Travail Easier to the Edge of AI with LIPSedge’s Robotic Vision

By Chad Cox

Production Editor

Embedded Computing Design

March 23, 2023


LIPS issued its LIPSedge F110 3DxAI Edge Accelerator centered on the NVIDIA Jetson AGX Xavier module, making it the original NVIDIA Jetson AGX Xavier-based PCIe PoE+ endpoint-mode edge accelerator. The accelerator enables the building of applications while delivering both x86 and NVIDIA Jetson architecture.

Luke Liu, CEO of LIPS Corporation said, "Processing multiple 3D camera streams in real time has always been a big challenge in the industry. LIPS is able to provide a low-latency and scale-up edge acceleration architecture that not only speeds up image processing but also AI inference.”

A single x86 system can scale-up and pre-process multiple 3D camera streams with low latency and perform AI inference while keeping the CPU utilization low. Users may implement up to three edge accelerators per single x86 system. The platform supplies easy installation, maintenance liabilities, and a lower cost for edge deployment.

“By enabling NVIDIA Jetson running PCIe end-point mode on x86, we are also providing a next-generation building block to drive new innovations in edge-AI," continued Liu.

Highlights include:

  • Industrial-ready, Jetson AGX Xavier-based, PCIe-based edge accelerator for 2.5 GigE/ PoE RGB-D camera with GenICam
  • Features 2 robust 2.5G PoE+ ports with PSE (up to 30 watts)
  • Run at PCIe endpoint mode
    • High-bandwidth data transfer with PCIe 4.0 up to 16GB/s
    • Low-latency direct data transmission to GPU via RDMA without consuming CPU resource
    • Features LIPS Cleverboot Interconnect for multi-OS systems without use of a hypervisor
    • Allows programmable edge computing with NVIDIA CUDA and OpenCL
  • Root Port Deployment from single lane to multi lanes
  • One jumper to switch between root port and endpoint mode
  • Features LIPS own 3D point cloud de-noise, stitching and fusion algorithms
  • Compatible with LIPSedge 3D Camera and SDK wrappers
    •  Extended library support of NVIDIA Isaac robotics software and latest NVIDIA JetPack SDK

For more information, visit

Chad Cox. Production Editor, Embedded Computing Design, has responsibilities that include handling the news cycle, newsletters, social media, and advertising. Chad graduated from the University of Cincinnati with a B.A. in Cultural and Analytical Literature.

More from Chad