Agent as Cerebrum, Controller as Cerebellum: Implementing an Embodied LMM-based Agent on Drones

TL;DR

This paper introduces AeroAgent, an embodied LMM-based drone agent architecture with ROS integration, demonstrating superior performance in search and rescue tasks over traditional DRL agents.

Contribution

It presents a novel embodied agent framework combining LMMs with robotic control, including a new linkage system and empirical validation in simulated and real-world scenarios.

Findings

AeroAgent outperforms existing DRL-based agents in complex tasks.

The ROSchain framework enables seamless integration with robotic systems.

Empirical results show improved efficiency in search and rescue operations.

Abstract

In this study, we present a novel paradigm for industrial robotic embodied agents, encapsulating an 'agent as cerebrum, controller as cerebellum' architecture. Our approach harnesses the power of Large Multimodal Models (LMMs) within an agent framework known as AeroAgent, tailored for drone technology in industrial settings. To facilitate seamless integration with robotic systems, we introduce ROSchain, a bespoke linkage framework connecting LMM-based agents to the Robot Operating System (ROS). We report findings from extensive empirical research, including simulated experiments on the Airgen and real-world case study, particularly in individual search and rescue operations. The results demonstrate AeroAgent's superior performance in comparison to existing Deep Reinforcement Learning (DRL)-based agents, highlighting the advantages of the embodied LMM in complex, real-world scenarios.