Autonomous Navigation for Flying Robots

What it is about

You will learn the fundamentals of autonomous navigation of quadrotors: from 3D geometry and probabilistic state estimation to visual odometry, SLAM and control engineering. In the end you will be able to estimate the position of a quadrotor from sensor data, implement an EKF and PID controller, and steer the quadrotor in simulation along a trajectory.

What you will be able to do

• Explain flight principles of quadcopters and application possibilities
• Fly a quadcopter safely in simulation
• Specify the position of a three-dimensional body and compute relative pose
• Explain the principle of Bayes' state estimation
• Implement, apply and parameterize Extended Kalman Filter (EKF)
• Implement PID controller and tune it correctly
• Understand fundamentals of visual motion estimation and mapping

What the module consists of

• VideovorlesungenDelivery of theoretical foundations in weekly videos
• Interaktive Übungs- und ProgrammieraufgabenApplication of the content; implementation of Python scripts for the simulator
• SimulationsexperimentePractical flight experiments in the browser-based quadrotor simulator

Teaching method

• MOOC / Videovorlesungfor theory delivery in asynchronous form
• Experimente (Simulation)to test practical skills and algorithms
• Einzelarbeit (Hausaufgaben)to consolidate what was learned through programming tasks