QoE Assessment of Adaptive Viscoelasticity Control in...

14
QoE Assessment of Adaptive Viscoelasticity Control in Remote Penmanship System Papa Faye 1 , Pingguo Huang 2 , Yutaka Ishibashi 1 , Yuichiro Tateiwa 1 1 Nagoya Institute of Technology, 2 Seijoh University Tokai-Section Joint Conference on Electrical, Electronics, Information, and Related Engineering Sept. 9, 2019 Daido University

Transcript of QoE Assessment of Adaptive Viscoelasticity Control in...

  • QoE Assessment of Adaptive Viscoelasticity Control in Remote Penmanship System

    Papa Faye 1, Pingguo Huang 2, Yutaka Ishibashi 1,Yuichiro Tateiwa 1

    1 Nagoya Institute of Technology, 2 Seijoh University

    Tokai-Section Joint Conference on Electrical, Electronics, Information, and Related EngineeringSept. 9, 2019 Daido University

  • BackgroundRecently, a number of researchers have been paying attention to remote control systems used as remote surgery systems, remote penmanship systems, and so on.

    In each system, a user can operate a haptic interface device at a remote location by using another haptic interface device while watching video.

    Transmission of haptic information in QoS (Quality of Service) non-guaranteed network such as the Internet

    Degradation ofQoE (Quality of

    Experience)

    QoS control

    Network delay,delay jitter, andpacket loss

    We can largely improve the efficiency of work by using haptic interface devices.

  • Previous work

    The adaptive elasticity control, adaptive viscosity control, and adaptive viscoelasticity control were proposed as QoS control for work of writing characters.

    The three types of control were compared with each other by QoE assessment.The experiment results demonstrate that the adaptive viscoelasticity is themost effective.

    *1 T. Abe et al., IEICE Technical Report, CQ2018-91, Jan. 2019.

    The adaptive viscoelasticity control dynamically changes the elasticitycoefficient and viscosity coefficient according to the network delay *1.

    Elasticity is perceived when stretching or pushing a spring.

    Viscosity is perceived when moving object in

    water or oil. Combination of elasticity and viscosity

  • Problems

    The remote control system with haptics was used as a remote penmanship system in which a teacher teaches a student how to write characters while feeling reaction force and watching video *2.

    The student also learns how to write characters while feeling the reaction force and watching video.

    *2 H. Mizuno et al., in Proc. IEEE International Conference onMultimedia and Expo, pp. 1202-1205, June/July 2009.

    The remote penmanship system may have different QoE (Quality of Experience) from the remote control system in which there is only one user who operates the haptic interface devices.

  • In order to improve the educational effect, we carry out QoS control such as the adaptive viscoelasticity control, which is the most effective for the remote control system.

    We compare the adaptive viscoelasticity control with the adaptive elasticity control and the case without adaptive control for the remote penmanship system by experiment.

    We illustrate the effectiveness of the adaptive viscoelasticity control.

    This work

    We do not handle the adaptive viscosity control because the control is not so effective.

  • Haptic interface device

    Whiteboard sheet

    Video camera

    Teacher terminal Student terminal

    French character ‘H’

    s Assessment system

    Whiteboard marker

  • Demo video Teacher terminal Student terminal

  • Calculation of Reaction force

  • Assessment method (1/3)

    Each subject behaved as a teacher to teach a student (one of the authors) how to write a French character ‘H’, and then behaved as a student to learn how to write the character (at this time, the author behaved as a teacher).

    Before assessment as each of teacher and student, each subject practiced for 60 seconds on the condition that there was no additional delay produced by a network emulator. The operability of haptic interface device at this time was regarded as the standard quality in the assessment.

    After practice, combinations of additional delay and type of control were presented in random order for each subject.

    Same person in whole assessment

  • Assessment method (2/3)

    Score Description5 Imperceptible4 Perceptible, but not annoying3 Slightly annoying2 Annoying1 Very annoying

    After each stimulus, each subject made his/her judgment about the operability of haptic interface device based on the five-grade impairment scale.The subject gave a score from 1 through 5 for each stimulus.

    Five-grade impairment scale

  • Assessment method (3/3)Constant additional delay: 0 ms, 50 ms, ..., 250 msThe assessment was carried out for the following three types of control: The adaptive viscoelasticity control The adaptive elasticity control Without adaptive control:

    Elasticity coefficient: 0.1 N/mmViscoelasticity coefficient: 0.0000 Nms/mm

    (Optimum values when the additional delay is 0 ms*1)

    The number of subjects (males and females) whose ages were between 21 and 28 was 15.

    *1 T. Abe et al., IEICE Technical Report, CQ2018-91, Jan. 2019.

  • Assessment results (1/2)MOS of teacher

  • Assessment results (2/2)MOS of student

  • We carried out the adaptive viscoelasticity control for a remote penmanship system and investigated the effect of the control by QoE assessment.

    The adaptive viscoelasticity control is the most effective in the remote penmanship system.

    The MOS value decreases as the network delay increases.

    Assessment of educational effect (e.g., reproducibility of written characters)

    Conclusions

    Future work

    スライド番号 1スライド番号 2スライド番号 3スライド番号 4スライド番号 5スライド番号 6スライド番号 7スライド番号 8スライド番号 9スライド番号 10スライド番号 11スライド番号 12スライド番号 13スライド番号 14