Roughness Synthesizing for an Encountered-type Haptic Display

INTRODUCTION

Encountered-type haptic rendering provides realistic, free-to-touch, and move-and-collide haptic sensation to a user. However, inducing haptic-texture sensation without complicated tactile actuators is challenging for encountered-type haptic rendering. In this paper, we propose a novel texture synthesizing method for an encountered-type haptic display using spatial and temporal encoding of roughness, which provides both active and passive touch sensation requiring no complicated tactile actuation. Focused on macro-scale roughness perception, we geometrically model the textured surface with a grid of hemiellipsoidal bumps, which can provide a variety of perceived roughness as the user explores the surface with one's bare hand. Our texture synthesis method is based on two important hypotheses. First, we assume that perceptual roughness can be spatially encoded along the radial direction of a textured surface with hemiellipsoidal bumps. Second, perceptual roughness temporally varies with the relative velocity of a scanning human hand with respect to the surface. To validate these hypotheses on our spatiotemporal encoding method, we implemented an encountered-type haptic texture rendering system using an off-the-shelf collaborative robot that can also track the user's hand using IR sensors. We performed psychophysical user tests with 25 participants and verified the main effects of spatiotemporal encoding of a textured model on the user's roughness perception. Our empirical experiments imply that the users perceive a more rough texture as the surface orientation or the relative hand motion increases. Based on these findings, we show that our visuo-haptic system can synthesize an appropriate level of roughness corresponding to diverse visual textures by suitably choosing encoding values.

OUR APPROACH

TEXTURED SURFACE MODELING

We encode the tactile roughness by arranging hemiellipsoidal textons on a flat surface.

Our two modeling parameters are texton width and the texton spacing that can be moulated in runtime.

When scanning direction is parallel to the x axis (θ = 0)

In general case (θ ≠ 0)

Perceptual width (w_θ ) of texton

Hemiellipsode shape was selected to control the texton width in runtime.

As θ increases, w_θ decreases.

As w_θ decreases, the perceived roughness increases.

=> The rougher roughness encoded as θ increase.

Variation of inter-distance of textons (d_θ ) according to a scanning direction

Inter-distance (d_θ ) of textons

Textons are arranged in a geometric way with d_θ decreasing as θ increase.

As d_θ decreases, the perceived roughness increases.

=> The rougher roughness encoded as θ increase.

OUR APPROACH

HAPTIC TEXTURE RENDERING

A textured surface is attached to the end-effector of the haptic device and interacts with the user's bare hand.

The end-effector rotates a textured surface model to set orientation relative to the user's hand scanning direction.

The end-effector translates a textured surface model to provide a proper relative velocity to the user's hand.

Encoded roughness of virtual objects are rendered to the ueser and make the user feel different roughness of textures.

SYSTEM

SYSTEM SETUP

KUKA LBR IIWA 7 R800 is used as an encountered-type haptic manipulator.

A 3D printed Textured Surface Model is attached to the IIWA end-effector and serves as a proxy for virtual obejct while synthesizing texture roughness.

Leap Motion is used as a motion sensor to track the user's hand.

Oculus Rift CV1 is used to visually render 3D virtual scene and track the user's head.

SYSTEM

ENCOUNTERED-TYPE HAPTIC TEXTURING SYSTEM

User Tracking

The user's head and hand positions are tracked by HMD and IR sensor.

Tracked data are transformed from the sensor frame to the robot base frame.

Midpoint Smoothing algorithm was used for reducing tracking noise and generating a smooth robot motion.

Contact Prediction and Detection

Contact is predicted in a VR environment using the concept of a region of interest.

Physical Contact is detected in the real-world using joint torque sensors embedded in the robot.

Haptic Rendering

A 7-DoF manipulator, capable of human-robot-coexistence, operates as a haptic device.

A textured surface model is attached to the end-effector of the manipulator and used as a proxy for a virtual object.

Kinesthetic feedback is provided by position control and impedance control.

Tactile feedback is provided by adjusting the orientation and velocity of the textured surface model.

Visual Rendering

A user's virtual hand and a virtual environment are rendered through the HMD.

USER STUDY

PRELIMINARY STUDY

Examine if altering users’ roughness perception by changing scanning direction of our model is feasible.

Decide on the surface design of our model for simulating a range of perceived roughness as wide as possible.

Apparatus

r_x = 1.5mm

r_x = 2.0mm

r_x = 2.5mm

r_x = 3.0mm

Four surface textured models with various raddi in x axis (3DS MAX and Form2 3D printer was used)

Five different scanning directions parameterized by angles

Experimental Setup

Subjects were asked to graze the patch in horizontal direction and evaluate roughness using a number
while each patch was covered by an opaque cloth.

Results

Average perceived roughness as a function of x-axis radius r_x with different scanning directions

Both radii (F₍₃₎ =20.4, p<.001) and orientation (F₍₄₎ =17.2, p<.001) have signiﬁcant main effects on roughness estimation.

It is feasible to change perceived roughness by varying the scanning direction using our model.

We chose a textured surface model with r_x = 3.0mm for our encountered-type haptic system which showed the widest range of average perceived roughness to maximize the effect.

USER STUDY

MAIN STUDY

Apparatus

3D textured surface model with r_x = 3.0mm

Physical textured surface model with r_x = 3.0mm

Five different orientation condition and five different velocity condition of surface model are provided to the user.

Experimental Setup

Subjects were asked to graze the surface and evaluate surface texture roughness using a number.

Results

Average perceived roughness as a function of x-axis radius r_x with different scanning directions

There exist significant main effect with large effect size of orientation on roughness estimation.
(F₍₄₎ =71.67, p<.001, η_p²=.3323)

There exist significant main effect with medium effect size of velocity on roughness estimation.
(F₍₄₎ =15.87, p<.001, η_p²=.0993)

Boxplots showing users’ perceived roughness on average varying orientation and velocity.
Stars were indicate p values from pairwise posthoc analysis results: ’*’ for p < .05, ’**’ for p < .01 and ’***’ for p < .001

CONTACT

ewha graphics lab

Department of Computer Science & Engineering, Ewha Womans University

52, Ewhayeodae-gil, Seodaemun-gu, Seoul, Korea, 03760
+82-2-3277-6798
  Yaesol Kim, yaesol91@ewhain.net

  Siyeon Kim, syunni33@ewhain.net

  Uran Oh, uran.oh@ewhain.net

  Young J. Kim, kimy@ewha.ac.kr

Synthesizing the Roughness of Textured Surfaces
for an Encountered-type Haptic Display
using Spatiotemporal Encoding

Yaesol Kim, Siyeon Kim, Uran Oh, and Young J. Kim

CONTACT

ewha graphics lab