Categories
Core Projects Projects

MEDAL

The MEDAL project, which stands for Mobilizing the Emerging Diverse AI Talent, brings together expertise in large language models, virtual reality, and robotic automation to produce collegiate-level coursework focused on using these tools to design and establish automated scientific laboratories. 

Project Leaders:

Dr. Sumit Kumar Jha Lead PI, University of Texas San Antonio 

Dr. Arvind Ramanathan – ANL Partner and Co-PI, Argonne National Laboratory 

Dr. Sreenivasan Ramamurthy – Co-PI, Bowie State University 

Dr. Sunny Raj – Co-PI, Oakland University  

Dr. Sathish Kumar – Co-PI, Cleveland State University  

Dr. Giri Narasimhan Co-PI, Florida International University 

Dr. Rickard Ewetz Co-PI, University of Central Florida 

Categories
Core Projects Projects

Mobile robotics for medical isotope production and processing (med-iso)

This project aims to modernize the field of medical isotope production by developing fully automated, robotically driven workflows to minimize radiation exposure for staff. The primary objective is to advance mobile robotics capable of reducing hands-on radiation risks by enabling precise, autonomous navigation and task execution. The project focuses on achieving reliable mobility for a two-arm robotic system, including precision docking at stationary workstations and the coordinated use of both robotic arms for complex tasks. By integrating advanced sensing and low-latency feedback, the system autonomously transports materials between workstations, establishing a foundation for future funding opportunities in autonomous labs and isotope production automation. Project in partnership with Physical Sciences and Engineering.

Watch our Locking Pin Demonstration

Watch our Hose Demonstration

Categories
Core Projects Projects

A Self-driving Laboratory for Precise and Efficient Inverse Design of Functional Polymers 

Polybot is an AI-driven self-driving laboratory revolutionizing the inverse design of functional polymers through automation and robotics. By integrating robotic platforms such as Chemspeed, UR5e, and Tecan with RPL’s Workflow Execution Interface (WEI), Polybot autonomously handles tasks ranging from monomer recipe formulation to polymer synthesis, purification, and characterization. WEI’s Python-based tool coordinates complex workflows using ROS and TCP sockets, enabling seamless communication between robots for synchronized operations. Polybot’s physics-informed ML model accurately predicts electrochromic polymer properties, refining itself through active learning, and achieving high-precision results in just a few iterations. With this streamlined, fully automated workflow and open-access ECP informatics database, Polybot paves the way for collaborative, high-throughput polymer research and AI-driven material discovery. Project in partnership with Physical Sciences and Engineering and Center for Nanoscale Materials.

Categories
Core Projects Projects

From Automated Light Scattering to Autonomous Material Design

Pipette

The RPL has developed a groundbreaking method that leverages robotics to automate the preparation, characterization, and disposal of complex liquid samples for synchrotron coherent X-ray scattering experiments. Using a pendant drop technique shielded from air turbulence, the automated experiments achieved results consistent with traditional containers, making this approach suitable for high-precision studies. The robotic system, featuring an electronic pipette mounted on a robotic arm, enables precise sample handling and high-throughput exchange. By integrating a single Python script with beamline and robot control libraries, this enables seamless automation, paving the way for AI-driven, fully autonomous material design at large-scale scientific facilities. This approach enhances experimental efficiency and consistency, revolutionizing the study of complex fluids. In partnership with the Advanced Photon Source.

Read the related paper.

Watch the Demonstration

Categories
Core Projects Projects

Autonomous Assembly of Soft Material Chambers

The RPL introduced a compact, fully automated robotic system designed for the precise assembly of small liquid/gel chambers with advanced robotics at its core. This system automates the entire process, from sample preparation to data collection, for X-ray and neutron scattering experiments. Featuring transparent polycarbonate windows and a metallic body for temperature control, it minimizes human interference, improving reliability, particularly for delicate biological samples. Its seamless integration into experimental stations like synchrotrons and X-ray free electron lasers enables efficient, automated sample exchange during beamtime, while AI-driven data analysis ensures a fully autonomous workflow. This robotic platform revolutionizes soft material discovery and biomaterial development through advanced automation and robotics. In partnership with the Advanced Photon Source.

Watch robotic assembly in action!

Categories
Core Projects Projects

Autonomous Design of Antimicrobial Peptides

Automating the design of antimicrobial peptides, which can be used to combat antimicrobial-resistant bacterial strains, was the key motivation for developing Argonne’s BIO-Workcell. This proof of concept was demonstrated on E. coli bacterial strains using high-throughput screening techniques in concert with state-of-the-art artificial intelligence. This approach shows promise for accelerating the drug discovery process, making it faster and more efficient to identify potential therapeutic compounds. In partnership with Biosciences.

Categories
Projects Student Projects

PyHamilton Protocol Implementation on Hudson Robotic Platform

In a recent paper entitled Flexible open-source automation for robotic engineering (Chroy et al. 2021), a new software package named PyHamilton was presented which allows the user to program the actions of Hamilton STAR and STARlet liquid handling robots using standard Python. The capabilities of PyHamilton were demonstrated by preforming several biological experiments including the use of a feedback loop to maintain culture turbidostats and a high-throughput perturbation analysis of metabolites.

For this project, we will reproduce one of the biological experiments presented in this paper using the Python interface that has been developed for the Hudson SOLO liquid handler and Hudson SoftLinx integration system. We aim to prove that our completely open-source Python API can execute the same biological experiments as PyHamilton with similar results. Students on this project will also contribute to the growing library of biological protocols written for our Hudson robotic experimentation platform.

Link to PyHamilton paper: https://pubmed.ncbi.nlm.nih.gov/33764680/

Mentors: Casey Stone, Abraham Stroka, Priyanka Setty

Students: Gillian Camacho, Arleen Hidalgo, Halona Dantes

Categories
Projects Student Projects

Automation of OT-2 Calibration

Trays of pipette tips waiting to be picked up by robotic arm

Biological protocols can be run via a python API on the Opentrons OT-2 liquid handler, but the calibration of the OT-2 is performed in an interactive session, preventing the development of fully autonomous workflows. This project aims to develop an automated calibration method using microswitches and an Arduino microcontroller. The Python-based utility will calibrate the pipette followed by the calibration of the deck.

Mentors: Gyorgy Babnigg, Casey Stone, Ryan Lewis

Students: Miriam Stevens (summer 2021)

Categories
Projects Student Projects

Modular Robotics for Science

Scientific workflows often rely on domain specific instruments. This project aims to engineer “modules” of commonly required instruments to assemble robotic experimentation platforms for certain scientific workflows. This project will encompass both the hardware and software aspects of developing new modules.

At first approach, the modules will replace already existing instruments to minimize the utilization gap. Furthermore, new instruments will be introduced and new scientific workflows purely based on the modules. The student will also be linked with domain scientists to bridge the gap between development and real-life utilization.

Mentors: Rafael Vescovi, Doga Ozgulbas, Mark Herald

Students: Sanjiv Parthasarathy, Kendrick Xie, Eric Codrea, Noah Grom

Categories
Projects Student Projects

DNA Assembly Protocol Development

DNA Assembly is the method of physically linking together fragments of DNA, in an end-to-end fashion, to achieve a desired, higher-order assembly prior to joining to a vector. This is one of the most important protocols in molecular biology and will be a vital capability of our larger autonomous laboratory in the future. There are also several existing projects at Argonne that would benefit greatly from this AI enabled DNA Assembly protocol. For now, this protocol will be developed on the Hudson robotic platform in building 446. In the future, we would like to move this protocol to a more open-source robotic platform like the ones that are being developed in the Rapid Prototyping Lab in building 240.

Mentors: Tom Brettin, Casey Stone, Abraham Stroka

Students: Priyanka Setty