Self Driving and ROS 2 - Learn by Doing

Course 1 - Odometry and Control

https://www.udemy.com/course/self-driving-and-ros-2-learn-by-doing-odometry-control/

• Differential kinematics (control): translating velocity commands from a joystick into movement

• Odometry: inverse problem, estimating robot movement from encoder signals

• Sensor fusion: improving odometry by fusing data from various sensors to reduce noise (Kalman filters)

Notes on Course 1

• Notes on Section 2: Setup
• Notes on Section 3: Introduction to ROS2. Why ROS2?
• Notes on Section 4: Locomotion. URDF
• Notes on Section 5: Control
• Notes on Section 6: Kinematics
• Notes on Section 7: Differential Kinematics
• Notes on Section 8: TF2 Library
• Notes on Section 10: Probability for Robotics
• Bonus: Notes on building the real bumperbot

Code examples for Course 1

• Python code examples
• C++ code examples

Course 2 - Mapping and Localization

https://www.udemy.com/course/self-driving-and-ros-2-learn-by-doing-map-localization/

• 2D laser sensor:
• Mapping: mapping with a laser scanner assuming we know where the robot is
• Localization: assuming we have a map
• SLAM: simultaneous localization and mapping

Course 3 - Planning and navigation

• Path planning: comparing different algorithms to plan trajectories to reach a desired location in the map
• Obstacle avoidance: sense the environment constantly and adjust trajectory if there is an unexpected obstacle
• Behaviour trees: more complex logics for flexible navigation techniques

Course 4 - Vision and Perception

• visual odometry: using camera for localization
• visual SLAM:
• object recognition and tracking: detection of moving obstacles using a camera