Upload

Loading...

Learning to Predict Obstacle Aerodynamics from Depth Images for Micro Air Vehicles

399 views

Loading...

Loading...

Loading...

Rating is available when the video has been rented.
This feature is not available right now. Please try again later.
Published on Apr 23, 2014

We describe a system for predicting at a distance, the effects that obstacles can produce on the flight of a micro air vehicle. The most common of these effects is often known as ground effect.
Predictions are based on learning from prior experience
gathered during training flights by linking depth images with the disturbance measured from a motion tracking system. We show that aerodynamic effects caused by obstacles are consistent, and demonstrate that it is practical to make predictions from experience without running a computationally expensive aerodynamic simulation. Our approach uses a Gaussian process regression, it requires minimal parameter tuning and is able to predict the acceleration that will be expected at a distance in the future. The method produces estimates within 12ms without any code optimisation and the results indicate good prediction ability with mean errors within 4-10cm/s2 on a database of various obstacles.

[We have also used this to close the loop with the controler but that is described in another paper]

From the paper:
Learning to Predict Obstacle Aerodynamics from Depth Images for Micro Air Vehicles by
John Bartholomew, Andrew Calway and Walterio Mayol-Cuevas.
IEEE International Conference on Robotics and Automation (ICRA 2014).
PDF at: http://www.cs.bris.ac.uk/Publications...

Loading...


to add this to Watch Later

Add to

Loading playlists...