Transport networks are the heart of economies and societies. An increasing number of vehicle types such as trains, buses, cars and ships to cargo containers communicate with processing centres, transmitting identification, positioning and other useful data. Digital journey planners help travellers plan intermodal trips and provide real time updates of travel routes, weather forecasts and traffic conditions. The AIT develops novel mobility data collection and analysis techniques for generating comprehensive and instantaneous views of movements in transport networks as, thus providing valuable decision support for reactive measures and interventions.
The objective of AIT’s research activities in mobility data analytics for dynamic and interactive transport management is to capture and analyse mobility data from different sensor sources in real time to provide accurate and comprehensive insights into transportation systems. AIT develops major building blocks for
- real-time identification of multimodal traffic state patterns (travel times, level-of-service) including unusual event detection and short-term and mid-term predictions based on various (combined) data sources such as cell phone communication signal datasets, GPS, Bluetooth, toll systems, etc.
- reliable detection of public transport journeys of people for innovative electronic ticketing concepts such as ‘Check In Be Out’ or ambient interaction-free ‘Be In – Be Out’ systems, based on robust pattern analysis of smartphone sensor data fused with static and operational public transport network data
- exploiting massive Automatic Identification System (AIS) vessel data streams in real time for intelligent maritime navigation (route prediction and collision avoidance, travel time prediction and abnormal manoeuvre detection for maritime control centres)
- reliable detection of motorized trips for innovative individual dynamic parking pricing parking and road user charging schemes based on robust pattern analysis of people’s smartphone sensor data
All techniques for mobility data analytics for dynamic and interactive transport management are strictly compliant with the EU General Data Protection Regulation.
A. Graser and P. Widhalm: “Modelling massive AIS streams with quad trees and Gaussian Mixtures”, 21st AGILE Conference on Geographic Information Science June 12-15, 2018 Lund, Sweden
H. Koller, Peter Widhalm, M. Dragaschnig, A. Graser:
"Fast Hidden Markov Model Map-Matching for Sparse and Noisy Trajectories"; in: "Proceedings 18th IEEE International Conference on Intelligent Transportation Systems (ITSC)", IEEE (Hrg.); (2015).
A. Graser, W. Ponweiser, M. Dragaschnig, N. Brändle, Peter Widhalm: "Assessing Traffic Performance using Position Density of Sparse FCD"; in: Proceedings 15th International IEEE Conference on Intelligent Transportation Systems (ITSC)", (2012).
Peter Widhalm, M. Piff, N. Brändle, H. Koller, M. Reinthaler, W. Ponweiser: "Robust Road Link Speed Estimates for Sparse or Missing Probe Vehicle Data"; in: "Proceedings 15th IEEE Intelligent Transportation Systems Conference (ITSC2012), 2012.