PRIVACY-PRESERVING AI FOR SUSTAINABLE TRANSPORT

A modal shift away from private vehicle use and towards sustainable public transport (PT) is essential to combat climate change. In addition to ticket costs, delays, lack of comfort and overcrowding are the main reasons that prevent people from using public transport. 

AI4PT develops privacy-aware AI solutions that predict and optimise passenger distribution and flows to increase the capacity and attractiveness of public transport. It thus aims to increase the capacity of the public transport network, especially on routes that already have tight schedules.

The lack of high-quality data in public transport environments is a major obstacle to the development of AI applications. AI4PT therefore relies on cost-effective camera-based systems with integrated object recognition (SmartCams) in combination with existing subway vehicle sensors. During the project, these data sources will be linked with Federated Learning (FL) methods to develop data protection-compliant, predictive AI models for subway operations.

Facts

• Project start: June 2024
• Project duration (in months): 30 months
• Budget: ca. 460k EUR
• Partners: Indicate Digital GmbH, Siemens Mobility Austria GmbH
• Funding: FFG Digitale Technologien, AI for Green 2023 Call https://projekte.ffg.at/projekt/5121351
• Project website: N/A