Altera Targets Physical AI Market With FPGAs for Nvidia Sensor Bridge

What happened

Lee Nam-hoon, a director at Altera, said at the "Semiconductors That Power Humanoids" conference held alongside the Semiconductor Exhibition (SEDEX) 2026 at COEX in Seoul on Oct. 16 that Altera is targeting the physical AI market, including robotics and autonomous vehicles, with FPGAs designed for Nvidia's Holoscan Sensor Bridge (HSB). Altera developed the Holoscan Sensor Bridge with Nvidia to connect numerous sensors, including cameras, lidar, radar and ultrasonic sensors, to GPU platforms. The architecture connects sensors, FPGAs, Ethernet and GPUs in sequence: the FPGA aggregates data from sensors using different communication protocols and handles latency-sensitive operations in real time, including positioning, motor control and image preprocessing; the aggregated data is converted into Ethernet packets and transmitted directly to Nvidia GPU memory, bypassing the CPU to reduce latency, while a single high-speed Ethernet connection simplifies wiring and the approach reduces the limitations of the short transmission distances supported by MIPI. Altera offers four FPGA families for the HSB: the entry-level Cyclone, Agilex 3 for compact robots, Agilex 5 for next-generation robots including humanoids, and the higher-end Agilex 7; developers can select chips according to robot size, sensor count and data bandwidth. The company also supplies two modular development kits for Agilex 5-based reference designs and is working with partners including Terasic and Critical Link to develop an ecosystem of dedicated sensor boards, including system-on-modules (SOMs).

Impact and signal

Affected components and suppliers

Source and time

Source: The Elec · 10/20/2026 · View original source