Skip to content
Course Roadmap

Learning Roadmap

12 lessons across 4 chapters — click any node to open it, or toggle completion directly on the map.

0%0 / 12 lessons done
0
Introduction
0/1
0.1
Introduction to SLAM
Introduction to SLAM
What SLAM is, why it matters, and the high-level taxonomy of approaches.
1
Kalman Filters
0/4
1.1
Recursive Bayes Filter
Recursive Bayes Filter
The probabilistic foundation: belief, prediction, and correction.
1.2
Motion & Sensor Models
Motion & Sensor Models
Odometry, velocity, range-bearing and other models used inside filters.
1.3
Kalman & Extended Kalman Filters
Kalman & Extended Kalman Filters
Linear KF, linearization via Jacobians, the EKF prediction/update cycle.
1.4
EKF-SLAM
EKF-SLAM
Estimating the robot pose and landmark map jointly with an EKF.
2
Particle Filters
0/4
2.1
Occupancy Grid Maps
Occupancy Grid Maps
Representing the world as a grid of free / occupied cells via log-odds.
2.2
Monte Carlo Localization
Monte Carlo Localization
Approximating beliefs with weighted samples; sampling, weighting, resampling.
2.3
FastSLAM
FastSLAM
Factoring SLAM into a particle filter over poses plus per-particle landmark EKFs.
2.4
Grid-based SLAM (RBPF)
Grid-based SLAM (RBPF)
Rao-Blackwellized particle filter for SLAM with occupancy grids.
3
Least Squares & Graph-based SLAM
0/3
3.1
Nonlinear Least Squares
Nonlinear Least Squares
Gauss–Newton, Levenberg–Marquardt, and the normal equations.
3.2
Least-Squares SLAM
Least-Squares SLAM
Pose-graph constraints and global optimization of trajectories.
3.3
Landmark Graph-based SLAM
Landmark Graph-based SLAM
Joint pose + landmark graphs and information-matrix sparsity.
END