Combining Edge Computing to Enhance Real-Time Control and Low-Latency Response of Line Robots

Abstract

 In the context of real-time control and latency response scenarios for production line robots, this study focuses on optimizing the “cloud-edge-end” three-layer heterogeneous deployment based on edge computing. Multi-agent reinforcement learning methods are employed for task scheduling optimization, enabling distributed control to reduce resource waste and further improve load balancing. Dynamic event-triggered communication optimization reduces latency while minimizing network resource consumption. Through simulation testing, applying this method to optimize real-time control latency on the cloud platform achieved over 85% higher latency reduction compared to traditional cloud platform optimization tests. The system operates stably in dynamic complex scenarios, providing both theoretical and practical research and exploration into production line robot control optimization.