Interpretable DRL-based Maneuver Decision of UCAV Dogfight

TL;DR

This paper introduces a DRL-based framework for UCAV dogfight maneuver decision-making, achieving high success rates and interpretability, with the agent developing advanced tactics like 'Dive and Chase'.

Contribution

It presents a three-layer UCAV dogfight framework using DDQN for maneuver selection, enhancing interpretability and tactical diversity in autonomous aerial combat.

Findings

Agent achieves 85.75% win rate against DT strategy.

The framework improves interpretability of DRL decisions.

Emergence of novel tactics like 'Dive and Chase'.

Abstract

This paper proposes a three-layer unmanned combat aerial vehicle (UCAV) dogfight frame where Deep reinforcement learning (DRL) is responsible for high-level maneuver decision. A four-channel low-level control law is firstly constructed, followed by a library containing eight basic flight maneuvers (BFMs). Double deep Q network (DDQN) is applied for BFM selection in UCAV dogfight, where the opponent strategy during the training process is constructed with DT. Our simulation result shows that, the agent can achieve a win rate of 85.75% against the DT strategy, and positive results when facing various unseen opponents. Based on the proposed frame, interpretability of the DRL-based dogfight is significantly improved. The agent performs yo-yo to adjust its turn rate and gain higher maneuverability. Emergence of "Dive and Chase" behavior also indicates the agent can generate a novel tactic that utilizes the drawback of its opponent.