Home >> Preservation Research and Testing >> Preservation Research Projects Updates >> Use of Haptic Technology to Train Conservators

The Use of Haptic Technology to Train Conservators

Background: Although much of the skill of conservators consists of analyzing material properties and conditions to decide which treatment methods to apply, conservators must also master delicate manipulations for treatments. Manual practice requires hands-on training to achieve the highest level of skill. Traditionally, conservators learn by working on collection items or on mock-ups. Training on collection items requires constant instructor monitoring to prevent accidental damage, and while mock-ups can safely support practice and mistakes, considerable labor and space are required to make and store them. In both cases, a trainee usually needs instructor input to identify problems in technique, and significant time is invested to cover a limited set of problems.

Haptic simulation uses computer technology to provide biofeedback through the sensation of touch, simulating different textures and resistances. Users perceive real-time differences in “feel,” allowing them to instantly auto- or self-correct, and providing a safe environment in which to experiment, make mistakes, and develop the right degree of “touch.” Widely used to train medical and dental students, haptic technology could supplement traditional hands-on conservation training by simulating a large range of treatments on a wide variety of substrates and conditions. Haptic simulation can let learners practice basic procedures such as adhesive and backing removal, try out new techniques, hone established skills, and revive skills they have not used recently. Instructors could then focus on procedures that cannot be simulated, such as enzyme use, washing, lining, mending, etc.

Building on explorations of haptic technology for conservation training In the UK in the late 1990's, the Preservation Directorate is collaborating with the University of Washington to develop a pilot simulation system for conservation training.

Contributing Study:
Leuschke, R., R. Donlin, M. Claus, M. Nugent, D. van der Reyden, and B. Hannaford. In press, 2008. Haptic Characteristics of Document Conservation Tasks. Haptics 2008 (proceedings of the symposium, Reno, NV, September 28, 2007. Los Alamitos, CA: IEEE. http://brl.ee.washington.edu/Publications/Publications_html/Rep211.html (accessed 4/8/08).

Project Description: This project tested the feasibility of capturing conservator hand skills to develop a simulator that can teach selected basic paper conservation procedures through haptic technology. Among anticipated benefits are:

Researchers combined video recording with sensors mounted on tools and substrates to capture the hand movements of five highly experienced conservators. The conservators used water, brush, scalpel, needle, and other tools to address five experimental conditions on control samples:


Video and time, force, and torque data were relayed to a laptop computer and stored to permit off-line analysis. Detailed descriptions of control samples can be found in the cited publication.

Outcomes/Findings: Broadly significant results of this study are summarized here:

  1. A Fourier analysis of three sample segments determined the frequency of force magnitude signals for the paper and cloth tape removals and mechanical cleaning tasks. Ninety-five percent of the energy frequency was contained below 25, 39, and 25 Hz for the three tasks respectively. All showed periodic behavior characteristic of the repeated actions typical of these procedures.

  2. Typical force and torque records showed that bursts of force above noise level corresponded to periods of contact between tool and work piece. The chosen conservation tasks featured highly intermittent contact.

  3. Force and torque variations among experienced conservators addressing the same simple problem appear to be surprisingly narrow.

  4. Based on this small number of relatively basic procedures, the range of information processing a conservator needs to achieve the “right” touch in a specific situation may be smaller than anticipated.

The data provides information for design of simulation training systems for conservator hand skills. Further analysis of the collected data should include directional information, clarification and filtering of noise, and refinement of the Fourier analysis parameters. Next steps should focus on collecting data on movement by measuring speed and amplitude in the same tasks. Additional materials should be tested, including Japanese tissue hinges, glassine, and other tapes.

Support:


Acknowledgements: Preservation Directorate staff (Lage Carlson, Elmer Eusman, and Tamara Ohanyan, and Diane Vogt-O’Connor, Eric Hansen, and Richard Herbert all contributed time and expertise to research and reporting); Library Services Management (Deanna Marcum and Robert Dizard)

Update and Images: A poster session has been accepted for presentation at the American Institute for Conservation of Artistic and Historic Works 36th Annual Meeting on April 22-23, 2008 in Denver, Colorado. Additional information will be posted following the meeting.

Home >> Preservation Research and Testing >> Preservation Research Projects Updates >> Use of Haptic Technology to Train Conservators