Strategic Objectives
• Master the algorithms that solve the infamous 'Kidnapped Robot Problem'.
• Understand the mechanics of visual and LiDAR-based place recognition.
• Learn to implement robust database retrieval systems for massive spatial datasets.
• Eliminate long-term odometry drift to create perfectly consistent global maps.
The Core Challenge
Autonomous systems inevitably suffer from 'drift'—a slow accumulation of positioning errors that turns a precise map into a tangled mess of digital hallucinations.
01
Foundations of Spatial Awareness
02
The Kidnapped Robot Problem
03
Probabilistic Navigation
04
Visual Feature Extraction
05
The Bag of Words Model
06
Invariant Keypoints
07
Efficient Binary Descriptors
08
Geometric Verification
09
The RANSAC Algorithm
10
Pose Graph Optimization
11
Bayesian Filtering
12
Monte Carlo Localization
13
LiDAR-Based Recognition
14
Visual Odometry Constraints
15
Information Theory in Robotics
16
Appearance-Based Mapping
17
The Kalman Filter Evolution
18
Robust Cost Functions
19
Deep Learning for Descriptors
20
Semantic SLAM
21
Large-Scale Database Management
Explore Research Ecosystem
Available eBook Editions
Arabic
English
French
German
Italian
Japanese
Korean
Portuguese
Spanish
