Real-time information and data on the means of transport used and the type of movement and transportation are a basis for the future viability of logistics and mobility in Hesse. This information will play an even more important role in the important fields of innovation, autonomous driving, accident prevention and for a networked transport infrastructure. For these applications, the recognition of human activities (such as standing, walking or running) provides an important input variable that must be recognized quickly or in real time. Mobile sensors, e.g. in smartphones, and machine learning algorithms are used for this purpose.

The aim of the FastAR project is the fundamentally new research and validation of methods for the evaluation and optimization of the recognition time and accuracy of human activities, compared to the usual focus in the research field on the optimization of recognition accuracy, which thus makes a fundamental contribution to the above-mentioned application possibilities.