Multi-Objective Reinforcement Learning for Cognitive Radar Resource Management

TL;DR

This paper applies deep reinforcement learning to optimize resource management in cognitive radar systems, balancing target detection and tracking through Pareto-efficient solutions.

Contribution

It introduces a multi-objective optimization framework using DDPG and SAC algorithms for cognitive radar resource management, comparing their performance.

Findings

SAC outperforms DDPG in stability and sample efficiency

Both algorithms effectively adapt to different scenarios

NSGA-II provides an upper bound on the Pareto front

Abstract

The time allocation problem in multi-function cognitive radar systems focuses on the trade-off between scanning for newly emerging targets and tracking the previously detected targets. We formulate this as a multi-objective optimization problem and employ deep reinforcement learning to find Pareto-optimal solutions and compare deep deterministic policy gradient (DDPG) and soft actor-critic (SAC) algorithms. Our results demonstrate the effectiveness of both algorithms in adapting to various scenarios, with SAC showing improved stability and sample efficiency compared to DDPG. We further employ the NSGA-II algorithm to estimate an upper bound on the Pareto front of the considered problem. This work contributes to the development of more efficient and adaptive cognitive radar systems capable of balancing multiple competing objectives in dynamic environments.