Construction site navigation for the autonomous excavator Thor
Daniel Schmidt, Karsten Berns
- Year
- 2015
- Citations
- 12
Abstract
The paper at hand is part of the autonomous excavator project Thor (Terraforming Heavy Outdoor Robot) who's goal is the development of a construction machine which performs landscaping on a construction site without an operator. So far the project mainly focused on the local excavation on one position. Due to the high digging forces the machine permanently changes its position during excavation. Furthermore, the global goal is to shape the complete construction site. Therefore, a final test platform needs to permanently reposition itself on the site. Within this paper the construction site navigation function is described which guarantees safe traveling from one pose to another one. It is based on an extended A* path planning algorithm, executed on a 2D gridmap including region growing for obstacles, including forward and backward movement. In combination with an intelligent path following algorithm the machine proved to safely reach its position with the desired orientation.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991