Victor Fiorin De Carvalho
Integration of VRUs in Cooperative Perception: Extending CARLA and VaN3Twin.
Rel. Claudio Ettore Casetti, Carlos Mateo Risma Carletti, Francesco Raviglione, Diego Gasco. Politecnico di Torino, Corso di laurea magistrale in Automotive Engineering (Ingegneria Dell'Autoveicolo), 2025
|
Preview |
PDF (Tesi_di_laurea)
- Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives. Download (8MB) | Preview |
Abstract
The integration of Vulnerable Road Users (VRUs) into Cooperative Intelligent Transportation System (C-ITS) is essential to enhance road safety, however, it introduces significant challenges related to network scalability. Evaluating their integration in real-world field testing is still limited given the related costs, complexity, and logistical constraints, therefore, simulation plays an essential role in this area of research. This thesis extends the VaN3Twin-CARLA co-simulation framework to support and investigate passive and active participation of VRUs in cooperative scenarios. The contributions of this work are as follows: (i) The integration of VRUs into the OpenCDA framework, enabling realistic perception of VRUs; (ii) the implementation of ETSI-compliant communication services, allowing VRUs to actively participate by transmitting Vulnerable Road User Awareness Messages (VAMs); (iii) the implementation of a perception-based clustering mechanism for pedestrians, to evaluate its effect on Collective Perception Messages (CPM) size and channel load; (iv) the assessment of network impact of connected VRUs under varying densities and penetration rate.
The results demonstrate that clustering significantly reduces the fluctuations in CPM size, being three times lower, on average, in the most dense scenario
Tipo di pubblicazione
URI
![]() |
Modifica (riservato agli operatori) |
