Localization in a known map
Particle filter based localization using a previously built map. Each black dot is a possible pose of the quadrotor; the green dot with the red laser beams marks the current best pose estimate.
Publications
Quadrotor Videos
Only a few navigation techniques have been designed for indoor environments, where the systems cannot rely on GPS and therefore have to use their exteroceptive sensors for navigation. The videos below demonstrate a general navigation system that enables a small-sized quadrotor to autonomously operate indoors, by systematically extending and adapting techniques successfully applied on ground robots.
Related documents
PhD Thesis (2011) TRO Journal (2012) RSS Workshop (2011) ICRA (2009)
Click a video to play it directly in your browser.
SLAM and pose stability
A small example of simultaneous localization and mapping (SLAM) combined with pose stability during autonomous hovering.
Multi-level SLAM
Extending SLAM towards the z-direction to map obstacles in height, resulting in a multi-level SLAM in an office environment.
Path planning and obstacle avoidance
Autonomous flight including path planning and dynamic obstacle avoidance.
Fuel-cell powered flight
Autonomous flight of the system powered by a TU-Berlin fuel cell driving the on-board modules (Gumstix, Hokuyo laser, XSens IMU).