Automated Gain Control Through Deep Reinforcement Learning for Downstream Radar Object Detection

TL;DR

This paper introduces a deep reinforcement learning approach for automated gain control in cognitive radars, optimizing object detection performance using a synthetic dataset generated from a gaming engine.

Contribution

It presents a novel method combining deep RL with a YOLOv3 detector and synthetic data for adaptive radar gain control in object detection tasks.

Findings

Deep RL effectively learns gain control policies.

Synthetic dataset enables extensive training scenarios.

Improved object detection accuracy achieved.

Abstract

Cognitive radars are systems that rely on learning through interactions of the radar with the surrounding environment. To realize this, radar transmit parameters can be adapted such that they facilitate some downstream task. This paper proposes the use of deep reinforcement learning (RL) to learn policies for gain control under the object detection task. The YOLOv3 single-shot object detector is used for the downstream task and will be concurrently used alongside the RL agent. Furthermore, a synthetic dataset is introduced which models the radar environment with use of the Grand Theft Auto V game engine. This approach allows for simulation of vast amounts of data with flexible assignment of the radar parameters to aid in the active learning process.