AI-powered vision and compute solutions accelerated by NVIDIA Jetson Thor

e-con Systems has announced support for the newly launched NVIDIA Jetson Thor series, covering a comprehensive range of vision solutions, including the USB series, RouteCAM GigE series, 10G Holoscan Camera solutions, and a robust ECU platform purpose-built for real-time edge AI applications. This strategic expansion enables high-performance, AI-accelerated edge imaging solutions across industries.

NVIDIA Jetson Thor modules deliver up to 2070 FP4 TFLOPS of AI compute, delivering a high-performance platform for next-generation robotics and AI-enabled machines. e-con’s vision portfolio fully supports Thor’s capabilities with multi-sensor fusion, ultra-low latency, and resolutions up to 20MP – accelerating development in industrial, automotive, medical, and humanoid robotics applications.

As part of its NVIDIA Jetson Thor support, the company is launching a versatile range of camera solutions across key interface options:

• See3CAM USB Cameras for rapid development and evaluation
• RouteCAM Ethernet Cameras with ONVIF compliance
• Compact 10G e-con HSB leveraging Camera Over Ethernet (CoE) Powered by its camera modules and custom-designed HSB board with FPGA-based TintE ISP, this solution supports ultra-low latency and various sensors up to 20MP resolution. Camera over Ethernet (CoE) enables direct data transfer to GPU memory with virtually zero CPU utilisation, allowing for quicker response times and real-time operation.

e-con Systems supports a wide range of sensors, including the Sony IMX715, Sony IMX568, onsemi AR0234, Sony IMX900, AR2020, and Sony ISX031, across its Holoscan-ready solutions, empowering developers to fully utilise the NVIDIA Jetson Thor platform for next-generation AI vision applications.

Alongside its advanced camera offerings, the company introduces a compact, production-ready ECU platform with NVIDIA Jetson Thor. This robust compute unit is purpose-built to support real-time reasoning and sensor processing in the field, making it an ideal fit for AMRs, humanoids, and other AI-enabled machines. It ensures efficient processing of high-resolution imaging data, allowing faster decision-making and deployment in dynamic edge environments.

“We are excited to bring our latest portfolio of vision and complete compute solutions to the NVIDIA Jetson Thor platform. With support for multiple interfaces, real-time processing, and AI-enabled superior imaging, our Thor-compatible solutions empower developers to accelerate innovation in robotics, autonomous systems, and next-gen edge AI applications. Our compact e-con 10Gbit/s HSB board, featuring a proprietary TintE ISP, enables ultra-low latency imaging and multi-sensor flexibility while efficiently offloading image processing from the NVIDIA GPU—freeing up compute resources and optimising performance for AI-driven tasks across industries,” said Prabu Kumar Kesavan, CTO at e-con Systems.