Multi-Target Defense Using Uncertain Differential Games and Reinforcement Learning
Implementing Organization
Indian Institute of Science
Principal Investigator
Dr. Aniruddha Roy
Indian Institute Of Science
roy9aniruddha@gmail.com
Project Overview
Pursuit and interception of moving targets are critical processes observed in both nature and engineering systems, and have attracted significant attention due to their wide range of applications. Representative scenarios include: guided missiles intercepting enemy aircraft, naval vessels defending against torpedoes, and coordinated unmanned aerial systems operations in rescue and surveillance missions. These problems are often modeled using differential games, particularly the Target–Attacker–Defender (TAD) framework. However, traditional models typically assume simplified dynamics and perfect state information and involve only a few agents, which fail to capture the complexity of real-world adversarial settings.
This proposal aims to develop a unified game-theoretic and learning-based framework for multi-target, multi-attacker, multi-defender (MT–MA–MD) scenarios in uncertain and constrained environments. The MT-MA-MD scenario modeled as finite-horizon uncertain differential games with finite-energy disturbances. A partial state information structure is considered, incorporating limited sensing, restricted communication, and obstacle avoidance to synthesize feedback strategies for real-time decision-making.
To enable decentralized coordination, the proposal introduces an opinion-driven role-switching mechanism in which each defender autonomously decides whether to intercept or rescue based on local observations and influence from neighboring agents. Given the computational complexity of solving nonlinear, high-dimensional differential games, we integrate multi-agent reinforcement learning (MARL) with dynamic game theory. Scalable MARL algorithms will be developed to learn equilibrium strategies in partially known environments while incorporating obstacle-aware safety constraints.
Simulation and real-world validation using UAV platforms will demonstrate the feasibility of the proposed methods. The outcomes will directly contribute to the development of autonomous, AI-driven defense capabilities, supporting India’s vision of technological self-reliance under Atmanirbhar Bharat.
Disclaimer:
Information available on this portal is sourced from various organizations and is provided for informational purposes only. Users are advised to verify details from the respective official sources.
Please enter your details
Please provide your name and email to continue. Your details are saved in this browser for future use.
Latest Updates
Loading…
⚠️
You are leaving this website
You are about to be redirected to an external website that is not operated by
India Science, Technology & Innovation (ISTI) Portal.