• Understand the main challenges in mobile robotics • Understand what are mobile robots made of – Sub-systems (perception, mobility, navigation, planning, power, …) – Basic approaches and terminology for each sub-system – Methodology and algorithms in mobile robotics • Learn the characteristics in three applications areas 1. Indoor mobile service robot 2. Field Robot: Outdoor semiautonomous vehicle / heavy duty machine UGV 3. Autonomous vehicle, basics • Not processing, manipulation and grasping
Teaching: Lectures, Team work: Design of a case robot system, Indv.: Testing algorithms under ROS on Pioneer robot platform
Lectures (estimate) Nro topics 1 Course introduction and arrangements, Introduction to mobile robots 2 Robot control system architecture , Reactive robotics 3 Team work: Design of a case robot system Individual project: Testing algorithms under ROS/ Pioneer Introduction to ROS Locomotion, Kinematics, and Low level Motion Control 4 Sensing and perception, overview. - Kalman-filter based traditional vehicle localization 5 SLAM = Simultaneous Localization and Mapping, Extended Kalman Filter SLAM (EKF SLAM) 6 Particle filter and Fast SLAM 7 Graph SLAM 8 Satellite Positioning - GNSS technologies 9 Pose Estimation, Inertial Navigation Systems (INS), Inertial Measurement Unit (IMU) 10 Visual localization and object recognition 11 Graph based Path and Motion Planning, Indoor A*. Path and Motion Planning, Outdoor D* 12 Outdoor Path and Motion Planning Motion Control
Introduction to mobile robots
• What is a service robot? Field Robot? • Which kinds of service and field robots exist? – Examples • Mobile robot subsystems • Control architectures • More on Field robotics • Autonomous vehicles
Lecture 2. Control architectures. 03/03/2021 Kansio
Contents of the lecture
• Motivation and state-of-the-art issues
• Short introduction to control paradigms
• The Hierarchical Paradigm
• Biological foundations for the reactive paradigm
• The Reactive Paradigm
• The Hybrid Paradigm
+Hopefully some good examples and some inspiration
• Characterizing sensors • Classification of sensors • Sensor types for mobile robots • Perception in mobile robotics
Lecture 5, March 16, 2021: Localization, mapping and SLAM = Simultaneous Localization and Mapping, idea; EKF SLAM Kansio
Localization, mapping and SLAM =
Simultaneous Localization and Mapping, idea
- Main material as subset of ETH slides - Kalman-filter based traditional vehicle
localization in the tractor-trailer -case.
Extended Kalman Filter SLAM (EKF SLAM)
Save videos for running. A video haehnel-RawOdometry-anim.avi
demonstrates mapping with raw odometry localization - Just mapping, not
SLAM. The odometry errors accumulate quickly too large. A video haehnel-ScanMatching-anim.avi demonstrates first real
SLAM, localization on the basis of laser scan matching, only one estimate for
pose, Mapping works. EKF SLAM video by Nebot can be run in the 4_EKF_SLAM 2020.ppt -slide set.
Lecture 6, March 17, 2021: Particle filter and Fast SLAM Kansio
filters are introduced in particle-filters 2020.pdf.
distribution for robot pose, for each robot pose particle, separate
EKFs for positions of landmarks. Own realization of grid type maps
You can run videos in the fastslam 2020.ppt -slide set. pdf available, too.
SLAM videos added as separate files. In order to run videos, save them.