HW/SW Co-design and Prototyping Approach for Embedded Smart Camera: ADAS Case Study

Abstract

In 1968, Volkswagen integrated an electronic circuit as a new control fuel injection system, called the “Little Black Box”, it is considered as the first embedded system in the automotive industry. Currently, automobile constructors integrate several embedded systems into any of their new model vehicles. Behind these automobile’s electronics systems, a sophisticated Hardware/Software (HW/SW) architecture, which is based on heterogeneous components, and multiple CPUs is built. At present, they are more oriented toward visionbased systems using tiny embedded smart camera. This visionbased system in real time aspects represents one of the most challenging issues, especially in the domain of automobile’s applications. On the design side, one of the optimal solutions adopted by embedded systems designer for system performance, is to associate CPUs and hardware accelerators in the same design, in order to reduce the computational burden on the CPU and to speed-up the data processing. In this paper, we present a hardware platform-based design approach for fast embedded smart Advanced Driver Assistant System (ADAS) design and prototyping, as an alternative for the pure time-consuming simulation technique. Based on a Multi-CPU/FPGA platform, we introduced a new methodology/flow to design the different HW and SW parts of the ADAS system. Then, we shared our experience in designing and prototyping a HW/SW vision based on smart embedded system as an ADAS that helps to increase the safety of car’s drivers. We presented a real HW/SW prototype of the vision ADAS based on a Zynq FPGA. The system detects the fatigue/drowsiness state of the driver by monitoring the eyes closure and generates a real time alert. A new HW Skin Segmentation step to locate the eyes/face is proposed. Our new approach migrates the skin segmentation step from processing system (SW) to programmable logic (HW) taking the advantage of High-Level Synthesis (HLS) tool flow to accelerate the implementation, and the prototyping of the Vision based ADAS on a hardware platform.