Institution profile

Ontario Tech University

Academic institutionnorthamerica · ca
Official website
Research library54linked papers
Opportunities0open roles
Selected work

Representative Papers

DMAVA: Distributed Multi-Autonomous Vehicle Architecture Using Autoware

Jan 22, 2026

This work proposes the first fully distributed, decentralized multi-vehicle cooperative simulation framework based on Autoware, addressing the limitations of existing autonomous driving simulators that are often restricted to single-vehicle scenarios or rely on centralized control. The architecture concurrently executes multiple instances of Autoware Universe within a shared Unity environment, integrating ROS 2 Humble, AWSIM Labs, and the Zenoh communication layer to achieve low-latency, cross-host real-time synchronization. It enables advanced cooperative applications such as multi-vehicle autonomous valet parking and demonstrates, in a multi-machine setup, the feasibility and scalability of stable localization, reliable inter-vehicle communication, and closed-loop control without centralized coordination.

1 citationsRead paper

DMV-AVP: Distributed Multi-Vehicle Autonomous Valet Parking Using Autoware

Jan 22, 2026

This work proposes the first autonomous valet parking (AVP) system based on a distributed multi-vehicle architecture, termed DMAVA, addressing the limitations of existing centralized simulation frameworks that struggle to support scalable, fully autonomous coordination among multiple vehicles. The system integrates Autoware, Unity, and a YOLOv5-based visual perception module, and leverages the Zenoh communication middleware to enable low-latency message passing. Through a combination of state coordination, queuing strategies, and a parking-slot reservation mechanism, DMAVA achieves conflict-free and deterministic cooperative parking in multi-host environments. Experimental validation on two to three hosts demonstrates strong scalability and collaborative performance, establishing a foundation for future real-vehicle deployment and hardware-in-the-loop testing.

1 citationsRead paper

An Analysis of LLM Fine-Tuning and Few-Shot Learning for Flaky Test Detection and Classification

Feb 04, 2025

This paper addresses the challenge of detecting and classifying flaky tests—non-deterministic test cases that intermittently pass or fail—in automated testing. We propose FlakyXbert, a lightweight and efficient framework based on a Siamese neural network architecture for few-shot learning (FSL). The work systematically compares full fine-tuning of large language models (LLMs) against FSL in terms of accuracy–cost trade-offs. Experiments on the FlakyCat and IDoFT benchmarks demonstrate that while full LLM fine-tuning achieves high accuracy, FlakyXbert attains competitive performance using only a small number of labeled examples, substantially reducing both annotation effort and computational overhead. To our knowledge, this is the first empirical study to delineate the practical applicability boundary of FSL versus full fine-tuning specifically for flaky test detection. Our findings provide a deployable, resource-efficient solution for settings with limited labeled data and computational capacity.

1 citationsRead paper
Recent publications

Latest Papers