Spot

Quadruped Robot

Spot is a quadruped robot provided by DARPA, serving as a core experimental platform for SAIR Lab’s research on real-world Spatial AI under mission-driven constraints. Rather than a lab-only system, Spot anchors our work in challenge-scale environments where robustness, repeatability, and recovery matter.

It is the platform where our methods are evaluated end-to-end: from perception and memory to planning and execution, across repeated deployments in the same evolving spaces.

Publications

  1. Point-Level Refinable Semantic Mapping for Spatial Understanding.
    Taimeng Fu, Yi Du, Zitong Zhan, Qiwei Du, Shaoshu Su, Zhipeng Zhao, Chen Wang.
    arXiv preprint arXiv:2610.31476, 2026.
    Continuously refinable semantic maps as spatial memory for embodied agents
  2. Neuro-Symbolic Learning for Long-Horizon Task Planning Under Complex Logical Constraints.
    Qiwei Du, Zitong Zhan, Shaoshu Su, Bowen Li, Yi Du, Zhipeng Zhao, Taimeng Fu, Sebastian Scherer, Jiaoyang Li, Chen Wang.
    arXiv preprint arXiv:2606.06877, 2026.
    Best Paper Award at IROS Workshop on Embodied Neuro-Symbolic AI for Reliable and Safe Robotics

Latest News