Research

Current Projects

Results from SmartScan: An intelligent scanning approach we have developed to reduce thermal-induced defects in additive manufacturing

Intelligent Scanning Strategies for Reducing Thermally-induced Defects in Additive Manufacturing

Additive manufacturing processes usually involve high temperature gradients, leading to various thermal-induced defects, like residual stress and shape distortion. This project seeks to develop new physics-based and data-driven strategies for predicting and compensating thermally induced defects in polymer and metal additive manufacturing

This project is partially supported by the National Science Foundation's Grants #2430109 and #2345676

This project is sponsored by Lawrence Livermore National Laboratory

Controlling the Melt Pool Size in Metal Powder Bed Fusion Additive Manufacturing

Live control of the metal laser powder bed fusion (LPBF) process is an active research topic of considerable interest to fields like aerospace and medicine. This project advances LPBF live control by designing control algorithms for the melt pool area, which is correlated with a wide variety of process specifications and defects, including the part microstructure and macroscopic engineering properties, the extent of porosity in the part, and certain mechanisms of surface roughness. We envision developing controllers that minimize the amount of required calibration data, which synergizes with the inherent flexibility of additive manufacturing processes like LPBF.

Microstructure Control for Powder Bed Fusion Additive Manufacturing

Metal Laser Powder Bed Fusion (LPBF) works by a laser selectively fusing 2D slices of the desired geometry out of a growing bed of metal powder, and the short spatiotemporal scales of the fusion permit a spatially-varying microstructure. This project helps develop this potential for microstructural control by leveraging dual-laser LPBF machines. The second laser enables heating and cooling behaviors that are otherwise impossible to achieve, and the control algorithms we are developing will use these behaviors to drive the observed microstructures within the part to a desired state.

This project being carried out in collaboration with the Air Force Research Laboratory

Closed-Loop Control of Metal Binder Jetting Additive Manufacturing

Despite the growth of metal binder jetting (MBJ) due to its scalability and cost-effectiveness, the high cost of powder impedes its broader industrial adoption. In this project, we investigate the use of cost-efficient water-atomized powder (WAP) as an alternative to the more expensive gas-atomized powder (GAP). We focus on the potential of powder surface topography and ultrasonic signals as indicators for MBJ powder bed density, with the aim of implementing closed-loop control.

Cloud-based CNC will (mostly) reside in the cloud and be provisioned as an on-demand service to manufacturing machines over Internet connections

This project is sponsored by the National Science Foundation's Grant # 1931950 and Cisco

Towards Control as a Service via Cloud-based CNC

Control as a Service (CaaS) is an emerging paradigm where low-level control (e.g., computer numeric control - CNC) of a machine is moved from a local computer on the machine to a remote computer in the cloud, from where it is provided to the machine as an on-demand service over the internet. A machine with access to CaaS can, therefore, leverage the vast computational resources available in the cloud to run advanced control algorithms that boost its performance at low cost. CaaS also facilitates easy upgrades to the control system of a machine and enables it to become more “intelligent” by leveraging data sharing and analytics. This project tackles challenges of safety and reliability of cloud-based CNC using strategic redundancy and machine learning.

Part printed on desktop 3D printer at (a) low speed, high quality and (c) high speed, low quality without vibration compensation; and (b) at high speed, high quality using vibration compensation

This project is sponsored by the National Science Foundation's Grant # 1825133, # 2032814, # 2054715, #2233481 and the Michigan Economic Development Corporation

Boosting the Speed and Accuracy of Vibration-prone Manufacturing Machines at Low Cost through Software

Most manufacturing machines vibrate as they move, due to weaknesses in their mechanical structures. The resultant motion-induced vibration adversely affects the accuracy and speed of the manufacturing machines, thus degrading the quality and productivity of the associated manufacturing processes. This project investigates a filtered basis functions approach for compensating the vibration and servo errors of manufacturing machines via software. The benefit of using software is that it does not add to the weight, size or hardware cost of the machine. An interesting application of this approach is desktop 3D printers, because it boosts their speed and precision at low cost. However, software vibration compensation also applies to high-end manufacturing machines.

MULTI-Robot Overspray Free Painting Of Vehicles

Robotic depowdering is a technique used in metal binder jetting additive manufacturing to remove excess powder from printed metal parts using robotic arms and automated systems. Robotic depowdering offers several advantages in metal binder jetting additive manufacturing. It provides a consistent and repeatable depowdering process, minimizing human error and variability. It also improves efficiency by automating the depowdering step, allowing for faster processing times. Additionally, robotic systems can be designed to handle delicate parts and complex geometries, ensuring gentle and precise depowdering.

Robotic Depowdering for Metal Binder Jetting Additive Manufacturing

Design and Control of a Silicon Wafer Handling Robot

Designing and controlling a silicon wafer handling robot for the semiconductor industry involves careful consideration of the specific requirements and challenges associated with wafer handling such as designing robotic arm, wafer alignment, control system planning and operational considerations. The specific design and control details of a silicon wafer handling robot may vary depending on the specific requirements and processes of the semiconductor industry. Close collaboration with semiconductor manufacturers and industry experts is crucial to ensure the robot's design meets the industry's needs and standards.

This project is sponsored by the EPICREW Corporation

Past Projects

Novel nanopositioning stage designed at S2A-lab

This project is sponsored by the National Science Foundation's Grant # 1562297 and #1855354

Improvement of Speed and Precision of Low-cost Nanopositioners

Nanopositioners are mechanical devices used for precise positioning in a wide range of nanotech processes. Rolling bearings enable low-cost nanopositioning but suffer from reduced speed and precision due to friction. This project explores two approaches to mitigate friction: (1) vibration assisted nanopositioning, i.e., the use of high-frequency vibration, combined with active vibration control; and (2) friction isolator, i.e., the use of a compliant joint between rolling bearings and nanopositioning stage. Both approaches have demonstrated significant improvements in posiitioning speed and precision while maintaining the cost-effectiveness of rolling bearings.

Roll-to-roll manufacturing

Novel friction isolated processing roller developed at S2A-Lab in collaboration with KIMM

Improvement of Speed and Precision of Roll-to-roll Manufacturing

High precision roll-to-roll processes are believed to be one of the most promising technologies for manufacturing flexible and large-area thin film electronics. However, broadband frictional disturbances caused by the guiding ball bearings affect the precision of the processing roller in the roll-to-roll system, severely hampering the quality of the manufactured products. This project extends the friction isolator approach for nanopositions to mitigate frictional disturbances in roll-to-roll manufacturing. Thus, it enables high speed, high precision roll-to-roll manufacturing at low cost.

This project is sponsored by the National Science Foundation's Grant #1855354 and the Korea Institute for Machinery and Materials (KIMM)

Novel hybrid feed drive built at S2A lab achieves superior positioning speeds and precision with 80% less energy.

This project is sponsored by the National Science Foundation's CAREER Award # 1350202

Dynamically Adaptive Feed Drives for High Performance and Energy Efficient Manufacturing

Linear motors are used for fast and accurate motion delivery in a wide range of precision manufacturing machines. They however consume a lot of energy and generate a lot of heat in the process. In this project, we explore a mechatronic design concept where the electro-mechanical structure of linear motor driven machines is designed much like the power train of hybrid electric vehicles; i.e., their structure is designed to intelligently vary in real-time as a function of the manufacturing operation being performed, so as to simultaneously achieve high performance and improved energy efficiency.

Mori Seiki's NN 1000 ultra precision 5-axis machine tool used for demonstrating earlier research results

This project is sponsored by the National Science Foundation's Grant # 1232915

Low-cost and Energy-efficient Reduction of Vibrations in Ultra-precision Manufacturing Machines

Ultra-precision manufacturing (UPM) machines (e.g. ultra-precision machine tools, wafer scanners and micro CMMs) are designed to fabricate and measure complex parts having micrometer-level features and nanometer-level surface finishes. They are often plagued by a combination of ground vibrations and residual vibrations which occur while executing motion commands. We are carrying out a model-based study of the dynamics of passively isolated UPM machines to determine ways of reducing vibrations via design-level mechatronic optimization – without resorting to active vibration control – thus reducing costs and energy consumption

Enabling Performance and Fuel Efficiency Improvements in Vehicles by Developing Dynamic Models for Electric Power Assist Steering Systems

Electric power assist steering (EPAS) systems deliver improved fuel efficiency and flexibility to modern-day automobiles. However, they are plagued by a so-called "sticky feel" which adversely affects their performance. This project, funded by Ford Motor Company, seeks to develop dynamic models that provide insights on how to design and control EPAS to minimize the sticky feel thus enhancing the performance and wide-spread use of EPAS.

GDI Photo courtesy of www.motivemagazine.com

Enabling Improvements in Fuel Efficiency via Modeling and Design Optimization of GDI Actuators

The objective of this project is to improve fuel injector performance in Gasoline Direct Injection (GDI) engines via multi-physics modeling that encompasses electro-magnetics, collision dynamics, and fluid dynamics. Compared to the traditional port fuel injection systems in IC engines, GDI provides major advantages such as increased fuel efficiency, higher power output, and lower emission levels. This project is funded by the Ford Motor Company and MCubed 2.0. It is performed in collaboration with the Precision Systems Design Lab

Photo Courtesy of www.gearheads.org

Steering Systems for Next-generation Automobiles

This pilot project seeks to develop effective, reliable and safe steering systems for next-generation semi-autonomous and fully autonomous vehicles.

(a) Model with variable-width pattern and Printed parts (b) without compensation, Using the state-of-the-art (c) Standard approach, and the compensated results using the (d) Proposed approach

Deposition Control for Material Extrusion Additive Manufacturing

High-quality material extrusion requires accurate synchronization between the dynamically dissimilar systems: the fast positioning and slow extrusion system. To improve the deposition accuracy in material extrusion additive manufacturing, this project aims at modeling the nonlinear extrusion dynamics, and developing model-based control strategies by efficiently using both the position and extrusion system.

Low-Resource Settings

Pathways to Sustainable Manufacturing in
Less Developed Countries

This pilot project explores a multi-disciplinary approach to foster economically, environmentally and socially sound manufacturing in less developed countries.

This project is partly sponsored by Procter and Gamble through MCubed Diamond