Robotic task execution faces challenges due to the inconsistency between symbolic planner models and the rich control structures actually running on the robot. In this paper, we present the first physical deployment of an integrated actor-planner system that shares hierarchical operational models for both acting and planning, interleaving the Reactive Acting Engine (RAE) with an anytime UCT-like Monte Carlo planner (UPOM). We implement RAE+UPOM on a mobile manipulator in a real-world deployment for an object collection task. Our experiments demonstrate robust task execution under action failures and sensor noise, and provide empirical insights into the interleaved acting-and-planning decision making process.
1.1 Task completion with high utility (box used)
1.2 Task completion with high utility (box unavailable)
2.1 Increased Object Spacing
3.1 Execution failure
(arm failure / cable entangled)
3.2 Execution failure
(navigation failure)
3.3 Execution failure
(perception failure)
@article{lima2025acting,
title={Acting and Planning with Hierarchical Operational Models on a Mobile Robot: A Study with RAE+ UPOM},
author={Lima, Oscar and Vinci, Marc and Patra, Sunandita and Stock, Sebastian and Hertzberg, Joachim and Atzmueller, Martin and Ghallab, Malik and Nau, Dana and Traverso, Paolo},
journal={arXiv preprint arXiv:2507.11345},
year={2025},
url={https://arxiv.org/abs/2507.11345},
}