RealSense and NVIDIA Collaborate to Accelerate Adoption of Physical AI

Strategic Insights -

RealSense, Inc. and NVIDIA have partnered to accelerate the adoption and capabilities of physical AI across humanoids and autonomous mobile robots (AMRs). This collaboration integrates RealSense's industry-leading AI depth cameras with NVIDIA's robotic platforms, including the NVIDIA Jetson Thor Series for robot run-time computing. The initiative aims to shorten time to market, discover new applications, and scale safely into production. RealSense's CEO, Nadav Orbach, stated that this partnership "cements RealSense's role as the perception platform of choice for AMRs and humanoids."

Key Takeaways:

  • RealSense's new RealSensetm D555 depth camera features a v5 Vision Processor and on-chip Power over Ethernet (PoE), delivering native NVIDIA Holoscan Sensor Bridge streaming and on-camera neural network to enable enhanced image post-processing.
  • The NVIDIA Jetson Thor Series delivers up to 2070 FP4 teraflops of AI compute, 7.5x higher than its predecessor, NVIDIA Jetson Orintm, and 3.5x greater energy efficiency.
  • The collaboration enables developers to integrate RealSense depth perception with NVIDIA's Isaac Sim, Jetson Thor, and Holoscan Sensor Bridge for real-time sensor fusion and ultra-low-latency data streaming.
  • This partnership positions RealSense and NVIDIA at the forefront of the industry's next wave of innovation, with the robotics market projected to become a multi-trillion-dollar opportunity.
  • Developers can now leverage fully validated system architectures optimized for NVIDIA Jetson Thor to ensure maximum performance for robotics and physical AI applications.
  • The initiative provides native integration and performance optimizations with NVIDIA Thor and Holoscan Sensor Bridge, accelerating the mainstream adoption of physical AI.

Statistics:

  • RealSense's new RealSensetm D555 depth camera features on-chip Power over Ethernet (PoE) ensuring high-fidelity perception data for humanoid robotics and beyond.
  • The NVIDIA Jetson Thor Series delivers up to 2070 FP4 teraflops of AI compute within a 130-watt power envelope.
  • NVIDIA Jetson Thor delivers up to 7.5x higher AI compute and 3.5x greater energy efficiency compared to its predecessor, NVIDIA Jetson Orintm.
  • The robotics market is projected to become a multi-trillion-dollar opportunity.

Sources:

  • RealSense, Inc., (2023) - No specific source text available
  • NVIDIA, (2023) - No specific source text available