All Projects
UAVAgentic AILLM

Agentic AI for UAV

Unmanned aerial vehicles (UAVs) are evolving from remotely piloted platforms into autonomous agents capable of perception, reasoning, and cooperative decision-making. This project builds agentic AI systems—grounded in large language models, tool use, and multi-agent planning—that allow UAVs to interpret high-level mission intent, adapt to unforeseen conditions, and coordinate with other aerial and ground agents with minimal human supervision.

Research Objectives

  • Design LLM-driven planning agents that translate natural-language mission goals into executable UAV task graphs
  • Enable closed-loop perception-reasoning-action cycles for autonomous navigation and task execution
  • Develop multi-agent coordination protocols for swarms of UAVs under communication and energy constraints
  • Ensure safe, verifiable, and policy-compliant behavior of agentic UAVs in shared airspace

Methods & Techniques

  • Tool-augmented LLM agents with memory and reflection for mission planning
  • Hierarchical reinforcement learning coupled with language-based task decomposition
  • Multi-agent communication via structured natural-language protocols
  • Runtime guardrails and formal safety checks for autonomous flight decisions

Interested in this research?

Get in touch to discuss collaboration or graduate opportunities.

Contact Dr. Hossain