View source for Driver Prediction for Automative Industry

{{StudentProjectTemplate
|Title=Predicting driver intions
|Summary=Investigate if and how it is possible to predict the drivers actions and inentions in a predefined limited number of scenarios
|Author=Stefan Karlsson, Cristofer Englund
|Supervisor=Stefan Karlsson, Cristofer Englund
|Level=Master
|Status=Open
}}
The goal is to investigate if and how it is possible to predict
the drivers actions and inentions in a predefined limited number of scenarios.

This master thesis will conduct experiments on the data from FOT projects. This data is from the vehicles during every-day-use.  The
current database consists of vehicle Controller Area Network (CAN) data and
video material from cameras in the vehicle both facing the road and the driver from
100 cars and 20 trucks. 

The project will develop knowledge in driver intention prediction. The scenarios
to study will be defined in the beginning of the project and the data will be
extracted from the FOT database. Investigations on and evaluation on methods
that are suitable will be performed. Based on the experimental investigations
a final method will be proposed and the results will be presented.

[[media: ThesisSuggestDriverPrediction.pdf‎ |read more]]