Novel high-performance multi-modal routing service for first and last mile trip requests with hierarchical multi agent reinforcement learning
Loading...
DOI of the original publication
Project type
angewandte Forschung
Project start
01.03.2025
Project end
29.02.2028
Project status
laufend
Project contact
Project manager
Contributors
Description
Abstract
The Meta-Router is a novel service to solve the first and last mile routing problem by ensuring trip request being handled over all available means of public transport. The unique hierarchical multi-agent reinforcement learning approach generates valid and user centred trip options in a short time.
Link
Created during FHNW affiliation
Yes
Strategic action fields FHNW
Zero Emission
School
Hochschule für Wirtschaft FHNW
Institute
Institut für Wirtschaftsinformatik
Financed by
Innosuisse
Project partner
Schweizerische Bundesbahn SBB
Contracting authority
SAP reference
Keywords
First and Last Mile Routing
Mulit-Modal Routing
Reinforcement Learning
Public Transport
Mulit-Modal Routing
Reinforcement Learning
Public Transport