Fighter Jet Navigation and Combat using Deep Reinforcement Learning with Explainable AI

TL;DR

This paper develops a deep reinforcement learning-based fighter jet AI within a custom simulation, achieving over 80% success in navigation and combat tasks while providing explainability through counterfactual analysis.

Contribution

It introduces a novel DRL framework for fighter jet navigation and combat with integrated explainable AI, enhancing transparency in decision-making.

Findings

Over 80% task completion rate achieved

Effective decision-making demonstrated through reward analysis

Explainability via counterfactual reward comparison

Abstract

This paper presents the development of an Artificial Intelligence (AI) based fighter jet agent within a customized Pygame simulation environment, designed to solve multi-objective tasks via deep reinforcement learning (DRL). The jet's primary objectives include efficiently navigating the environment, reaching a target, and selectively engaging or evading an enemy. A reward function balances these goals while optimized hyperparameters enhance learning efficiency. Results show more than 80\% task completion rate, demonstrating effective decision-making. To enhance transparency, the jet's action choices are analyzed by comparing the rewards of the actual chosen action (factual action) with those of alternate actions (counterfactual actions), providing insights into the decision-making rationale. This study illustrates DRL's potential for multi-objective problem-solving with explainable AI. Project page is available at: \href{https://github.com/swatikar95/Autonomous-Fighter-Jet-Navigation-and-Combat}{Project GitHub Link}.