Open Source Robotics

Una raccolta curata di progetti, strumenti e software di robotica open source di alta qualità per la comunità della robotica.

SLAM Course

  • Erogato da: Leibniz University Hannover
  • Docente: Prof. Claus Brenner
  • Livello: intermediate
  • Formato: video-lectures
  • Durata: Self-paced (77 videos, 7 units)
  • Categoria: SLAM
  • Lingua: English
  • Prezzo: Gratuito
  • Tag: SLAM, localization, mapping, Kalman-filter, EKF, particle-filter, FastSLAM, path-planning, Python

Un MOOC completo sulla localizzazione e mappatura simultanee (SLAM) per la robotica. Strutturato in 7 unità con 77 videolezioni, il corso procede dai modelli di moto del robot e dai dati dei sensori ai filtri bayesiani, ai filtri di Kalman, a EKF-SLAM e allo SLAM con filtro particellare (FastSLAM), per concludersi con algoritmi di pianificazione del percorso tra cui Dijkstra e A*. Comprende esercizi di programmazione in Python per implementare in pratica tutti gli algoritmi. È ampiamente considerato una delle migliori risorse gratuite per imparare lo SLAM.

Perché seguirlo

SLAM is one of the most important capabilities for any autonomous mobile robot, and this course is widely regarded as the best free resource for learning it from scratch. With 77 video lectures and Python exercises, it takes you from basic robot motion models all the way to implementing complete SLAM algorithms. The hands-on Python coding makes abstract concepts concrete and immediately applicable.

Cosa imparerai

  • Robot motion models and sensor measurement models
  • Bayes filter framework for recursive state estimation
  • Kalman filters and Extended Kalman Filters for localization
  • EKF-SLAM: simultaneous localization and mapping with Kalman filters
  • Particle filters and FastSLAM for non-linear SLAM problems
  • Path planning algorithms: Dijkstra's and A* search

Programma

Seven structured units with 77 video lectures total. Progresses logically from robot motion and sensor models through increasingly sophisticated estimation techniques. Each unit includes Python programming exercises where you implement the algorithms from scratch. Starts with basic motion models, builds through Bayes filters and Kalman filters, tackles full SLAM with both EKF and particle filter approaches, and concludes with path planning.

Argomenti principali

  • Robot motion models and odometry
  • Sensor data and scan matching
  • Bayes filter and histogram filter
  • Kalman filter and EKF localization
  • EKF-SLAM with landmark-based maps
  • Particle filter and FastSLAM
  • Path planning (Dijkstra, A*)

Prerequisiti

  • Basic Python programming
  • Linear algebra fundamentals (matrices, vectors)
  • Basic probability (Bayes rule, Gaussian distributions)

A chi si rivolge

  • Robotics hobbyists building autonomous mobile robots
  • Engineers implementing navigation on ROS-based platforms
  • Students taking their first deep dive into SLAM
  • Anyone who learns best through hands-on coding exercises

Link

  • Sito web: https://www.youtube.com/playlist?list=PLpUPoM7Rgzi_7YWn14Va2FODh7LzADBSm
  • GitHub: https://github.com/Passant-Abdelgalil/SLAM-Course-Codes

Questo sito richiede JavaScript. Attiva JavaScript per esplorare i nostri progetti di robotica open source.