space-robots.orgpapers

Top papers of the week / 2026-W44

Inspection planning, distributed lunar SLAM, and gaze-driven teleoperation

Oct 26 – Nov 1, 2026 / curated by Space Robots

IAC season brings a cluster of on-orbit servicing planners built on reinforcement learning and POMDPs. On the surface side: a field-tested multi-robot SLAM stack for planetary missions, plus new tools for rover adaptation and lunar digital twins.

  1. #1Risk-Sensitive Mission Planning for Multi-Target On-Orbit Inspection via Distributional Reinforcement Learning

    Adam AbdinIAC 2026

    A QR-DQN agent picks which target a low-thrust servicer should inspect next in LEO. Because it learns the full return distribution, operators can trade expected value against risk at each step.

    PDF HTML arXiv
  2. #3SKiD-SLAM: Resource-Aware Distributed Multi-Robot LiDAR SLAM for Planetary Missions with Field Deployments at Analog Sites

    Hogyun Kim, Juwon Kim, Dongjin Cho, Jiwon Choi +6 moreJFR / Field Robotics 2026

    Distributed multi-robot LiDAR SLAM sized to the communication and compute budgets of real planetary missions. Deployed at analog sites for both cave and surface exploration concepts.

    PDF HTML arXiv
  3. #5Explore Where It Matters: Targeted Exploration for Efficient Rover Adaptation

    Jacob Levy, Arthur Bouton, Tristan D. Hassler, Hari Nayar +1 moreCoRL 2026

    TERRA adapts a simulation-trained world model on deformable terrain by seeking out deployment data where task-relevant model errors are likely. Fewer trial-and-error drives for planetary rovers.

    PDF HTML arXiv
  4. #8A Reproducible Reconstruction and Evaluation Framework for Photorealistic Lunar Analog Scene Reconstruction Toward Digital Twins

    Antar Mazumder, Ryan Hartzell, Luke Boyd, Alexander Robinson +3 morePreprint 2026

    Benchmarks six 3D Gaussian Splatting variants on lunar analog scenes. Traversal effects such as motion blur and lofted-regolith haze cause digital-twin accuracy to diverge from test-set novel-view metrics.

    PDF HTML arXiv

Previous weeks

Curators edit content/weekly-picks.json. Picks are also published as JSON and RSS feeds.