# ConceptBot: Enhancing Robot's Autonomy through Task Decomposition with Large Language Models and Knowledge Graph

**Authors:** Alessandro Leanza, Angelo Moroncelli, Giuseppe Vizzari, Francesco Braghin, Loris Roveda, Blerina Spahiu

arXiv: 2509.00570 · 2025-09-03

## TL;DR

ConceptBot is a modular robotic planning framework that leverages Large Language Models and Knowledge Graphs to improve task understanding, risk-awareness, and success rates in unstructured environments without domain-specific training.

## Contribution

It introduces a novel integration of LLMs with knowledge graphs for task decomposition and risk-aware planning in robotics, outperforming existing methods like SayCan.

## Key findings

- Achieved 100% success on explicit tasks
- Maintained 87% accuracy on implicit tasks
- Reached 76% success on risk-aware tasks

## Abstract

ConceptBot is a modular robotic planning framework that combines Large Language Models and Knowledge Graphs to generate feasible and risk-aware plans despite ambiguities in natural language instructions and correctly analyzing the objects present in the environment - challenges that typically arise from a lack of commonsense reasoning. To do that, ConceptBot integrates (i) an Object Property Extraction (OPE) module that enriches scene understanding with semantic concepts from ConceptNet, (ii) a User Request Processing (URP) module that disambiguates and structures instructions, and (iii) a Planner that generates context-aware, feasible pick-and-place policies. In comparative evaluations against Google SayCan, ConceptBot achieved 100% success on explicit tasks, maintained 87% accuracy on implicit tasks (versus 31% for SayCan), reached 76% on risk-aware tasks (versus 15%), and outperformed SayCan in application-specific scenarios, including material classification (70% vs. 20%) and toxicity detection (86% vs. 36%). On SafeAgentBench, ConceptBot achieved an overall score of 80% (versus 46% for the next-best baseline). These results, validated in both simulation and laboratory experiments, demonstrate ConceptBot's ability to generalize without domain-specific training and to significantly improve the reliability of robotic policies in unstructured environments. Website: https://sites.google.com/view/conceptbot

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/2509.00570/full.md

## Figures

4 figures with captions in the complete paper: https://tomesphere.com/paper/2509.00570/full.md

## References

49 references — full list in the complete paper: https://tomesphere.com/paper/2509.00570/full.md

---
Source: https://tomesphere.com/paper/2509.00570