[ros-users] new ROS tutorials for vision based navigation for micro helicopters

Weiss Stephan stephan.weiss at mavt.ethz.ch
Mon Oct 8 08:44:57 UTC 2012


Dear ROS users,

We are pleased to announce new comprehensive tutorials about our code release concerning two ROS stacks for vision based navigation for micro helicopters (and other platforms). These ROS Stacks were successfully used in the European project sFly (www.sfly.org) to estimate a 6DoF pose for long-term vision based navigation of a micro helicopter in large outdoor environments. For corresponding video contributions at IROS see below.

www.ros.org/wiki/ethzasl_ptam<http://www.ros.org/wiki/ethzasl_ptam> : Visual odometry framework with constant computational complexity  running at 20 Hz on an ATOM 1.6 GHz single core computer and robust in self-similar scenes (e.g. grass, asphalt).
New: tutorials from camera calibration to framework usage

www.ros.org/wiki/ethzasl_sensor_fusion<http://www.ros.org/wiki/ethzasl_sensor_fusion> : EKF-based sensor-fusion framework for vehicle pose-estimation and intra- and inter-sensor self-calibration designed for robots (particularly micro helicopters) carrying an IMU and additional sensors (e.g. camera).
New: tutorial as introduction to the system;  tutorial on how to use your own sensors in the sensor-fusion stack; tutorial on how to use the sFly-framework for long-term vision based navigation of a micro helicopter in large outdoor environments.

Corresponding IROS video contributions and interactive sessions:
- "Inertial-Visual SLAM for a Small Helicopter in Large Outdoor Environments"
A video showing the performance of our framework on pure vision based navigation for MAVs
=> interactive presentation TODAY on Monday, October 8 at 13:30 - 14:00

- "SFly: Swarm of Micro Flying Robots"
A Video showing high level tasks that can be performed using our framework for low-level vision based MAV navigation
=> interactive presentation TODAY on Monday, October 8 at 13:30 - 14:00

Best, your ASL-sFly team
Stephan Weiss, Markus Achtelik,
Simon Lynen, Laurent Kneip, Margarita Chli





More information about the ros-users mailing list