Problem Statement

Increasing efficiency often involves predicting the behavior or the future state of a vehicle beyond the range of built-in radar or vision based sensors. The approach of a predictive electronic horizon combines the preselected route with information on the upcoming road sections. The information available is based on a selected trip destination and mostly statistical data. 

Objective

  • Prediction of most probable upcoming road segments based on historical data without an information about trip destination
  • Prediction of velocity profile on the road segments depending on the actual traffic state

Realization

  • Crowdsourcing of velocity profiles
  • Data preprocessing and transformation 
  • Data Mining for the applications trip and velocity prediction
  • Development of a prediction model for multidimensional state transitions
  • Feedback to the vehicle