PatrolDrone

PatrolDrone
Short Description:	The goal of the project is to develop a drone able to patrol an indoor ambient to search for intruders
Coordinator:	AndreaBonarini (andrea.bonarini@polimi.it)
Tutor:	AndreaBonarini (andrea.bonarini@polimi.it)
Collaborator:
Students:	DanieleIasella (daniele.iasella@mail.polimi.it), StefanoFossati (stefanofossati.86@gmail.com)
Research Area:	Robotics
Research Topic:	Robot development
Start:	24/10/2012
End:	20/07/2013
Status:	Active
Level:	Ms

Goal

The aim of the project is to develop an architecture for indoor autonomous exploration, including door and hole detection ability.

Motivation

Environment reconstruction and exploration with limited sensorial capacity (monocular view, altitude sensor, gyroscope) using the low cost quadrocopter A.R. Drone 2.0

Useful readings

SLAM

• SLAM [1]

• EKF [2]

• PTAM [3]

Navigation

• Indoor Navigation with Laser Scanners [4]
• Depth Camera Localization and Navigation [5]
• BIM Based Navigation [6]
• Visibility of Point Clouds and Exploratory Path Planning in Unknown environment [7]

ROS packages

• ardrone driver [8]
• ardrone PTAM and EKF wrapper [9]
• Point Cloud Library bridge [10]

Video

• External link

Real-time exploration and mapping of environment with low-cost drone and low-cost camera, using PTAM and special exploration heuristics taking into account both the nature of the drone and that of the fast algorithm. The three views are: a view from the external camera, the view of the on-board camera, with point cloud visualization, a view of the obtained map. In the last part of the movie there is a visit of the map, showing in red the outliers eliminated by the algorithm.