Large-scale manufacturing is common in industries such as aerospace, transportation, maritime, energy, and construction. The final manufacturing steps are finishing operations (e.g. trimming, surface treatment, grinding, sanding, or painting) that are needed to achieve the required geometrical tolerances and surface properties. Finishing operations on large-scale components often require repositioning of the part, as well as multiple workers. These tasks are tedious for humans, leading to physical discomfort and repetitive strain injuries. In the composites manufacturing industry, where human workers are predominant in performing finishing operations, specific barriers hinder the automation of such processes: (1) final part shapes may vary from the planned geometry, (2) the nature and duration of tasks varies from one part to the next, and (3) task completion is based on human judgment and experience. The project will address these barriers by improving the quality and cost-effectiveness of large-scale manufacturing and enhancing safety for human workers. Research outcomes will be integrated into outreach and educational activities to provide K12, undergraduate, and graduate students from diverse backgrounds with learning and training opportunities.
The project will research a collaborative robotic system that leverages robot-robot collaboration with trained human supervisors in the finishing of composite wind turbine blades. The implemented architecture will create scalable and customizable capabilities for high-dimensional robotic systems performing tasks in dynamic and uncertain environments and will include research thrusts in computer vision, control theory, motion planning, and artificial intelligence. Affordable sensing capabilities, machine learning and localization techniques will be employed to develop autonomous environment characterization and cognitive systems for robotic task completion through human-robot interaction. Furthermore, a decentralized planning-control architecture based on reachable set analysis will be designed to coordinate motion and interaction capabilities for teams of robots in human-occupied spaces. The researched theories will facilitate simple, but comprehensive, human-robot information exchange and ensure safe human-robot cooperation. This project is jointly funded by the Division of Civil, Mechanical and Manufacturing Innovation (CMMI) and the Established Program to Stimulate Competitive Research (EPSCoR).
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.
