The PolarFire FPGA Ethernet Sensor Bridge, which is compatible with the NVIDIA Holoscan sensor processing platform, was launched by Microchip Technology to help developers create artificial intelligence (AI)-driven sensor processing systems.
The first solution to be offered as part of Microchip’s platform is compatible with MIPI® CSI-2®-based sensors and the MIPI D-PHY physical layer. PolarFire FPGAs allow for multi-protocol support.
A variety of sensors with various interfaces, such as SLVS-ECTM 2.0, 12G SDI, CoaXPress® 2.0, and JESD204B, will be supported by future solutions. The platform enables designers to take advantage of the power-efficient PolarFire FPGA technology with low-latency connectivity and multi-protocol sensor support, as well as the capabilities of the NVIDIA Holoscan ecosystem.
For real-time insights, NVIDIA Holoscan facilitates the creation and implementation of AI and high-performance computing (HPC) applications at the edge. It combines the hardware and software systems required for network connectivity and low-latency sensor streaming into a single platform.
The platform comprises basic microservices to run streaming, imaging, and other applications, template applications to speed up prototyping, sample AI models to jumpstart the creation of AI inference pipelines, and optimized libraries for data processing.
The PolarFire FPGA Ethernet Sensor Bridge opens up new edge-to-cloud applications, permits AI/ML inferencing, and speeds up the adoption of AI in the industrial, medical, and automotive sectors by bridging real-time sensor data to NVIDIA Holoscan and the NVIDIA IGX and NVIDIA Jetson platforms for edge AI and robotics.
The NVIDIA Holoscan Sensor Bridge effectively handles high-bandwidth data from various sensors across Ethernet by leveraging the low power consumption of Microchip’s PolarFire FPGA technology, allowing for real-time, high-performance edge AI processing on NVIDIA AI platforms. Applications that are cost- or energy-sensitive and have a tiny footprint benefit from the power-efficient design.
By including built-in security and safety features to assist guard against any cyberthreats and ensure physical, device, design, and data integrity, PolarFire FPGAs address security issues in sensor applications.
Additionally, they are built with single event upset (SEU) immunity, which makes them extremely dependable in radiation-prone areas like space or high-altitude applications and medical settings. Additionally, the SEU immunity lowers the chance of system failures and data corruption.
Key Comments
“The Ethernet sensor bridge is based on Microchip’s highly power-efficient, secure and reliable PolarFire FPGA platform,” said Bruce Weyer, VP of Microchip’s FPGA business unit. “By combining our flexible FPGA fabric with NVIDIA’s advanced AI platform and multi-protocol support, we’re empowering developers to create innovative, real-time solutions that will revolutionize sensor interfaces across a wide range of powerful AI-driven edge applications.”
For further info about how Microchip’s development tool supports NVIDIA Holoscan and other applications, visit the PolarFire FPGA Ethernet Sensor Bridge web page.