Low Power and Learning Wearables for the Next Generation of Footballers
The project FussbaLLLiebe has the goal to build a wearable prototype for the use in football. The wearable sensor module captures movements through an inertial measurement unit (IMU) and processes the data on the device with a neuromorphic processor. By designing spiking neural networks combined with classical physics-based algorithms, valuable insights for athletes should be provided. Leveraging neuromorphic technologies within the wearables aims to drastically reduce their power consumption and extend their battery lifetime.
The fortiss contribution to the project lies in developing spiking neural networks (SNNs) for movement-based event detection and activity recognition. Therefore, methods from literature and internal preliminary work are extended to the specific use-case. The wearable system will be optimized for the best accuracy and low power tradeoff. Together with our partners we aim at developing one of the first efficient prototypes which includes a neuromorphic processor. A specific focus lies in the development of methods for solving challenges along this task of research transfer.
Bavarian Ministry of Economic Affairs, Regional Development and Energy (StMWi)
Reference number: DIE0572/01
01.10.2025 - 31.09.2027