DC15. Use of advanced AI Chat models for advising road authorities on Vulnerable Road User safety improvements

Sobia Bibi

Education
M.Sc in Computer Science in 2022 from COMSATS University Islamabad
B.Sc in Computer Science in 2016 from COMSATS University Islamabad 
About
Experienced researcher and lecturer currently pursuing a PhD at the University of Zagreb, Croatia. Specializes in artificial intelligence, computer vision, and machine learning, with a strong track record in translating complex research into actionable solutions. Adept at managing academic and professional responsibilities, known for working efficiently within dynamic environments and collaborating effectively with academic peers and stakeholders

Hosts: FPZ & ABEONA

Objectives:

  • To adapt available AI chat models (such as ChatGPT), to provide recommendations and advice to road authorities on enhancing road safety for VRUs (cyclists, pedestrians) groups, based on inputs from national handbooks, infrastructure guidelines and the iRAP assessment methodology.
  • To evaluate the effectiveness and efficiency of the AI chat model in providing accurate, relevant, and timely advice based on national and proactive infrastructure safety standards.
  • To facilitate increased interaction and communication between road authorities and AI systems for efficient decision-making.

Expected results:

  • A comprehensive, fine-tuned AI chat model, capable of providing context-specific recommendations to road authorities on improving road safety for VRUs.
  • An evaluation of the efficiency and effectiveness of the AI chat model in advising road authorities.
  • Enhanced understanding and acceptance of AI systems in decision-making processes among road authorities.

Planned secondment(s):

iRAP-UK, Purpose: international exposure and explore global issues of Chat AI assistance in a non-European context.