RU-SLAM: A Robust Deep-Learning Visual Simultaneous Localization and Mapping (SLAM) System for Weakly Textured Underwater Environments.
Zhuo WangQin ChengXiaokai MuPublished in: Sensors (2024)
Keyphrases
- simultaneous localization and mapping
- deep learning
- dynamic environments
- mobile robot
- visual slam
- kalman filter
- mobile robotics
- information filter
- particle filter
- robot navigation
- extended kalman filter
- data association
- indoor environments
- real environment
- loop closing
- object and scene recognition
- visual odometry
- unsupervised learning
- autonomous robots
- machine learning
- map building
- learning algorithm
- partial occlusion
- path planning
- single image
- topological map
- robotic systems
- image classification
- loop closure
- probabilistic model