Autonomous 3D Exploration of Large Structures Using an UAV Equipped with a 2D LIDAR

Application of the Lazy Theta* implementation to the exploration problem

The system:

1. Achieves truly autonomous inspection of large constructions in an unknown, unstructured environment.

2. Preforms all the calculations on-board and online with the limited memory and processing power compatible with the vehicle’s payload;

3. Implements a scalable, deterministic, 3D path planner tested in extensive outdoor experimental campaigns as a ready to use ROS library;

4. Employs a geometrical simplification transforming a 3D discretisation into twodimensions in order to reduce the processing time;

5. Guarantees obstacle detection within a tubular, operator safe, flight corridor;

6. Achieves 3D exploration with a low cost 2D laser sensor by complementing the sensor mounting position with a sampling manoeuvre;

7. Is able to store the world information in a single multi-resolution octree to support all the decision-making algorithms involved;

8. Showcases the applicability of software engineering tools and techniques to produce reproducible research and robust software implementations