Industrial ReMEmbR: Scaling Spatiotemporal Memory to Robot Fleets

Key points

• Warehouse robot fleets continuously observe operational conditions, but preserving these observations as reusable semantic memory incurs substantial computation, communication, and storage costs.1
• On each robot, an NVIDIA Jetson encodes video clips into retrieval embeddings with Cosmos-Embed1, vision tokens with Cosmos3-Edge, and reconstruction latents with the Wan2.2 VAE, without uploading raw camera frames.1
• We split Cosmos3-Edge at its vision-language boundary, placing the vision encoder on the robot and the language decoder on an on-premises server.1
• A Nemotron agent retrieves memories by content, position, and time and queries the decoder using cached vision tokens.1

Sources (1)

• [1]Industrial ReMEmbR: Scaling Spatiotemporal Memory to Robot Fleets
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 20, 05:58 PM

  We present IndustrialRemembr, a shared spatiotemporal memory system that extends NVIDIA ReMEmbR to robot fleets and reduces server-side computation by generating scene descriptions only on demand.

  Warehouse robot fleets continuously observe operational conditions, but preserving these observations as reusable semantic memory incurs substantial computation, communication, and storage costs.

Extractive summary: sentences quoted from the sources.