Date of Award
9-20-2024
Publication Type
Thesis
Degree Name
M.A.Sc.
Department
Electrical and Computer Engineering
Supervisor
Ahmed Sakr
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Abstract
In traditional cooperative awareness (CA) and cooperative perception (CP), connected vehicles exchange periodic messages containing their current status and perceived objects. This thesis investigates the concept of intent sharing-based cooperative perception (IBCP) for connected vehicles. IBCP introduces a more interactive framework wherein an ego vehicle communicates its intent (i.e., future trajectory), while other connected road users and infrastructure share alerts concerning potential conflicts with the ego vehicle. I compare and analyze the performance of the IBCP protocol in contrast to the existing CA and CP systems. The primary objectives of this thesis, are to assess the anticipation of potential conflicts by the ego vehicle and to measure how effectively the ego vehicle can react comfortably. Experimental data is employed to evaluate the performance of the IBCP protocol in two distinct scenarios: one at an intersection and another at a roundabout, each presenting different types of occlusions. We use simulated scenarios to assess the manifold benefits that IBCP brings to a platoon of vehicles executing a left turn at an intersection. Results show that IBCP enhances situational awareness and comfort compared to traditional CA and CP. Furthermore, adding IBCP on top of the CA and CP services does not add a significant channel load. We use experimental data and simulated scenarios to assess the manifold benefits that IBCP brings to a platoon of vehicles executing a left turn at an intersection. Our evaluation parameters encompass safety improvements, enhanced situational awareness, heightened comfort levels, and the mitigation of disruptions within the vehicular platoon.
Recommended Citation
Manjunatha, Shishir, "Intent Sharing-Based Cooperative Perception for Connected Vehicles" (2024). Electronic Theses and Dissertations. 9393.
https://scholar.uwindsor.ca/etd/9393