Back to search

Carleton University

Intelligent Cooperative Sensing for Connected and Autonomous Vehicles

Abstract

dc:description.abstract

Connected and autonomous vehicles (CAVs) enhance driving safety and efficiency through onboard sensors and vehicle-to-everything (V2X) communications. However, the limited sensing capabilities of individual vehicles and constraints in computational and communication resources pose significant challenges for reliable autonomous driving. This thesis addresses these challenges by developing innovative methods for intelligent cooperative sensing in CAVs, focusing on three key aspects: optimal sensor deployment, efficient data transmission, and multi-level information fusion. First, we propose an enhanced 3D sensor deployment method using Decision Transformers (DT) to optimize the placement of roadside sensors while considering traffic patterns and CAV sensing requirements. The method incorporates a bi-level optimization framework with reward redistribution to effectively handle long-delayed rewards, significantly improving sensing coverage and detection accuracy in complex road environments. Second, we introduce an intelligence-guided rewardless reinforcement learning (IRL) framework to enhance sensing information processing and transmission. Unlike conventional reinforcement learning approaches that rely on explicit reward functions, our method leverages intelligence as a metric to guide the learning process, enabling improved adaptability in diverse driving environments. The framework considers heterogeneous sensing requirements and resource constraints while ensuring reliable sensing assistance through efficient resource utilization. Third, we develop an improved DT-based cooperative sensing framework that intelligently integrates multi-source and multi-level sensing information. The framework employs a novel reward shaping mechanism and bi-level optimization to effectively handle long-delayed rewards and adapt to dynamic driving scenarios. This approach enables efficient fusion of sensor data at various abstraction levels while optimizing resource allocation. Comprehensive simulations and experiments demonstrate the effectiveness of our proposed methods. The enhanced 3D sensor deployment method achieves up to 20\% improvement in sensing coverage with fewer sensors compared to existing approaches. The IRL framework shows significant enhancements in detection accuracy and latency reduction across various traffic scenarios. The improved DT-based fusion approach demonstrates superior performance in sensing accuracy and resource utilization. These results validate the practical applicability of our solutions in real-world CAV environments, contributing to the advancement of reliable and efficient autonomous driving systems.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy (Ph.D.)
Level thesis:degree_level
Doctoral
Discipline thesis:degree_discipline
Information Technology
Grantor dc:publisher
Carleton University
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Zhao, Pincan

Rights

dc:rights
Statement dc:rights
  • Copyright © 2025 the author(s). Theses may be used for non-commercial research, educational, or related academic purposes only. Such uses include personal study, distribution to students, research and scholarship. Theses may only be shared by linking to the Carleton University Institutional Repository and no part may be copied without proper attribution to the author; no part may be used for commercial purposes directly or indirectly via a for-profit platform; no adaptation or derivative works are permitted without consent from the copyright owner.
Language dc:language.iso
en

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Carleton University
Base URL
carleton.scholaris.ca/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Zhao, Pincan. Intelligent Cooperative Sensing for Connected and Autonomous Vehicles. Doctoral thesis, Carleton University, 2025. https://hdl.handle.net/20.500.14718/43891