Lattice Semiconductor has announced a collaboration with Texas Instruments aimed at accelerating the development of real-time edge artificial intelligence (AI) systems for robotics and industrial applications.
The partnership focuses on simplifying sensor integration and enabling scalable, low-latency data processing at the edge. By combining Texas Instruments’ sensing technologies with Lattice’s Holoscan Sensor Bridge solution—built on low-power FPGA architecture—the companies aim to provide developers with a flexible platform for managing synchronized sensor data pipelines.
The collaboration demonstrates a real-time AI sensor fusion architecture that integrates mmWave radar and camera sensors from Texas Instruments with NVIDIA’s Holoscan Sensor Bridge, running on Lattice’s FPGA technology. In this setup, the FPGA acts as a companion chip, synchronizing sensor inputs and delivering data directly into GPU-accessible memory, enabling faster processing and improved system responsiveness.
According to the companies, the approach addresses a growing need for deterministic, real-time performance in edge AI systems, particularly as applications in robotics and industrial automation become more complex. Low-latency perception is critical for tasks such as object detection, navigation, and safety monitoring in physical AI environments.
Executives from both companies emphasized the importance of combining sensing, compute, and data integration capabilities to support scalable deployment. The joint solution is designed to help developers transition more efficiently from early-stage evaluation to production-ready systems.
The initiative also aligns with a broader industry trend toward integrating multiple sensing modalities, such as radar and vision, to improve reliability and performance in AI-driven systems operating in dynamic environments.
Lattice Semiconductor noted that its Holoscan ecosystem is expanding through partnerships across sensor, compute, and software domains. The ecosystem is intended to support deterministic data movement, flexible connectivity, and energy-efficient operation, which are increasingly important for edge deployments.
The collaboration is expected to benefit developers working on next-generation robotics, industrial automation, and other emerging physical AI applications, where real-time decision-making and efficient sensor integration are critical to performance and scalability.
