Workshop on Robotics Use in Accelerators, Targets and Detectors

Wednesday 23 Mar 2022, 09:00 → 15:20 US/Central

Adam Watts, Malachi Schram, William Pellico (FNAL), Mayling Wong-Squires (FNAL)

To address the challenges of operating and maintaining large accelerator complexes as well as associated target systems and HEP detectors, scientists and engineers are investigating the use of robotic systems.  These robotic systems vary greatly from small rovers used for inspections to aerial drones to large mechanical maintenance and repair fixturing systems. Due to the variety of hardware and software needed, Robotic teams are often required to build and operate the systems.  Like other shared DOE initiatives, robotic development will be local but with shared toolsets and hardware concepts.  The DOE accelerator laboratory environment is unique and therefor the robotic systems deployed will also be unique.  Industry will play a role, but ultimately, it will be scientists and engineers at the laboratories that need to find the solution that fits.

We hope this workshop allows for an exchange of ideas in areas of mechanical, diagnostics, and deep learning.

Zoom information has been mailed separately to those that registered.  If you have not yet received the information, please contact Mayling Wong-Squires at mlwong@fnal.gov.

09:00 → 09:15 Welcome - Intro: Introduction to Robotics Systems in Accelerators, Targets and Detectors

A introduction will be to the workshop attendees with a brief overview of content and goals.

Convener: Dr Malachi Schram (JLab)

09:15 → 09:30 DOE Perspective

DOE , Office of HEP, will provide perspective on robotic technologies and applications

Speaker: Dr Glen Crawford (DOE)

09:30 → 09:45 FNAL Accelerator Division Robotics Initiative

The Fermilab Accelerator Division Robotics Initiative has been established to demonstrate how robots can support current and future beamlines in order to improve personnel safety and increase operational beam time. This presentation shows the goals of the initiative, current work, and future plans.

Speaker: Ms Mayling Wong-Squires (FNAL)

RoboticsWorkshopHEP20220323.pdf

09:45 → 10:00 Fermilab Target System Robotics for Remote Handling

Fermilab operates several target systems to provide secondary and tertiary beam to various experiments. Many opportunities for remote handling systems are presented within the target halls of the experiments and within the work cells for autopsy of decommissioned targets in order to maximize ALARA efforts with regards to radiation dose that workers absorb. In collaboration with the UIC senior design groups we built a radiation mapping drone, a 3D image AI processor for accelerator facilities, a remote handling crane for transporting radioactive material, and a robotic arm for PIE.

Speaker: Katsuya Yonehara (Fermilab)

FermiRoboTarget032322.pptx

10:00 → 10:30 ROAM – a Remotely Operated Accelerator Monitor

Monitoring accelerators in operation is a well-known challenge due to the radiation environment. However, there are significant benefits in being able to deploy particular sensors in specific locations in accelerator enclosures for monitoring or troubleshooting purposes. Learning from experience at other labs (ie. XFEL) we used components off the shelf (COTS) and an open source robot control software framework (ROS) to build a remotely controlled rover with both a standard suite of instruments such as cameras (including infrared), LIDAR, and ultrasound as well as the ability to incorporate other ad hoc sensors for specific measurements. For its first deployment, we equipped ROAM with a Gamma radiation monitor to observe field emission from LCLS-II accelerating cavities during the initial conditioning and early beam commissioning phases. Among the many features of the device, here we discuss its ability to simultaneously localize and map the accelerator environment (SLAM), to overcome network outages, and to avoid collisions.

Speaker: Dr Thomas Thayer (SLAC)

robotics_in_accelerators_workshop_2022-AR.pptx

10:30 → 10:40 BREAK

10:40 → 11:10 LANL: A Remote Controlled Robot

LANL AOT-IC has been working with a remote controlled robot in recent years that is instrumented with a variety of sensors and intended to be in a beam line tunnel with beam running and using its instrument suite to obtain info about things like radiation levels, etc. while beam is running. It was used a few years ago to help diagnose a performance problem in our drift tube linac.

Speaker: Lucas Montoya (LANL)

LANL - A Remote Controlled Robot.pptx

11:10 → 11:25 NREC: Introduction to Robotic Solutions

An overview of robotics capabilities and technology at NREC that are relevant to the large array of applications in an accelerator and other complex facilities

Speaker: Mr Jeff Legault (NREC)

NREC Robotics Overview for Accelerators_DoE_23Mar2022.pdf

11:55 → 12:25 Break

12:25 → 12:55 FNAL Mu2E Automated Target Changeout System

The Mu2e Project at FNAL presents some unique challenges due to the highly radioactive tungsten target which must be swapped out once per year. A remote handling system consisting of 2 autonomous machines has been designed and built at FNAL to accomplish this task. The first machine removes/replaces the vacuum chamber access window, and the second machine reaches in with a long arm to remove/replace the target. A detailed description of these 2 machines, as well as other system components is presented, including videos of machine prototypes successfully performing these 2 operations.

Speaker: Mr Michael Campbell (FNAL)

Mu2e RH Presentation for FNAL Robotics Workshop 20220323.pptx

12:55 → 13:10 LANL: Accelerator Remote Handling Telerobotic Techniques and Challenges

At the Los Alamos Neutron Science Center (LANSCE) there is a need to work with and dispose of high radioactive items. Due to the high radiation dose rates the need for remote handling techniques have been pursued. Remote handling techniques and specialized equipment, in conjunction with purchased telerobotic systems, are being developed to accomplish the necessary remote handling work. The use of a Brokk 200 hydraulic mobile unit with Kraftt Telerobotic Predator manipulator arms provide the basis of our haptic manipulator system. This manipulator system allows us to lift items up to 500lbs, torque up to 100ft-lbs, grip up to 300 lbf, fit within large doorways and extend up to approximately 200 inches from its base unit. The purpose of this system was to mainly provide gross dexterity capabilities to disassemble and dispose of high radioactive targets and associated systems. Although its original purpose was to mainly accomplish gross dexterity work we have also accomplished some fine dexterity work. We have developed suction pickup tools for picking up small objects and saw tooling for sizing objects up to 5 inches in diameter. We have developed and we are still in the process of developing the use of removable contamination protection layers and contamination control techniques. We currently use off-the-shelf POE cameras to provide real-time high resolution streaming with minimal latency. Our challenges have provided us with successes and failures to develop remote handling techniques using our telerobotic system. We present the successes, failures and challenges that has been faced while developing these remote handling capabilities.

Speaker: Mr Jordon Marquis (LANL)

Accelerator Remote Handling Telerobotics_LANL_MARQUSI Final.pdf

13:10 → 13:25 FRIB: status and future needs in remote handling

FRIB has successfully transitioned to the beginning of user operations. To support targetry maintenance and user operations, traditional remote handling tooling and equipment is being used. Due to the frequency and commonality of target and wedge changes, FRIB acknowledges the potential benefit to implement industrial robotic and build cell technology. DC torque tools with 0-g reaction arms are currently in the process of being integrated and a 0-g hoist that will be able to be operated with thru-wall-manipulators is in the initial stages of being developed.

Speaker: Mr Michael Larmann (FRIB)

FRIB_FNAL_Robotics_Workshop.pdf

13:25 → 13:35 BREAK

13:35 → 13:50 ORNL: Meca500 Six-Axis Industrial Robot Arm Sample Changer

The Meca500 is a Six-Axis Industrial Robot Arm with 5 micrometer precision. It enhances the throughput of the BIO-SANS beam line at the Oak Ridge National Laboratory High Flux Isotope Reactor. The current robotic arm configuration is capable of changing up to 96 samples. The movements and logic can be modified from a configuration file making the software adaptable to other hardware designs. Furthermore, the robot can be scripted using the beam line experiment variables. The software to control the sample changer was developed using the Experimental Physics and Industrial Control System toolkit and Python.

Speaker: Mariano Ruiz-Rodriguez (ORNL)

Meca_robot.pptx

13:50 → 14:20 JLAB: CEBAF Accelerator Tunnel Robot Project

CEBAF Accelerator Tunnel Robot Project: Advancements in the field of robotics allow for the use of remotely controlled robots to perform tasks that have typically required human intervention. Jefferson Lab has explored the possibilities of robots placed in the accelerator tunnel to perform remote inspections that would otherwise require powering down equipment and completing lengthy radiological surveys that impact beam operations. A robotic solution – driven by a human operator – could not only reduce accelerator down time but also support the Lab's radiation exposure policy (known as "ALARA" - As Low As Reasonably Acceptable) by handling routine tunnel and beamline investigations in place of a human being. However, operating a remote device in a radioactive, high electronic-noise area is not a trivial task and drop-in solutions are rarely as simple as they seem. We will present the past prototypes that were tested as well follow-on technology developed from the project. We will also briefly survey current applications of robots at accelerator facilities with an eye toward future applications.

Speakers: Christopher Tennant (JLab), Mr Theo McGuckin (Jefferson LAb)

CEBAFAcceleratorTunnelRobotProject.pptx

JLab_Robotics_TENNANT (03-23-2022) FINAL.pdf

Advancements in the field of robotics allow for the use of remotely controlled robots to perform tasks that have typically required human intervention. Jefferson Lab has explored the possibilities of robots placed in the accelerator tunnel to perform remote inspections that would otherwise require powering down equipment and completing lengthy radiological surveys that impact beam operations. A robotic solution – driven by a human operator – could not only reduce accelerator down time but also support the Lab's radiation exposure policy (known as "ALARA" - As Low As Reasonably Acceptable) by handling routine tunnel and beamline investigations in place of a human being. However, operating a remote device in a radioactive, high electronic-noise area is not a trivial task and drop-in solutions are rarely as simple as they seem. We will present the past prototypes that were tested as well follow-on technology developed from the project. We will also briefly survey current applications of robots at accelerator facilities with an eye toward future applications. Notice: Authored by Jefferson Science Associates, LLC under U.S. DOE Contract No. DE-AC05-06OR23177. The U.S. Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce this manuscript for U.S. Government purposes.

14:20 → 14:40 PNNL: Fieldable AI for Robotics

Deep learning-based methods have enabled tremendous advances in real-world autonomy and robotics applications. Critical systems such as robots that operate near humans and sensitive machinery must meet strict safety requirements to avoid serious harm to people and property. A significant amount of research efforts have focused on methods that provide safety guarantees for learning-based systems in general settings. This talk presents control methods developed at PNNL that meet the challenges of modeling and controlling constrained, safety-critical systems, such as autonomous robots. In particular, we focus on identifying stable system models and reference tracking control policies for autonomous underwater vehicles for safe navigation in open waters. Our methods bring together the rich expressivity of advanced deep learning with provable theoretical guarantees of modern control methods to build synergistic, domain-aware models that outperform their counterparts. Our techniques leverage data-driven strategies for fast development compared to traditional physics-based models and lead to suitable, deployable models with trusted autonomy.

Speakers: Dr Jan Strube (PNNL), Soumya Vasisht (PNNL), Dr Wenceslao Shaw Cortez (PNNL)

PNNL_FNALworkshopSlides.pdf

14:40 → 14:55 JPL Robotics

A look at JPLs Robotic efforts

Speaker: Dr Fernando Mier-Hicks (JPL)

JPL Robotics - FermiLab.pdf

14:55 → 15:05 Close of Worshop with discussion

Robotics Workshope Discussion

Convener: Mayling Wong-Squires (FNAL)

ChatFromRoboticsWorkshop.txt