Miles Jamison 
- Intel and Altera are building a real-time sensor bridge for OEMs
- The collaboration focuses on aerospace and defense, healthcare, and robotics markets
- The companies will pair Intel compute processors and connectivity with Altera FPGA technology
Intel and Altera are collaborating on a technology designed to help original equipment manufacturers, or OEMs, manage data collection and processing in increasingly complex, sensor-heavy edge environments.
The companies said Tuesday they are combining Intel compute, connectivity and software with Altera field-programmable gate arrays to create a programmable, real-time sensor bridge for applications in healthcare, defense, aerospace and robotics.
What Challenges Will the Sensor Bridge Address?
Intel and Altera plan to use modular reference designs to accommodate the varying hardware and processing requirements of edge and physical artificial intelligence applications while addressing performance, timing, power and cost considerations.
The designs will combine programmable components with Intel’s computing and software technologies, giving OEMs flexibility to configure systems for different sensors and workloads.
These applications can involve high-bandwidth data transfers and hardware arrangements that separate input/output connections from the main computing board. Sensor information must also move into memory quickly enough for Intel processors to access it and support edge, robotics and other workloads.
Dan Rodriguez, corporate vice president and general manager of Intel’s edge computing group, said the collaboration will address customers’ need for a simpler way to turn real-world sensor data into timely action as sensor use expands.
Which Markets Will Intel & Altera Target?
In aerospace and defense, the planned designs will address the hardware constraints of mission systems that combine AI and signal processing, including size, weight and power requirements.
The healthcare effort addresses latency and real-time processing requirements for advanced medical equipment. In robotics, the companies intend to use Intel’s computing, connectivity and robotics software alongside Altera FPGAs to support the sensing and control functions of intelligent machines.
Developers will be able to access the capabilities through Intel’s Open Edge Platform SDK and an Altera SDK.
Intel and Altera also aim to help OEMs move from prototyping to deployment with reference designs that can accommodate changing applications, sensors and workloads.
Farhad Shafai, head of business solutions and marketing at Altera, said the company’s FPGAs provide sensor connectivity, data processing and deterministic control. He added that the chips “stay programmable as sensors and AI models evolve.”
What Other Edge AI & Defense Work Have Intel & Altera Pursued?
In July, Intel Foundry completed the Department of War’s Rapid Assured Microelectronics Prototypes-Commercial, or RAMP-C, program, which seeks to establish leading-edge semiconductor manufacturing in the U.S. The War Department selected Intel Foundry in April 2024 for the program’s third phase, which supported the tapeout and testing of early defense prototypes. Participants used Intel 18A, which Intel said helped defense customers meet size, weight and power constraints.
Intel named its Secure Enclave program as the next step for RAMP-C customers, with the goal of providing trusted manufacturing of leading-edge chips for the U.S. government at production volumes.
In June, Altera introduced its Agilex 9 Direct RF-Series SoC FPGA for aerospace, defense and advanced communications applications, including radar and RF systems.
In April, the company released version 26.1.1 of its FPGA AI Suite with a spatial compiler that maps AI models directly onto Agilex FPGAs. Altera said the tool supports deterministic, low-latency inference for robotics and autonomous machines running vision, video analytics, language model and sensor processing workloads. The suite works with PyTorch, TensorFlow and OpenVINO model optimization.
