of Machines

TL;DR

SynthSoM-Twin is a comprehensive multi-modal dataset designed for effective Sim2Real transfer in autonomous systems, validated through cross-modal tasks showing high efficiency with minimal real-world data injection.

Contribution

The paper introduces SynthSoM-Twin, a novel multi-modal sensing-communication dataset with a new framework for extending existing datasets and ensuring spatio-temporal consistency for Sim2Real transfer.

Findings

Models trained on SynthSoM-Twin perform well in real-world tasks.

Injecting less than 15% real-world data achieves comparable or better results.

The dataset enables effective cross-modal generative model training for Sim2Real transfer.

Abstract

This paper constructs a novel multi-modal sensing-communication digital-twin dataset, named SynthSoM-Twin, which is spatio-temporally consistent with the real world, for Sim2Real transfer via Synesthesia of Machines (SoM). To construct the SynthSoM-Twin dataset, we propose a new framework that can extend the quantity and missing modality of existing real-world multi-modal sensing-communication dataset. Specifically, we exploit multi-modal sensing-assisted object detection and tracking algorithms to ensure spatio-temporal consistency of static objects and dynamic objects across real world and simulation environments. The constructed scenario is imported into three high-fidelity simulators, i.e., AirSim, WaveFarer, and Sionna RT. The SynthSoM-Twin dataset contains spatio-temporally consistent data with the real world, including 66,868 snapshots of synthetic RGB images, depth maps, light detection and ranging (LiDAR) point clouds, millimeter wave (mmWave) radar point clouds, and large-scale and small-scale channel fading data. To validate the utility of SynthSoM-Twin dataset, we conduct Sim2Real transfer investigation by implementing two cross-modal downstream tasks via cross-modal generative models (CMGMs), i.e., cross-modal channel generation model and multi-modal sensing-assisted beam generation model. Based on the downstream tasks, we explore the threshold of real-world data injection that can achieve a decent trade-off between real-world data usage and models' practical performance. Experimental results show that the model training on the SynthSoM-Twin dataset achieves a decent practical performance, and the injection of real-world data further facilitates Sim2Real transferability. Based on the SynthSoM-Twin dataset, injecting less than 15% of real-world data can achieve similar and even better performance compared to that trained with all the real-world data only.