I-Corps: Artificial Intelligence Generated Labeled Maps from Reality Capture Data

The broader impact/commercial potential of this I-Corps project is to streamline the Architecture, Engineering and Construction community workflows so that considerable cost savings are passed on to their customers. This I-Corps project will enhance the scientific understanding of automatic feature extraction from a variety of multiple sensor data and imagery over a wide range of field conditions. This project will require innovative algorithms for processing and compression of huge datasets. This project has potential impact on combating climate change through automatic processing of massive amounts of data collected from earth orbiting satellites, drones, or ships. Elements from this project can be used to enhance workforce development programs to include occupational training on geospatial information systems. This I-Corps project’s visualization modules can support spatial thinking development in pre-college programs through hands-on manipulation of digital models of the built environment. Spatial thinking is highly correlated with competency in mathematics. The commercial impact of this project will transform the geospatial mapping industry by eliminating manual intervention and allowing machines to perform repetitive mapping tasks with greater accuracy and speed. Accordingly, this project will enable more accurate and efficient data collection, analysis, and decision-making processes, relieve human stress and strain, and enhance productivity.This I-Corps project is based on the development of an automated data processing system that will increase productivity in mapping workflows using reality capture data. Architecture, Engineering and Construction companies are heavily invested in reality capture technologies to map the infrastructure of the built environment from point clouds and imagery. Such investments are meant to reduce operational costs associated with field data collection and subsequent production of design plans for construction and the development of information systems for civil asset management projects. However, outdated workflows that incorporate reality capture data often become entangled with unproductive processes that slow down mapping operations, increase human-induced errors, and yield a negative revenue impact. A viable solution is an innovative software system that uses artificial intelligence algorithms to extract meaningful information from point clouds and imagery to produce industry-standard documents and annotated plans that satisfy pre-construction approvals/permitting requirements. This I-Corps project will be capable of producing digital information for intelligent infrastructure asset management, condition assessment, risk management, and overall project management. This project reduces human error and decreases the mapping production time with significant cost saving to customers.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.