Enhanced Multi-Robot Systems with Generative AI and Machine Learning: A Cost-Effective Approach for Swarm Robotics

Abstract

Saurabh Hitendra Patel*

This study addresses the challenge of enhancing multi-robot systems by integrating generative AI and machine learning to optimize performance and cost-effectiveness. Traditional swarm robotics systems often face limitations in adaptability, efficiency, and scalability, primarily due to hardware constraints. Our approach builds on an existing master-slave architecture, leveraging Raspberry Pi and MSP430 microcontrollers, to incorporate distributed intelligence through federated learning and advanced AI techniques. The enhanced system demonstrates a 17.9 % improvement in object detection accuracy, a 65.3 % reduction in response latency, and a 31.2 % faster task completion rate, all while maintaining cost efficiency with an average upgrade cost of $86.37 per unit. By combining adaptive model compression, intelligent resource management, and optimized communication protocols, the framework reduces overhead by 66.7 % and ensures robust performance in dynamic environments. These advancements establish a pathway for achieving sophisticated swarm behaviors through software innovations rather than costly hardware upgrades, offering significant implications for industrial applications such as manufacturing, agriculture, and disaster response. This work highlights the potential for AI-driven transformations in swarm robotics, demonstrating how strategic integration of machine learning can enhance operational capabilities without compromising accessibility or affordability.

HTML PDF