DEVELOPING EDGE COMPUTING SOLUTIONS FOR IOT DEVICES TO REDUCE LATENCY AND ENHANCE REAL-TIME DECISION-MAKING

Abstract

Multimodal AI integrated with Edge Computing provides better real-time decisions through their synergy by efficiently processing data nearby its origin points along with analyzing multiple input data such as images and sensors. Both technologies create essential functionality when combined for speeding up autonomous car operations and healthcare patient monitoring systems. Various features of edge devices limit their processing ability as well as spending energy and  securing data. The study evaluates three optimization approaches that reduce model size through pruning and use precision quantization for accuracy reduction along with customized AI processors to boost processing speed which resolves these limitations. The purpose behind these solutions generates minimal performance loss for sophisticated AI models which execute on constrained edge devices. Various domains stand to benefit from data-driven real-time decision-making applications because of multimodal AI's power when merged with edge computing processes. Advanced hardware and software systems develop continuously which enlarges existing boundaries to produce increasingly intelligent and quick systems.