20.500.12556/RUL-182457 Interagent beliefs for learning to communicate in large-scale multirobot visual object search We present Century Maze, a simulated environment for cooperative navigation of groups numbering over 100 robots with primarily visual observations. The underlying task of large-scale object search is designed to push the scalability and efficiency of decentralized multirobot systems relying on communication to overcome the limitations of separate units. To enable the training of robotic agents in large-scale visual embodied applications such as Century Maze, we propose two reinforcement learning (RL) components that facilitate the emergence of effective communication: Differentiable Interagent Belief Learning (DIABL), which provides a clear learning signal by optimizing the agents' beliefs about their goals and propagating gradients back over a differentiable communication channel to include the appropriate observers, and informative event replay (infer), which compensates for the scarcity of samples relevant to DIABL by maintaining a dedicated buffer for auxiliary learning. We demonstrate our combined method through an actor-critic neural network architecture encompassing a decentralized multiagent system of variable size with individual goal-based belief modules. Results on Century Maze show that our method is able to significantly improve performance through cooperation over 100 agents against comparable baselines and maintains its effectiveness in realistic conditions. cooperating robots multirobot systems reinforcement learning vision-based navigation sodelujoči roboti večrobotski sistemi spodbujevalno učenje vizualna navigacija true false true Angleški jezik Slovenski jezik Članek v reviji 2026-05-12 12:27:42 2026-05-12 12:27:49 2026-05-13 04:51:36 0000-00-00 00:00:00 2026 0 0 20 str. Vol. 42 2026 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 007.52 1941-0468 10.1109/TRO.2026.3677944 277860355 RAZ_Puc_Jernej_2026.pdf RAZ_Puc_Jernej_2026.pdf 1 6362C72405971F7529959D5E41F39136 e22832ce4d2cdaa3ae14186149549d40bd0831f9c852259bb8f3a72238bff305 3eb5e59a-4ded-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=233734 https://ieeexplore.ieee.org/document/11456614 1 ed96c3e9-4dec-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=233733 Fakulteta za strojništvo 0 0 0