Synthetic Enclosed Echoes: A New Dataset to Mitigate the Gap Between Simulated and Real-World Sonar Data

TL;DR

Addressing the sim-to-real gap in underwater perception, this paper introduces SEE, a predominantly synthetic sonar dataset with a real-world subset collected in a controlled tank. The authors build a high-fidelity simulation pipeline using HoloOcean/Unreal Engine to generate ground-truth-rich data (CAD models and rangefinder point clouds) across four tank-scene scenarios with 40 props, totaling 15,536 synthetic images and 142,361 files, plus metadata. They evaluate classical reconstruction methods and learning-based approaches (including a modified ElevateNET variant, ElevateNET R*) and demonstrate improved performance over baselines on synthetic data, with quantitative measures like RMS and Hausdorff distance and a generalization test. The work provides a publicly available dataset and code, enabling future research in robust underwater 3D reconstruction and sim-to-real transfer.

Abstract

This paper introduces Synthetic Enclosed Echoes (SEE), a novel dataset designed to enhance robot perception and 3D reconstruction capabilities in underwater environments. SEE comprises high-fidelity synthetic sonar data, complemented by a smaller subset of real-world sonar data. To facilitate flexible data acquisition, a simulated environment has been developed, enabling the generation of additional data through modifications such as the inclusion of new structures or imaging sonar configurations. This hybrid approach leverages the advantages of synthetic data, including readily available ground truth and the ability to generate diverse datasets, while bridging the simulation-to-reality gap with real-world data acquired in a similar environment. The SEE dataset comprehensively evaluates acoustic data-based methods, including mathematics-based sonar approaches and deep learning algorithms. These techniques were employed to validate the dataset, confirming its suitability for underwater 3D reconstruction. Furthermore, this paper proposes a novel modification to a state-of-the-art algorithm, demonstrating improved performance compared to existing methods. The SEE dataset enables the evaluation of acoustic data-based methods in realistic scenarios, thereby improving their feasibility for real-world underwater applications.

Figures (7)

• Figure 1: Synthetic Enclosed Echoes: a new dataset of synthetic and real-world sonar data for a closed underwater environment
• Figure 2: Indoor Tank Facilities.
• Figure 3: An overview of the SEE structure, exemplifying the scenarios developed and their respective collected data.
• Figure 4: An overview of all props present in the SEE dataset.
• Figure 5: Proposed scenarios for the dataset and their respective way-points