Constant improvements in sensor technologies mean that machine-generated mobility data is usually at a higher spatial or temporal resolution than was possible before. Looking at the scale at which mobility data is being created, it is beyond the scope of a human’s capability to process this data and hence there is a clear need for automated information processing and analysis.
The AIT exploits this new wealth of mobility data to unravel large and complex spatial-temporal sensor datasets for modeling mobility patterns and identifying deviations from typical mobility behavior. Modeling mobility patterns is of strong interest to many stakeholders in the (multimodal) mobility domain, including cities, traffic planners, and transport operators.
The AIT offers consulting services and data mining software modules for mobility pattern learning giving insights to
- Trip purpose recognition by integration context information (e.g. land use)
- Major routes / paths and deviations
- Behavior profiles and preferences (routine trips and points of routine)
- recognition of spatial-temporal dependencies and rules