A Virtual Environment System with Virtual Force Display and Feedback based on Master Arm

1
A Virtual EnvironmentSystem with Virtual Force
Display and Feedback based on Master Arm
Wu Juan, Song Aiguo, Cao Xiaoying
Department of Instrument Science and Engineering
Southeast University, Nanjing, 210096, China
2
Introduction

• Background
• Teleoperation technique has been applied in
  hazardous or uncertain environments, such as
  nuclear plants, outer space, or deep oceans.
• Successful implementation of teleoperation
  systems demands the creation of virtual
  environments that make the operator feels as if
  she or he is actually present at remote site.
• With the inclusion of visual and haptic
  interfaces, such as head mounted displays(HMD),
  data gloves, and force-reflection joysticks,
  operators can visualize, manipulate, and
  interact with objects in the virtual world more
  naturally.

3
Typical haptic interface device(1)
Fig2.SAFiRE(EXOS, Inc.)
Fig1.PHANToM(MIT)
4
Fig3. 7 DOF Stylus (McGill University)
Typical haptic interface device (2)
Fig4. SPICE (Suzuki Motor
Corporation)
5
Simulation System With Haptic Display

• Master arm
• Graphic display
• Simulator

6
Human System Interface

• 5 active DOFs, The arm rotates around the axis Z,
  the axis Y. The fingers have two active freedoms
  that can cooperate to grasp the object.
• When contact force is calculated and sent to
  master arm, then the motor on the arm generates
  the analog force directly acting on the human
  operator.

5 DOF master arm device with force/tactile
feedback
7
Graphic Interface
8
The Composition of Graphic Interface

• The virtual environment consists of a virtual
  slave Motoman manipulator, a virtual rigid ball,
  and a virtual sheet. The operators task is to
  explore the ball, manoenvre the slave virtual
  robot hand to grasp it and move to the virtual
  deformable sheet, and then put it on the sheet
  through directly interacting with the master
  manipulator

Control Area
Graphic Area
Data Area
9
Algorithms of Interaction

• Collision detectionTo find the geometrical
  contacts between the objects
• Collision response To integrate the resulting
  reaction and force effects in virtual environment.

10
Collision Detection

• Main tasks
• 1. Detecting the intersection region
• 2. Estimating the penetration depth.

11
Two Main Phases of Collision Detection

• 1.Broad phase using projection method rapidly
  eliminates objects which are clearly not
  colliding.
• 2.Narrow phase handles possible candidates that
  board phase of collision detection.
• Advantages
• Easy to compute,
• Small memory requirements,
• Fast transformable,
• Simple overlap test,
• Tight fitting.

12
The Main Operation of Sphere Box Collision
Detection

• Building a bounding sphere
• To generate the coarsest level of
  bounding is to find the smallest radius of sphere
  required to completely encompass any particular
  part of virtual object.

13

• the information of the sphere contains
• Radius,
• Centre relative to the origin,
• Centre relative to the entity in its update
  state,
• Comparing the distance between their centers to
  the sum of their radii.

14
Collision Response

• Virtual Force
• 1. Penalty method allowing interpenetration
  of objects and introducing virtual springs.
• 2. Analytical method treating the case of
  non-penetrating rigid bodies.
• Deformation

15
Virtual Force Penalty method

• The contact force is often calculated based on
  penalty methods that require the penetration
  depth.
• FKX
• Where K is the spring stiffness constant, X
  is the depth of penetration.
• The force is applied equally and in opposite
  directions to the two colliding objects. The
  direction of the force is such as to push the two
  objects apart.

16
Haptic Rendering

• The calculated virtual force can be
  represented in graphic and reflected to the
  master hand
• When contact force is calculated and sent to
  the master arm, then the motor on the arm
  generates the analog force directly acting on the
  human operator.
• In the graphic interface, the histogram shows
  the magnitude of virtual force, and the 3D cursor
  show the direction of it.

17
Deformation

• We define the deformation as the change of a
  surface caused by the displacement of surface
  points relative to each other over time.
• We characterize the deformation behavior of
  surface without using the mass distributions. We
  calculate the displacement of each surface gird
  according to Hirota and Michitaka Hiroses method
  as following.

Where d is the plane distance from the rods in
millimeters, F is the intensity of
force exerted on the rod in Newtons,
the constants are defined according to the
experiences
18
Gauss Deformation Model
19
Conclusions and future Work

• In this paper, we introduced our research on
  haptic display system and presented a method to
  realize haptic interaction for teleoperation
  based on master arm.
• Future work will focus on the collision detection
  calculation between complex surface and massive
  models. Hierarchical methods and subdivision
  techniques can be used. Another area for more
  study is the modeling of contact force. We would
  combine different algorithms corresponding to
  different phenomena.

20
Acknowledgment We would like to thank Koichi
Hirota and Michitaka Hirose of Tokyo university,
OSullivan and C.Dingliana of Trinity College
Dublin for the ideas addressed in their papers.
The End