Powering Next-Generation Automotive Systems
As the automotive market builds highly automated and fully self-driving vehicles, there is a growing demand for advanced sensor capabilities and domain controllers with artificial intelligence to control the vehicle. Visit the Xilinx booth to see how this is being accomplished withthe adaptable and intelligent Xilinx Automotive (XA) product portfolio.
| Demo Title | Description | Xilinx Technology Platform |
|---|---|---|
| YOLOv5 real-time AI inference processing with multiple cameras | Seamlessly port Neural Networks trained on GPUs with less than 0.5% accuracy degradation. Enabled by one Xilinx 7nm process product and Mipsology's Zebra CNN accelerator. | Versal ACAP |
| Model Predictive Control using accelerator IP for Automotive Systems | A demonstration system for model predictive control, which uses plant models and meta-heuristic solvers to achieve optimal control. The NSITEXE accelerator IP DR1000C offloads heavy load processing and is ready-certified for ASIL D by SGS-TÜV. | Virtex UltraScale+ FPGA |
| Dynamic Function eXchange (DFX)Framework | The multi-camera vision system demonstrates DFX-enabled feature swapping by reconfiguring parts of a continually operating programmable SoC chip. | Zynq UltraScale+ MPSoC |
| Occupancy Monitoring System | Simultaneous monitoring of driver and passenger using a wide-angle RGB-IR image sensor by Seeing Machines on Xilinx Zynq®-7000 based FOVIO Driver Monitoring (FDM) chip. Demonstration in new Levorg, SUBARU. | Zynq-7000 SoC |
| 8M Pixel, Mono Vision, Forward Looking Camera - Proof of Concept | PoC combines Xilinx Zynq® UltraScale+™ MPSoC, OmniVision's 8M pixel imager and Motovis' deep learning networks, with demonstration in a running car. | Zynq UltraScale+ MPSoC |