Efficient Learning-Based Robotic Navigation Using Feature-Based RGB-D Pose Estimation and Topological Maps
Authors:
- Eder A. Rodríguez-MartínezEREder A. Rodríguez-Martínez
- Jesús Elías Miranda-VegaJesús Elías Miranda-Vega
- Farouk AchakirFAFarouk Achakir
- Oleg SergiyenkoOleg Sergiyenko
- Julio C. Rodríguez-QuiñonezJulio C. Rodríguez-Quiñonez
- Daniel Hernández BalbuenaDBDaniel Hernández Balbuena
- Wendy Flores-FuentesWendy Flores-Fuentes
Open Access
Abstract
Robust indoor robot navigation typically demands either costly sensors or extensive training data. We propose a cost-effective RGB-D navigation pipeline that couples feature-based relative pose estimation with a lightweight multi-layer-perceptron (MLP) policy. RGB-D keyframes extracted from human-driven traversals form nodes of a topological map; edges are added when visual similarity and geometric–kinematic constraints are jointly satisfied. During autonomy, LightGlue features and SVD give six-DoF relative pose to the active keyframe, and the MLP predicts one of four discrete actions. Low visual similarity or detected obstacles trigger graph editing and Dijkstra replanning in real time. Across eight tasks in four Habitat-Sim environments, the agent covered 190.44 m, replanning when required, and consistently stopped within 0.1 m of the goal while running on commodity hardware. An information-theoretic analysis over the Multi-Illumination dataset shows that LightGlue maximizes per-second information gain under lighting changes, motivating its selection. The modular design attains reliable navigation without metric SLAM or large-scale learning, and seamlessly accommodates future perception or policy upgrades.Keywords
Funding Information
- Mexican Ministry of Science, Humanities, Technology and Innovation (SECIHTI)
This publication has 85 references indexed in Scilit:
- NavTopo: Leveraging Topological Maps for Autonomous Navigation of a Mobile Robot2024
- ExplORB-SLAM: Active Visual SLAM Exploiting the Pose-graph Topology2022
- A comprehensive overview of dynamic visual SLAM and deep learning: concepts, methods and challengesMachine Vision and Applications, 2022
- Visual language navigation: a survey and open challengesArtificial Intelligence Review, 2022
- Object Goal Visual Navigation Using Semantic Spatial Relationships2021
- Learning to Plan with Uncertain Topological Maps2020
- Visual teach and repeat for long‐range rover autonomyJournal of Field Robotics, 2010
- Rigid body dynamics in terms of quaternions: Hamiltonian formulation and conserving numerical integrationInternational Journal for Numerical Methods in Engineering, 2009
- A note on two problems in connexion with graphsNumerische Mathematik, 1959
- A Mathematical Theory of CommunicationBell System Technical Journal, 1948