The field of Intelligent Transport Systems (ITS), and more specifically the development of intelligent vehicles with fully automated driving capabilities, is becoming a promising domain of application and technology transfer from robotics research. It also gives rise to new research themes, such as heterogeneous wireless communications.
The MobiVIP project, following the 'Automated Road Scenario', focuses on low speed applications (<30 km/h) in an open, or partly open, urban context.
The level of automation can vary from limited driving assistance to full autonomy. In all cases, an accurate (< 1 metre) and frequently updated (10 Hz) location of the vehicle in its environment is necessary. With the GPS (differential and/or dynamic RTK) it is now possible to reach such an accuracy in open environments, but the problem has not yet been solved in dense urban areas (urban canyon).
Another issue is the accurate location referencing, in terms of the ground environment, needed to implement autonomous driving control techniques. Research is currently very active in this field known as SLAM (Simultaneous Localisation and Mapping) or CML (Concurrent Mapping and Localisation) problem.
In the MobiVIP project, a major effort is devoted to the use of on-board vision coupled with hybrid GPS data, for modelling the urban environment.
Such a model is then used in automatic guidance by applying visual serving techniques developed by the research partners.
Experiments are carried out on the CyCab, a small electric vehicle, equipped with stereo cameras, differential hybrid GPS and inertial sensors (gyro-meter and odometers).
This project brings together 11 partners in order to define and then to then experiment, demonstrate and evaluate the impact of new information and communication technologies on a mobility service in city centres. This service, made up of small clean urban vehicles will work complementary to existing public transport.
The project deals with the management of multimodal information and optimisation of public space (localisation, models, etc). Demonstrations take place on 5 different sites, with different experimentation themes (information system, modelling, environment, navigation, evaluation).