Active Inference and Behavior Trees for Reactive Action Planning and Execution in Robotics

Journal Article (2023)
Author(s)

C. Pezzato (TU Delft - Robot Dynamics)

Carlos Hernández Hernández Corbato (TU Delft - Robot Dynamics)

S.D. Bonhof (TU Delft - Robot Dynamics)

Martijn Wisse (TU Delft - Robot Dynamics)

Research Group
Robot Dynamics
Copyright
© 2023 C. Pezzato, Carlos Hernández, S.D. Bonhof, M. Wisse
DOI related publication
https://doi.org/10.1109/TRO.2022.3226144
More Info
expand_more
Publication Year
2023
Language
English
Copyright
© 2023 C. Pezzato, Carlos Hernández, S.D. Bonhof, M. Wisse
Research Group
Robot Dynamics
Issue number
2
Volume number
39
Pages (from-to)
1050-1069
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

In this article, we propose a hybrid combination of active inference and behavior trees (BTs) for reactive action planning and execution in dynamic environments, showing how robotic tasks can be formulated as a free-energy minimization problem. The proposed approach allows handling partially observable initial states and improves the robustness of classical BTs against unexpected contingencies while at the same time reducing the number of nodes in a tree. In this work, we specify the nominal behavior offline, through BTs. However, in contrast to previous approaches, we introduce a new type of leaf node to specify the desired state to be achieved rather than an action to execute. The decision of which action to execute to reach the desired state is performed online through active inference. This results in continual online planning and hierarchical deliberation. By doing so, an agent can follow a predefined offline plan while still keeping the ability to locally adapt and take autonomous decisions at runtime, respecting safety constraints. We provide proof of convergence and robustness analysis, and we validate our method in two different mobile manipulators performing similar tasks, both in a simulated and real retail environment. The results showed improved runtime adaptability with a fraction of the hand-coded nodes compared to classical BTs.

Files

Active_Inference_and_Behavior_... (pdf)
(pdf | 2.61 Mb)
- Embargo expired in 02-07-2023
License info not available