Behavior Trees

Ask about this topic

Overview

Behavior trees (BTs) organize tasks into condition, action, and control-flow nodes that can react to changing conditions, recover from failures, and be inspected by humans. They are used to build modular robot and agent policies, with recent approaches generating trees from language or vision-language models and grounding them in available skills before runtime validation. Other methods synthesize BTs from temporal-logic specifications for formally constrained control, or use tree structure to shape rewards for reinforcement learning; evaluations span robot manipulation and navigation, BEHAVIOR-1K tasks, and planning benchmarks such as TravelPlanner. The central tension is balancing flexibility and adaptation to uncertainty with reliable execution: generated trees still need safety checks, and classical BTs can struggle with learning, runtime restructuring, and limited tooling.

Latest papers

CardsList

19 papers3 papers in the last four weeks, down 25% on the four weeks before. 0.0% of all new papers.

  1. Self-Crystallizing Behavior Trees: Amortizing Foundation-Model Reasoning into Deterministic Control for Autonomous Drones

    Oct 18, 2026Behnam MoradiBehavior TreesRobot Policy Adaptation

    Software Department, KavAI. California, USA

  2. BT-Agent: Deploying Onboard Robot Agents with Behavior Trees and Multi-Fidelity Evaluation

    Oct 9, 2026Fabio Hübel, Adam Lilja, Yue Yin +5Large Language Model-Based Robot PlanningMobile Robotics

    Stanford University · University of California, Berkeley · NVIDIA

  3. SharedKV-BT: Node-Local Typed Decisions for Behavior-Tree Agents

    Oct 5, 2026Naoki Wake, Justin WagleLanguage Model DecodingRobot Task Planning

    Microsoft

  4. Towards An LLM-Driven Unified Conversion Framework for BT and FSM in Autonomous Intelligent Systems

    Sep 24, 2026Zhang Qi, Yang Shuo, Zhu Zhengqiu +2Finite-State MachineBehavior Trees

    College of Systems Engineering, National University of Defense Technology CHANGSHA, HUNAN, China

  5. Behavior Trees for Robotic Systems: An Empirical Study on Practices and Experiences

    Sep 18, 2026Razan Ghzouli, Jennifer Horkoff, Daniel Strüber +1Behavior TreesRobotics

    Department of Computer Science and Engineering, Chalmers University of Technology and University of Gothenburg, Gothenburg, Sweden · Department of Software Science, Radboud University, Nijmegen, The Netherlands · Software and Societal Systems Department, Carnegie Mellon University, Pittsburgh, PA, USA

  6. Adaptation Needs in Robotic Systems: Assessing Behavior Trees and Their Enhancement

    Sep 4, 2026Mehran Rostamnia, Gianluca Filippone, Ricardo Caldas +1Behavior TreesRobotic Control

    Gran Sasso Science Institute (GSSI), L’Aquila, Italy

  7. Programming and execution of skill-based human-robot-crane collaborative tasks

    Sep 3, 2026Taneli Lohi, Markku Suomalainen, Roope Mellanen +1RoboticsHuman-Robot Collaboration

    Intelligent Robotics VTT Technical Research Centre of Finland Ltd Oulu, Finland · Konecranes oyj. Hyvinkää, Finland

  8. Correct-by-Construction Behavior Tree Synthesis from Signal Temporal Logic Specifications with Application to Robotic Missions

    Jul 21, 2026Jiaheng Dong, Jingyi Huang, Liang HanSignal Temporal LogicRobot Task Planning

    Sino-French Engineer School,Beihang University, Beijing 100191, China

  9. Contract-Grounded Behavior Tree Synthesis via Coding Agents

    Jul 13, 2026Jonathan Salfity, Robert Blake Anderson, Mitch PryorLarge Language Model-Based Robot PlanningLLM-Based Program Synthesis

    Nuclear and Applied Robotics Group, Department of Mechanical Engineering, The University of Texas at Austin

  10. A Reward-Petri-Net Interpretation of Temporal Behavior Trees

    Jun 19, 2026Till Schmeil, Günther Waxenegger-Wilfing, Sebastian SchirmerReinforcement LearningReward Shaping

    University of Würzburg, Germany · German Aerospace Center (DLR), Germany

  11. Coordinating Task Switching in a Robotics Multi-Agent System Using Behavior Trees

    May 31, 2026Lucas Haug, Anarosa Alves Franco Brandão, Arthur CasalsRoboticsMulti-Robot Coordination

    ∗LTI - Laborat´orio de T´ecnicas Inteligentes, Universidade de S˜ao Paulo, SP

  12. CommandSwarm: Safety-Aware Natural Language-to-Behavior-Tree Generation for Robotic Swarms

    May 8, 2026Mohammed Majid, Amjad Yousef MajidMulti-Robot SystemsSwarm Robotics

  13. Decoupled Travel Planning with Behavior Forest

    Apr 23, 2026Duanyang Yuan, Sihang Zhou, Yanning Hou +7LLM PlanningTravel Planning

    College of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, China · College of Computer Science and Technology, National University of Defense Technology, Changsha 410073, China · College of Systems Engineering, National University of Defense Technology, Changsha 410073, China +1

  14. Ternary Logic Encodings of Temporal Behavior Trees with Application to Control Synthesis

    Apr 13, 2026Ryan Matheu, John S. Baras, Calin BeltaBehavior TreesSignal Temporal Logic

    Institute for Systems Research, University of Maryland, College Park, USA.

  15. Learning Structured Robot Policies from Vision-Language Models via Synthetic Neuro-Symbolic Supervision

    Apr 3, 2026Alessandro Adami, Tommaso Tubaldo, Marco Todescato +2RoboticsVLMs for Robotics

    University of Padova, Dept. of Information Engineering, Italy. · Polytechnic of Bari Dept. of Electrical and Information Engineering, Italy. · Fraunhofer Italia Research, 39100, Bozen, Italy.

  16. VersualRL: Closed-Loop Verbal Reinforcement Learning with Visual Execution Feedback for Task-Level Robot Planning

    Mar 23, 2026Dmitrii Plotnikov, Iaroslav Kolomiets, Dmitrii Maliukov +9Robot Policy LearningLarge Language Model-Based Robot Planning

    The authors are with the Intelligent Space Robotics Laboratory, Skolkovo Institute of Science and Technology.

  17. CABTO: Context-Aware Behavior Tree Grounding for Robot Manipulation

    Mar 17, 2026Yishuai Cai, Xinglin Chen, Yunxin Mao +6Robotic ControlRobotic Manipulation

    National University of Defense Technology · PsiBot · PKU-Psibot Lab +1

  18. Formalisms for Robotic Mission Specification and Execution: A Comparative Analysis

    Mar 16, 2026Gianluca Filippone, Sara Pettinari, Patrizio PelliccioneBehavior TreesRobot Task Planning

    Gran Sasso Science Institute (GSSI), L’Aquila, Italy

  19. Multimodal Behavior Tree Generation: A Small Vision-Language Model for Robot Task Planning

    Mar 6, 2026Riccardo Andrea Izzo, Cristiano Battistini, Gianluca Bardaro +1Large Language Model-Based Robot PlanningSynthetic Data Generation for Robotics

    Department of Electronics, Information, and Bioengineering, Politecnico di Milano, Milan, Italy