In the „KI-FLEX“ project, the project partners are developing a powerful, energy-efficient hardware platform and the associated software framework for autonomous driving. The „KI-FLEX“ platform will process and combine data from laser, camera and radar sensors in the car reliably and quickly. Methods of artificial intelligence (AI) are used for this purpose. The vehicle thus has a precise image of the real traffic conditions at all times, can locate itself in this environment and makes the right decision in every driving situation on the basis of this information. This makes autonomous driving safe and reliable.
For the necessary AI inference and signal processing tasks, an application adapted hardware architecture, the FlexAISIC, is developed by members from the Fraunhofer IIS and the Chair of Computer Architecture at FAU.
Consisting of a fixed ASIC and a reconfigurable FPGA part, the platform is both energy-efficient and flexible to use.
While the task of designing the fixed function circuit is accomplished by the Fraunhofer team, the FAU is responsible for the necessary software environment built around the ML framework TensorFlow.
This environment provides control over the hardware as well as programming the FPGA, by principles of the HSA standard.
Fraunhofer Institute for Integrated Circuits IIS (consortium leader)
Ibeo Automotive Systems GmbH
Infineon Technologies AG
TU Munich (Chair of Robotics, Artificial Intelligence and Real-Time Systems)
Fraunhofer Institute for Open Communication Systems FOKUS
Daimler Center for Automotive IT Innovations (DCAITI), TU Berlin