Skip to content

Edge AI for Automotive with Control Loop-Based QoS Optimization

License

sachinkum0009/edge-ai-research-thesis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 

Repository files navigation

edge-ai-research-thesis

Abstract

The motivation of this thesis is to examine the offloading of AI perception tasks from vehicles to edge computing nodes, utilizing YOLOP algorithm for concurrent lane detection, road segmentation, and object detection. The Emulate Edge Diagnostic Platform tests system performance under varying conditions. The proposed solution integrates Zenoh protocol and V2X communication with dynamic switching algorithms that respond to network latency and CPU stress metrics. Performance evaluation across same-network and cross-network edge scenarios demonstrates improved energy efficiency while maintaining responsive perception capabilities essential for autonomous driving applications.

Instructions to run

Coming soon.

About

Edge AI for Automotive with Control Loop-Based QoS Optimization

Topics

Resources

License

Stars

Watchers

Forks