Robotic Platforms Developed for Diverse Applications
Ferrier will present a brief overview of some recent projects, primarily focusing on
- A High Throughput Robot Machine Vision System for Plant Phenotype Studies
- Robotic Friction Stir Welding
A High Throughput Robot Machine Vision System for Plant Phenotype Studies: A robotic gantry system that uses visual servoing to find and monitor the growth of numerous seedlings has been developed to facilitate plant phenotype studies. Research in the area of gene sequencing has reached levels of maturity where entire genomes have been sequenced and genes can be marked with relative ease and speed. However techniques for defining phenotypes quantitatively are needed because studies require observations from numerous plants for each mutant type and there are roughly 10^4 genes and corresponding mutants per model organism. Machine vision has been used for monitoring individual plant growth and identifying these phenotypes. Incorporating a robot system with machine vision enables the monitoring of many plants concurrently, increasing the throughput of the phenotype studies. An overview of our system will be presented, describing some of the necessary capabilities and challenges to automating plant phenotype studies. Results of initial data collected using the system for gravitropic response of a model organism, Arabidopsis thalia, will be presented.
Robotic Friction Stir Welding: During friction stir welding (FSW), numerous parameter and condition variations are present in a production environment that can affect weld quality. There is a significant need to control the process to assure high quality under the inherent workpiece variability (thermal constraints, material properties, geometry, etc.). Reliable FSW for a wide range of applications will require closed-loop control of certain process parameters. Two current projects include a method to compensate for gaps when butt welding and a real-time temperature measurement and closed-loop control system for stir zone. Ferrier, with her collaborators at UW and Friction Stir Link, Inc. (Brookfield, WI), has demonstrated these systems for welding of aluminum alloys in the presence of real-world manufacturing disturbances. On going efforts are looking to use this system to reduce the need of post-weld inspections (currently over 50% of the cost of FSW) for naval applications.
Dr. Nicola Ferrier joined Energy Systems in September 2012 as Principal Automation Engineer. At Argonne she will build a collaborative program where automation, robotics, and sensing accelerate concepts from research idea to development and commercialization. Ferrier’s research interests are in the use of computer vision (digital images) to control robots, machinery, and devices, with applications as diverse as medical systems, manufacturing, and projects that facilitate “scientific discovery” (such as her current project using machine vision and robotics for plant phenotype studies).
The use of visual feedback in the control is advantageous when precise fixturing is impractical or impossible. Her work has been applied to visual monitoring of polymer processing, visual control of crystal formation processes, visual monitoring of nano-scale assembly processes, visual monitoring of plant growth and development for plant phenotype studies and friction stir welding systems. Ferrier did her undergraduate studies in mathematics, and a master’s degree in computer science at the University of Alberta (1984, 1987). Her doctoral studies, on “control of robot binocular vision systems”, was completed at Harvard University in 1992.
After a post-doctoral fellowship at Oxford University (1992-1994) and a post-doc at Harvard (1995), she joined the Department of Mechanical Engineering at the University of Wisconsin-Madison in 1996. She became an Associate Professor in 2003 and Professor in 2009. She received the NSF CAREER award (1997) and the UW Vilas Associates Assistant Professorship (1999) and the UW Honored Instructor Award (2009). She has published 28 journal papers, over 50 refereed conference papers and awarded 3 patents since joining the faculty at UW. Ferrier has been PI or co-PI on grants totaling over $11M during her career at UW. She served as Chair of the Executive Committee of the Physical Sciences Divisional Committee for the University of Wisconsin-Madison (2011-2012).