System Architecture in a Heavy DutyVehicle Platooning System using xPCTargetTM

Sammanfattning: The ongoing environmental debate emphasizes the need for reducing CO2 emissions in order to counteract rapid climate change. The transportation industry is growing and is already today a signicant contributor to green house gas emissions. Information and Communication Technology (ICT) can be utilized to create intelligent transportation systems which have proven to be able to reduce fuel consumption signicantly. The concept of vehicle platooning involves vehicles traveling in a longitudinal convoy with a short intermediate distance. By introducing ICT to a vehicle platoon the intermediate distance can be even shorter without compromising trac safety. The shorter distance also allows for reduced drag, thus reduced fuel consumption. A research group from KTH Royal Institute of Technology and Scania CV AB competed successfully with an autonomous platooning system in the Grand Cooperative Driving Challenge in 2011. In 2012 the cooperation continued and a new platooning system including lateral maneuvers, i.e. lane changing, was implemented. This master thesis presents the design and implementation of a new hardware and software architecture for the platooning system. For exibility and control prototyping issues a new real-time platform supporting the system was tested. The choice was to evaluate and test the xPC Target TM system developed by MathWorks. The xPC Target platform supports real-time code generation from Simulink Rand also provides an extensive I/O library needed for communication with external physical systems. In order to adapt to the xPC Target platform and lateral platooning scenarios a new system architectural framework was suggested. The architecture emphasized the need for separated components each with clearly delimited tasks. Another important requirement was to provide a system with a short learning curve for new master students that will work with the system in the future. A new system architecture approach was presented and implemented in two real-time computer systems. The rst computer executed a real-time software created in C++ which primarily handled wireless transmission. The second unit utilized a real-time software created in Simulink Rand executed on the xPC Target platform. Successful vehicle platooning experiments, including lateral maneuvers, were performed with the new system in November 2012.