On the Discrimination of Stiffness during Pressing and Pinching of Virtual Springs

On the Discrimination of Stiffness during

Pressing and Pinching of Virtual Springs

Chair of Media Technology

Technische Universität München

IEEE International Symposium on Haptic Audio-Visual Environments

and Games, Dallas, TX

11 October 2014

Giulia [email protected]

Burak [email protected]

Eckehard [email protected]

Cem [email protected]



Teleoperation with perceptual coding

11 October 2014 2Burak Cizmeci: Discrimination of Stiffness during Pressing and Pinching Events

position/velocity

Network

force

operatorteleoperator

force

position/velocity

Hinterseer et al. “Perception-based data reduction and transmission of haptic data in TPTA systems„ IEEE Trans. on Signal

Proc., 2008

Ferrel and Sheridan, “Supervisory Control of Remote Manipulation„ IEEE Spectrum, 1967

stabilitytransparency



3

force force

sensors

Model

Estimator

Local

Rendering

position/velocity position/velocity

Network

operator teleoperator

P. Mitra and G. Niemeyer, “Model mediated telemanipulation”, International Journal of Robotics Research, 2008

Model parameters

Model Parameters

o Object shapes

o Object movement

o Material properties: stiffness, damping and friction

transparency stability

Model-mediated teleoperation

11 October 2014 Burak Cizmeci: Discrimination of Stiffness during Pressing and Pinching Events



Overview

4Burak Cizmeci: Discrimination of Stiffness during Pressing and Pinching Events

• Introduction to stiffness perception

• Experimental conditions and procedure

• Exp. 1: Pressing vs. Pinching

• Exp. 2: WFs for various references

• Conclusion & Future Work

11 October 2014



Introduction to stiffness perception

Discriminating stiffness


Deformable

surface

Non-deformable

surface

11 October 2014

by pinchingby pressing with haptic devices

Perception with toolsPerception with direct contact



Stiffness perception with haptic devices

Challenges to utilize previously determined perceptual thresholds:

A wide array of applied methodologies and technologies [1,2,3,4]

□ From custom-designed to commercial haptic devices

□ Different stiffness ranges

□ Different test conditions from real and virtual scenes to augmented reality

A broad range of stiffness thresholds from 0.036 [3] to 0.30 [1]

6Burak Cizmeci: Discrimination of Stiffness during Pressing and Pinching Events11 October 2014

[3] N. Gurari, K.J. Kuchenbecker, and A.M. Okamura. “Stiness discrimination with visual and proprioceptive cues”, EuroHaptics 2009

[1] M.K. O'Malley and M. Goldfarb. “The implications of surface stiffness for size identification and perceived surface hardness in

haptic interfaces”, ICRA 2002

[2]S. Yamakawa, H. Fujimoto, S. Manabe, and Y. Kobayashi “The necessary conditions of the scaling ratio in master-slave systems

based on human difference limen of force sense” IEEE Trans. On Systems, Man and Cybernetics, 2005.

[4] M. Kuschel, M. Di Luca, M. Buss, and R. L. Klatzky, “Combination and integration in the perception of visual-haptic compliance

information”, IEEE Trans. on Haptics, 2010



Research questions

1. Does the exploration event make any difference?

□ Experiment 1: Detection of WFs for pressing and pinching with

respect to a reference stiffness 200 N/m

2. Is there any influence of the reference‘s stiffness on the

differential sensitivity for stiffness?

□ Experiment 2: Detection of WFs for pinching with respect to

reference stiffnesses 50, 135, 220, 305 and 390 N/m




Experimental conditions Haptic Device: Force Dimension Sigma 7

Interacting object: Accordion shaped elastic cubes with non

deformable surface rendered with linear model (F=kx).




Experimental procedure


1-up 4-down adaptive double staircase detection is applied.

2-IFC (interval forced choice) procedure:

□ 500 ms pause between each stimuli

□ 1600 ms pause between each trial

□ After the second interval, the subject inputs his/her answer

□ A break of 30 sec after 20 trials

□ 5 min break in the middle

Training session with 40 trials

11 October 2014

H. Levitt, “Transformed up-down methods in psychoacoustics” , The Journal of Acoustical society of America, 1971



Experiment 1: Pressing vs. Pinching


Group Pressing

Starting stiffness value

Pinching

Starting stiffness value

202.5 N/m (below threshold) 202.5 N/m (below threshold)

262.5 N/m (above threshold) 262.5 N/m (above threshold)

197.5 N/m (below threshold) 197.5 N/m (below threshold)

137.5 N/m (above threshold) 137.5 N/m (above threshold)

Group Bbelow reference

Group Aabove reference

Reference stiffness value 200 N/m compared with harder and softer

stimuli

10 male and 4 female (7 subjects per group), aged between 22 and 30



Experiment 1: Results

Groups Pressing

(WF±σ)

Pinching

(WF±σ)

Group A 0.147±0.047 0.166±0.032

Group B 0.134±0.032 0.143±0.041

11

ANOVA factors F(1,12) p-value significant

pressing vs. pinching 1.93 0.19 No

above vs. below reference 0.97 0.34 No

Interaction between event

and group

0.28 0.60 No




Experiment 2 : WFs for various references


Pinch the object using the gripper fingers

Reference stiffness values 50, 135, 220, 305, 390 N/m compared

with harder and softer stimuli

13 new participants (5 female) aged between 23 and 28

Reference Starting point

above threshold

Starting point

below threshold

50 N/m 47.5 N/m 7.5 N/m

135 N/m 132.5 N/m 92.5 N/m

220 N/m 217.5 N/m 177.5 N/m

305 N/m 302.5 N/m 242.5 N/m

390 N/m 387.5 N/m 287.5 N/m

11 October 2014



Experiment 2 : Results


Reference 50 N/m 135 N/m 220 N/m 305 N/m 390 N/m

WF±σ 0.281±0.127 0.151±0.05 0.143±0.032 0.127±0.038 0.132±0.034

Significant difference is detected between softest reference 50 N/m

and the other reference values. F(4,48)=15.94, p=0

No statistical difference between 135, 220, 305 and 390 N/m

11 October 2014



Conclusion

We studied stiffness perception of virtual springs

We analyzed two events: pressing and pinching

Observation 1:

□ No significant difference between pressing and pinching

Observation 2:

□ No statistical perceptual difference between stimuli (135 – 390 N/m) for

pinching




Future Work

15

• Further investigation for the stiffness range (50 - 135 N/m )

Perform experiments on deformable surface compliant objects

Perform experiments on incongruent visual-haptic information

o Under delay conditions

Validation of results on a real physical interaction

Usage of these thresholds in stiffness data reduction for model-

mediated teleoperation




Thank you for your attention!


On the Discrimination of Stiffness during Pressing and Pinching of Virtual Springs

Technology

discrimination of stiffness

haptic deviceschallenges

haptic interfaces

pinching events11

stiness discrimination

transmission of haptic

haptic compliance information

tpta systems ieee