Thesis Defense Presentation

1
Improving the Accuracy of Target Engagement
Presented by Lane Carlson1 M. Tillack1, T.
Lorentz1, N. Alexander2, D. Goodin2, R.
Petzoldt2 (1UCSD, 2General Atomics) HAPL Project
Review Santa Fe, NM April 8-9, 2008
2
Hit-on-the-fly experiment has demonstrated
improved engagement on moving targets

• Improvements in imaging and relaying the glint
  return signal to the PSD give a more repeatable
  dependable signal.
• Improvements have yielded better engagement
• HAPL Oct 2007 - 150 µm (1?) for targets in 1.5
  mm range.
• 80 µm (1?) engagement for targets in 1.5 mm
  range.

• Final Requirement
• 20 µm engagement accuracy in (x,y,z) at 20 m
  (10-6)

3
Surface issues have been resolved with glint
return imaging improvements

• Last time, concerned about surface roughness and
  laser beam profile.
• However, imaging system was looking at an
  intermediate, incoherent field.
• With simplified setup, found proper focusing
  techniques that allow more precise imaging of the
  glints true apparent source.

Surface finish issues are resolved with proper
focusing and aperture.
Proper focusing
Improper focusing
4
Glint return provides a final reference point
5
Systematic identification and reduction of errors
yields improved target engagement

• Error sources as of Oct 2007
• Relaying glint return....50 µm
• PSD accuracy limit..50 µm (spec)
• Target motion from
• glint to driver..24 µm (1?)
• PSD non-linearity..40 µm (1?)
• Verification camera7 µm
• Mirror pointing..6 µm

• Simplified setup helped identify four areas of
  improvement related to relaying glint return
• Proper focusing of the glint return on the PSD.
• Aperturing the glint return signal.
• Changing the systems magnification.
• Higher-order PSD calibration fit.

6
Proper imaging of the glint return requires
focusing and aperturing

• The ideal glint return is an Airy disk with
  apparent location displaced from the targets
  center.
• Previously, we were defocused due to saturation
  concerns.
• Proper focusing and aperturing of the glint
  return minimizes coma, surface roughness speckle,
  and wave aberrations.

Glint off target
Aperture
Intermediate field
At focus
Glint return relayed and Imaged onto sensor
Focus planes
(Not to scale)
7
Improvements to imaging the glint return on the
PSD yield a more consistent return
Glint off target is imaged onto the PSD. How the
glint is imaged is very important.
8
Magnification leverages target-PSD calibration

• PSD specified position accuracy is 50 µm.
• gt can correspond to a large target position
  error.
• By leveraging magnification ratio, we use more of
  sensors area and reduce glint return positioning
  error.

9
Imaging improvements contribute to total
engagement improvement

• Cycle through improvements, then X, Y individual

10
Imaging improvements contribute to total
engagement improvement

• Cycle through improvements, then X, Y individual

11
Imaging improvements contribute to total
engagement improvement

• Cycle through improvements, then X, Y individual

12
Imaging improvements contribute to total
engagement improvement

• Cycle through improvements, then X, Y individual

13
Imaging improvements contribute to total
engagement improvement

• Cycle through improvements, then X, Y individual

14
Improvements in relaying the glint return have
eliminated some errors

• Current error sources
• Relaying glint return....50 µm..eliminated
  with improvements
• PSD accuracy limit..50 µm (spec)
• Target motion from
• glint to driver..24 µm (1?)
• PSD non-linearity..40 µm (1?)reduced to 19
  µm (1?) with higher
• Verification camera7 µm order
  calibration fit
• Mirror pointing..6 µm

15
PSD limit necessitates an alternative sensor

• PSD spec limits position accuracy to 50 µm.
• Deviation from improved PSD calibration 19 µm
  (1?).
• Using a camera will allow for direct geometric
  position reporting of the glint centroid.
• Camera accuracy and resolution lt 1 µm with energy
  centroiding techniques.

6 x 4 mm camera image of glint return
Basler GigE camera
16
Vacuum chamber design progresses to permit
engagement of lightweight targets
2 m tall
Dropping chamber

• Wake effects on target minimal for SS BBs but
  substantial for a lightweight targets.
• Better placement accuracy anticipated.
• Will permit dropping and engagement of
  lightweight targets.

Crossing sensors
Engagement chamber
17
We expect substantially reduced errors with
implementation of camera and vacuum chamber

• Predicted error sources
• Relaying glint return. --
• PSD accuracy limit..50 µm (spec)1 µm with
  camera
• Target motion from
• glint to driver..24 µm (1?).reduced to
  5µm by vacuum
• chamber
• PSD non-linearity..19 µm (1?)eliminated
  with camera
• Verification camera7 µm
• Mirror pointing..6 µm

• With full implementation of the camera and vacuum
  chamber, we can reasonably expect to attain 20
  µm engagement based on these numbers.

18
Future effort focuses on completing demo and
achieving 20 µm engagement goal

• In summary
• We are continuing to use the glint return off a
  falling target to engage targets to 80 µm (1?).
• Next steps
• Construct a more rigorous 3D calibration.
• Replace PSD with camera to better resolve glint
  return.
• Long-term effort
• Mate with a prototypic injector in vacuum, engage
  real targets.

19
End of slideshow
20
Our new imaging setup resolves surface issues

• In practice, engagement error was more than seen
  with a stationary target.
• Concerned that a falling, rotating target might
  present a different surface to be imaged.

• With a simple camera setup, demonstrated that a
  target rotated on a kinematic mount has a
  position repeatability equivalent to a stationary
  target.

Rotated target 5 µm
Stationary target 3 µm
Target rotation/different surface presentation is
not a concern.
21
Imaging improvements contribute to total
engagement improvement

22
Additional information

• Wake effects between glint and driver beam
  location? Minimal for a steel BB in air,
  substantial for a lightweight target.
• Found that divergence is too great to adequately
  homogenize glint laser beam.
• Verified that glint lasers beam profile does not
  affect glint return.