Application of robotics methods to Neutron and Synchrotron diffraction instrumentation Jon James, Nov 2008 Department of Design, Development, Environment.

Application of robotics methods to Neutron and Synchrotron

diffraction instrumentation

Jon James, Nov 2008

Department of Design, Development, Environment and Materials (DDEM)

The Open University (UK)

User group software: An example from Engineering diffraction

Strain Scanning Simulation Software

Context: Engineering diffraction

• Measuring residual stress in Engineering samples/objects

1.Aerospace, 2. Power generation,

3.Materials research, 4. Cultural heritage…….

The Problem:

Which measurement points ?

Will it fit on the instrument?

Choice of hardware?

How long will it take?

Instrument control

Possible errors / collisions

Sample repositioning

Counting Neutrons

BeamtimeSScanSS

Typical SScanSS usage - Planning

• Planning– Acquire sample model– Setup measurement points and strain

components– Simulate experiment calculating;

• Instrument movements• Count times• Collisions

– Save measurement plan to HDF file

Typical SScanSS usage - Execution

• Execution– Load measurement plan– Place sample on instrument – measure exact

location and input into software– Run measurement plan simulation and output

• Instrument control file• Collision warnings• Count time estimations

– Archive complete experiment to plan to HDF file

Sample model setup

• Complex sample Articulated arm + Laser head

Setting up Measurement points

• Measurement points are positioned within the sample

Setting up Measurement vectors

• Strain component(s) are defined at each measurement point

Initial sample position

• Touch probe Sample + fiducial balls

Experiment execution

• Simulation of experiment to give:– Instrument motor commands – Neutron path lengths

Robotics

• Common frame-work for variety of systems

• Mathematically compact – minimises code

• Soluble forward and inverse problems

• Very difficult to do any other way

• Limitations….

Robot types

Serial robots Parallel robots

Alignment: The inverse problem

Path length minimisation

• Sample is rotated about Q-vector is search of orientation that minimises path length

Collision Prevention

• Opportunities minimised by: – Training on virtual instrument– Careful planning of experiment – Visual inspection of simulation prior to motor

movements

• + last resort – Numerical Collision detection………

Collision Detection

Bounding box tree -> computationally

economic solution

Collision detected!

Future plans

• Responding to requests from existing collaborators

• New collaborations:– KOWARI (ANSTO), VULCAN (SNS),– JEEP (DIAMOND)

• Special projects– Joint imaging / diffraction instruments (IMAT)– Enable input of tomography models within

SScanSS for locating hidden features

Neutron Tomography models

Using sample models derived from segmented Neutron tomography data will allow access to internal geometric and compositional features.

A cultural heritage illustration 1,2

Ticked boxes ?

• Increases scientific output

• Enables New science

• Unifies user experience across facilities

• Good technical model

• Good collaboration model

• Good funding model

?

?

?

Thank you for your attention and

support

Jon James

Acknowledgments

1. Dr. Salvatore Siano, Istituto di Fisica Applicata "N. Carrara"- IFAC, CNR-Italy and the Archaeological Museum of Florence, for their kind permission to use the material presented in panels (a), (b) and (c) of Figure 4.

2. Dr Robert van Langh, for kind permission to use the image reproduced in panel (d) of Figure 4, (neutron image from NEUTRA, PSI, Switzerland).

3. Research at ORNL sponsored by the Assistant Secretary for Energy Efficiency and Renewable Energy, Office of FreedomCAR and Vehicle Technologies, as part of the High Temperature Materials Laboratory User Program, Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U.S. Department of Energy under contract number DE-AC05-00OR22725. The authors would like to also include an acknowledgement to William Barton Bailey for his efforts on the ORNL-NRSF2 facilities, drawings of the NRSF2 instrument and accessories and contributions to the implementation of SScanSS for NRSF2.

4. The Open University for it’s continued support of this research.