Software-in-the-loop simulation as testing and visualisation framework for ITS algorithms

Testing intelligent transportation systems and their underlying algorithms in the field is a necessary step towards near market prototypes. Such tests, however, are always timeconsuming, costly and potentially cumbersome. It usually involves moving the prototype vehicle or vehicles to a certified test site, weather conditions need to be favourable, and the participating researchers need to be able to operate the vehicle. Running preliminary simulations in a realistic virtual environment can allow for faster prototyping of the underlying algorithms. Moreover, if recorded data from field trials can be replayed in simulation, the algorithms can be tested with realistic data as if they were being run on the actual physical systems. This paper presents the usage of a robotics simulator, MORSE, to address this issue. It also shows how real motion data from field experiments with autonomous vehicles is replayed in simulation, by combining it with virtual sensors and actuators. Additionally, it shows how a cooperative situation awareness application in which a set of intelligent vehicles, road side units and pedestrians on an intersection can be represented and pedestrian detection and motion flow estimation could be communicated.status: publishe