XIT: Exploration and Exploitation Informed Trees for Active Gas Distribution Mapping in Unknown Environments

1. 05:00
   arXiv Robotics — research abstractsSignal

   XIT: Exploration and Exploitation Informed Trees for Active Gas Distribution Mapping in Unknown Environments

   This research presents XIT, a sampling-based planner designed for autonomous mobile robotic gas distribution mapping in unknown environments. The system addresses the challenge of simultaneously exploring traversable space, mapping the environment, and inferring gas distribution from sparse measurements by formulating active mapping as a next-best-trajectory informative path planning problem. Key contributions include the development of XIT, which uses an Upper Confidence Bound information field to balance travel effort with information acquisition, and the introduction of the 'gas frontier' concept alongside the Wavefront Gas Frontier Detection (WGFD) algorithm for identifying unobserved regions near high gas concentrations. High-fidelity simulations and a scaled real-world experiment demonstrate the method's benefits in terms of mapping quality, efficiency, and practicality.

Current attention 3·Peak within the comparable range 4（2026-11-10 13:00 UTC）·Change within the comparable range over 24 hours –

The trend compares participants observed continuously; its scope may be narrower than current attention totals. Hover or click to explore hourly observations, or use the left and right arrow keys.