DISTRIBUTED DEEP LEARNING FOR REAL-TIME OBJECT DETECTION IN AUTONOMOUS VEHICLES: A PERFORMANCE OPTIMIZATION STUDY

Read full text Summarise with VIRAT
Authors: Ronak Goyal -, Ashwini Somani, Naveena Rukumani Kannan -
Cited by: 0
References: 48

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This study examines the impact of key computational variables—Degree of Parallelism (DP), Compute Utilization (CU), Bandwidth Throughput (BT), and Model Complexity (MC)—on Real-Time Detection (RTD) performance in distributed deep learning systems for autonomous vehicles. Using a structured questionnaire and a sample of 327 respondents from New York’s autonomous systems industry, the research applied multiple linear regression analysis via R Studio. Results revealed that DP and BT significantly enhance RTD performance, while MC negatively affects it; CU showed no significant influence. The findings highlight the critical role of scalable computation and efficient data flow in optimizing real-time AI-based object detection. This study offers valuable insights for AI engineers, system architects, and policymakers aiming to advance intelligent mobility infrastructure in high-density urban environments. It also presents a framework for future research in AI deployment across real-time applications.

Fields of Study & Subjects

Advanced Neural Network Applications Parallel Computing and Optimization Techniques Autonomous Vehicle Technology and Safety

Publication details

Journal / source
International journal of innovative research and creative technology
Volume
12
Issue
2
Publication date
30 March 2026
Language
EN