There and Back Again: Learning to Simulate Radar Data for Real-World Applications

TL;DR

This paper introduces a learned radar sensor model that synthesizes realistic radar data from elevation maps, enabling effective training of models on simulated data with minimal real-world data, thus accelerating radar processing development.

Contribution

It presents a novel adversarial and cyclical consistency-based approach to learn a radar sensor model from unaligned radar data and lidar ground truth, improving simulation fidelity.

Findings

Achieves real-world segmentation performance within 4% of models trained on real data.

Demonstrates effective radar data synthesis from elevation maps.

Validates the approach with downstream radar segmentation tasks.

Abstract

Simulating realistic radar data has the potential to significantly accelerate the development of data-driven approaches to radar processing. However, it is fraught with difficulty due to the notoriously complex image formation process. Here we propose to learn a radar sensor model capable of synthesising faithful radar observations based on simulated elevation maps. In particular, we adopt an adversarial approach to learning a forward sensor model from unaligned radar examples. In addition, modelling the backward model encourages the output to remain aligned to the world state through a cyclical consistency criterion. The backward model is further constrained to predict elevation maps from real radar data that are grounded by partial measurements obtained from corresponding lidar scans. Both models are trained in a joint optimisation. We demonstrate the efficacy of our approach by evaluating a down-stream segmentation model trained purely on simulated data in a real-world deployment. This achieves performance within four percentage points of the same model trained entirely on real data.