Jump to: navigation, search
Mini-Course on SLAM: TU Kaiserlsautern, Germany

Summer 2012


Dr Abubakr Muhammad

Email: abubakr [at]


Office Hours:

Mini-Course Description

Introduction to basic concepts in robotics, Review of basic probability, random signals and systems, classical parameter estimation (min variance, least squares, max likelihood), Noisy sensor and motion models in robotics, concepts of state-space, Kalman filtering, Extended Kalman filter (EKF), robot configuration spaces, motion planning problems, Simultaneous Localization and Mapping (SLAM), Bayesian filtering, Monte-Carlo methods, particle filtering, SLAM variants; practical issues in autonomous robotics; examples/exercises from control, signal processing.


  • Using sensor-based information to determine robot’s own state and of the world.
  • Linear and non-linear statistical estimation techniques applied to problems in robotics and dynamical systems.
  • Introduce practical applications


The following textbooks will be helpful as reference material. The first text matches closely the presentation followed in this mini-course.

(Quick overview for first encounter with SLAM.)

(Contains details, derivations and state of the art techniques.)

(Mathematical background on stochastic systems.)

Reference Slides

The following are reference slides (companion with textbook by Choset) to help prepare before the mini-course. The instructor will post his own slides, exercises and material as the lectures proceed.

Similar courses

KAUST. (my course in Summer 2011)

LUMS. (my course in Fall 2010)



Centro de Investigacion en Matematicas, Mexico.

Univ of Washington.


June 05, 2012 Lecture. Introduction to probabilistic robotics; Basic laws of probability and Bayes rule; linear transformations of Gaussian random variables; interpreting and visualizing the covariance matrix; state and measurement;

Hands-on / Demo. Modeling sensor noise and process noise in robotic systems;

Choset CH7, CH8;
June 11, 2012

Lecture. Objectives of Bayesian & Kalman filtering; belief about state; derivation of the Bayesian filter; prediction and update; dealing with beliefs and distributions for nonlinear non-Gaussian systems; Kalman Filtering as a special case; Derivation of the Kalman filter; combining estimates from two scalar Gaussians; innovation and Kalman gain; estimation in the absence of process and sensor noise; state and covariance prediction; combining measurement and prediction in the presence of process noise;

Thrun CH 2; Choset CH8
June 12, 2012

Lecture. Basic SLAM using Kalman filtering; non-linear models of sensing and robot motion; Extended Kalman filtering; data association problem; Advanced techniques: Brief overview of occupancy grid maps and GraphSLAM algorithm.

Hands-on/Demo Setting up GraphSLAM for RGB-D SLAM.

Thrun CH 2,3,4; Choset CH8
Personal tools