Hybrid Manipulation: Force-Vision

1
Hybrid Manipulation Force-Vision

• CMPUT 610
• Martin Jagersand

2
Today

• How to incorporate other sensory feedback.
• Focus on tasks where the number of contact
  constraints increase during manipulations.
• Acquire necessary constraint geometry on-line.

3
PreliminariesVisual Servoing Applications
4
Limitations of Visual Control

• Accuracy is limited by Visual tracking
• and Visual goal specification
• (Jagersand et. al. ICRA 97)
• Specifying well defined visual encodings can be
  difficult (Hager, Hespana, Dodds,
  Jagersand)
• Limited to non-occluded settings
• Not all tasks lend themselves to visual
  specification.

5
ObservationForce constraints natural in tasks

• Example
• Inserting a box
• Sequence of motions
• Increasing number of contact constraints

6
Preliminaries Uncalibrated Visual Servoing

• Let y visual observation x motor control
• Linear system model
• y
• Linear p-controller
• Estimate the Visual-Motor Jacobian

7
What do we need to add to do things like this?
8
Learning Constraint Geometry

• Impact force along surface normal
• Sliding motion
• 3rd vector

9
Constraint Frame

• With force frame tool frame we get
• Assume frictionless gt Can update each time step

P2
P1
P3
10
Hybrid Control Law

• Let Q Joint -gt Tool Jacobian
• Let S be a switching matrix, e.g. diag(0,1,1)
• Velocity control u

Visual part
Force part
11
Accounting for Friction

• Friction force is along motion direction!
• Subtract out to recover surface normal

12
Final Hybrid Controller Structure
13
Motion Sequence
14
Switching of Constraints inHybrid Force-Vision
Feedback
15
Sliding along a Curved Surface

• In principle any smooth surface can be followed

Detail of measured force
Estimated surface normals
16
Estimation Accuracy

• The constraint coordinate frame was acquired
  within a few degrees of the true surface frame

17
Conclusions

• Hybrid Force-Vision control allow more accurate
  positioning than vision or force alone
• Many tasks naturally involve an increasing number
  of contact constraints
• We showed that the contact geometry can be
  successfully estimated
• Combined Uncalibrated Vision Force control