Event
Paper on VORL-EXPLORE Multi-Robot Exploration Framework Released
1 reports · 1 sources
Ning Liu and co-authors submitted a preprint proposing the VORL-EXPLORE multi-robot exploration framework. The framework uses "execution fidelity" to estimate local traversability and execution reliability, and couples it into a Voronoi allocation objective with inter-robot repulsion, arbitrating between global A* guidance and a reactive reinforcement learning policy during navigation; the fidelity model is recalibrated with self-supervised pseudo-labels generated from recent progress and safety outcomes. The authors report that experiments in stochastic dynamic grids and an ablation under severe congestion show higher success rates, shorter exploration paths, and lower inter-robot coverage overlap, and they provide a deployment example in a Gazebo factory scenario; the code has been open-sourced. The paper has been accepted to IROS 2026. All the above results are as reported by the authors and await independent verification.
Report timeline
- VORL-EXPLORE: A Hybrid Learning Planning Approach for Multi-Robot Exploration in Dynamic Environments
The authors propose VORL-EXPLORE, a hybrid learning and planning framework that uses execution fidelity to estimate local navigability and execution reliability, coupling it into a Voronoi allocation objective with inter-robot repulsion and arbitrating during navigation between global A* guidance and a reactive reinforcement learning policy. The fidelity model is recalibrated using self-supervised pseudo-labels generated from recent progress and safety outcomes. Experiments on randomized dynamic grids and an ablation under severe congestion show higher success rates, shorter exploration paths, and lower inter-robot coverage overlap, along with a deployment example in a Gazebo factory scenario; the code is open-sourced.
arXiv cs.RO ·