Full-Scale Continuous Synthetic Sonar Data Generation with Markov Conditional Generative Adversarial Networks
This addresses the high cost and time of deploying autonomous underwater vehicles by enabling faster, controlled synthetic data generation for training and validation, though it is incremental as it builds on existing generative models.
The paper tackles the challenge of generating realistic synthetic sonar data for underwater vehicles by proposing MC-pix2pix, a method that produces data almost indistinguishable from real sonar imagery and improves autonomous target recognition performance when used for bootstrapping.
Deployment and operation of autonomous underwater vehicles is expensive and time-consuming. High-quality realistic sonar data simulation could be of benefit to multiple applications, including training of human operators for post-mission analysis, as well as tuning and validation of autonomous target recognition (ATR) systems for underwater vehicles. Producing realistic synthetic sonar imagery is a challenging problem as the model has to account for specific artefacts of real acoustic sensors, vehicle altitude, and a variety of environmental factors. We propose a novel method for generating realistic-looking sonar side-scans of full-length missions, called Markov Conditional pix2pix (MC-pix2pix). Quantitative assessment results confirm that the quality of the produced data is almost indistinguishable from real. Furthermore, we show that bootstrapping ATR systems with MC-pix2pix data can improve the performance. Synthetic data is generated 18 times faster than real acquisition speed, with full user control over the topography of the generated data.