A Comparative Study of Navigability and Cleaning Efficiency in Robot Vacuum Cleaners
This study evaluates 2025 robot vacuum models (WAP Robot W90, Positivo Smart Robô Aspirador Wi-Fi Laser PRA1000, and Xiaomi Mi Robot Vacuum-Mop 2 Lite) using random, LiDAR, and vSLAM navigation, respectively. It combines a controlled experimental study (featuring OpenCV-based heatmaps, area-coverage analysis, and debris-collection ratios) with an LLM-based sentiment and thematic analysis of user reviews on navigability and cleaning efficiency. The paper provides a repository that includes recorded videos, calibration data, analysis code, and LLM prompts, supporting full reproducibility of the reported metrics and enabling future research on robot vacuum navigation and cleaning performance.
Authors
- Gustavo B. Figueiredo (ORCID: https://orcid.org/0000-0001-9756-378X)
- Cássio Prazeres (ORCID: https://orcid.org/0000-0003-0197-0909)
- Frederico Araújo Dur�ão (ORCID: https://orcid.org/0000-0002-7766-6666)
- Eduardo Santana de Almeida (ORCID: https://orcid.org/0000-0002-9312-6715)
- Ivan Machado (ORCID: https://orcid.org/0000-0001-9027-2293)
- Danilo Barbosa Coimbra (ORCID: https://orcid.org/0000-0003-2218-1351)
- Maycon Peixoto (ORCID: https://orcid.org/0000-0002-4851-5228)
- Rodrigo Rocha Gomes e Souza (ORCID: https://orcid.org/0000-0001-8186-0069)
- Enio Garcia de Santana Junior (ORCID: https://orcid.org/0009-0000-7625-8583)
- Guilherme Castro Hora Fontes (ORCID: https://orcid.org/0009-0009-6316-6300)
Institutions
- Universidade Federal da Bahia (BR)
- Hospital Ana Nery (BR)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-08-20
- DOI
- https://doi.org/10.5281/zenodo.22030564
- Primary Topic
- Robotics and Sensor-Based Localization
- Type
- preprint