Title
Simulator for Planning Collision-Free Trajectories in Manipulation of Objects Applications
DOI
https://doi.org/10.1007/978-3-030-45096-0_58
Document Type
Conference Proceeding
Publication Date
1-1-2020
Publication Title
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Abstract
This article presents the functional description of a simulator that plans trajectories in an automatic fashion in order for a Motoman HP20D industrial robot to move box-like objects from an initial to a final position avoiding obstacles in its workspace. The simulator sets up robot positioning based on the data entered by the user as well as the position of obstacles (boxes) by using an offline workspace reconstruction based on a computer vision tool, the main component of which is an RGB-D camera. The vision system estimates the position, orientation and size of a set of boxes to create a virtual representation in the simulator. In addition to that, a collision detection module, cost-efficient in computational terms, is described as the structural block of the trajectory planning process. Finally, the work space road map is obtained through a lazy PRM algorithm, as a global planner, and a type-A* search algorithm, as a local planner. This way, a collision-free trajectory is obtained and animated in the simulator as the final result of the general planning process.
Volume
12014 LNCS
First Page
477
Last Page
484
ISSN
03029743
ISBN
9783030450953
Recommended Citation
Martinez-Caldas, J. S.; Rodriguez-Garavito, C. H.; Peña-Ortega, Carolina; Patiño-Forero, Alvaro A.; and Camacho, Guillermo A., "Simulator for Planning Collision-Free Trajectories in Manipulation of Objects Applications" (2020). Scopus Unisalle. 97.
https://ciencia.lasalle.edu.co/scopus_unisalle/97
Identifier
SCOPUS_ID:85083998552