CABTO: Context-Aware Behavior Tree Grounding for Robot Manipulation

2026-03-17

AI Summary

This work addresses the reliance of behavior tree (BT) systems on expert-designed high-level action models and low-level control policies by proposing CABTO, a novel framework that formalizes and efficiently solves the BT grounding problem for the first time. CABTO integrates prior knowledge from pretrained large language models, heuristic search, a BT planner, and contextual feedback from the environment to enable end-to-end automatic synthesis of complete and consistent behavior trees. Evaluated across three robotic manipulation scenarios and seven tasks, CABTO demonstrates significant effectiveness and efficiency, markedly reducing dependence on manual design. Its key contribution lies in introducing a context-aware mechanism that synergistically combines large-model reasoning with planning feedback to achieve automated BT construction.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: Behavior Learning & Control
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Behavior Trees (BTs) offer a powerful paradigm for designing modular and reactive robot controllers. BT planning, an emerging field, provides theoretical guarantees for the automated generation of reliable BTs. However, BT planning typically assumes that a well-designed BT system is already grounded -- comprising high-level action models and low-level control policies -- which often requires extensive expert knowledge and manual effort. In this paper, we formalize the BT Grounding problem: the automated construction of a complete and consistent BT system. We analyze its complexity and introduce CABTO (Context-Aware Behavior Tree grOunding), the first framework to efficiently solve this challenge. CABTO leverages pre-trained Large Models (LMs) to heuristically search the space of action models and control policies, guided by contextual feedback from BT planners and environmental observations. Experiments spanning seven task sets across three distinct robotic manipulation scenarios demonstrate CABTO's effectiveness and efficiency in generating complete and consistent behavior tree systems.

Problem

• Behavior Tree Grounding
• Robot Manipulation
• Automated Planning
• Action Models
• Control Policies

Innovation

• Behavior Tree Grounding
• Context-Aware Planning
• Large Models
• Robot Manipulation
• Automated Controller Synthesis