Automation of construction processes can result in reduced project costs and increased worker safety. A process that lends itself to automation is the picking and placing of objects. However, determining the pose (position and orientation) of an object is critical. LADAR (laser detection and ranging) data provides 3D information of a scene, but the data are noisy, contain outliers, and have phantom points along edges of objects. A preliminary algorithm to preprocess the data and to compute the object pose is presented. The algorithm was validated through comparison with experimental measurements.
Proceedings Title: Proceedings of the 21st Symposium
Conference Dates: September 21-25, 2004
Conference Location: Jeju Island, KO
Conference Title: International Symposium on Automation and Robotics in Construction
Pub Type: Conferences
binning, LADAR, object recognition, object segmentation, pose determination