Skip to Main Content
Outlined is a new approach to sensor space planning and its application to the construction industry. The software planning tool described here generates sensor placements automatically for use in assessing deviations in construction environments. The first step is to separate construction information goals into clusters, simplifying the planning space in order to reduce computational complexity. For each cluster, the planner generates the space of potential sensor placements for a set of information goals and selects a minimal set of subspaces to take advantage of views that can achieve multiple goals simultaneously. Sensing locations are chosen that maximize the probability of achieving each goal and a path is generated to minimize the transit cost between the various sensing locations within each cluster. Finally, paths are generated that minimize the transit cost between clusters. This method is demonstrated on a desktop computer and shown to support LIDAR information goal sensor planning within a construction site.
Robotics and Automation, 2004. Proceedings. ICRA '04. 2004 IEEE International Conference on (Volume:5 )
Date of Conference: 26 April-1 May 2004