Anti-sway clone

1
Anti-sway clone
2
Reverse engineering of car suspension system

• Given Quarter Car Model, reverse engineer
  controller for semi-active car suspension
• Matlab/Simulink model provided by CTU and Intec

3
Car suspension model
4
Reverse engineering of car suspension system

• Qualitative controller was induced with QUIN
• Induced controller applied to control car model
• Performance compared with original controller

5
Performance comparison
Original controller
Qualitative controller
Comparatively better
Comparatively worse
6
Summary of approach

• Reverse engineering in qualitative terms, useful
  to grasp the intuition behind the design
• Induction of qualitative trees (QUIN)
• Qualitative-to-quantitative transformation of
  qualitative control strategies (involves
  optimisation)

7
Qualitative vs. quantitative models

• Induced qualitative models superior as
  explanations (unsurprisingly)
• Induced qualitative controllers also perform
  better (surprisingly!)
• Qualitative controller means
• qualitative control strategy induced by
  QUIN, applied as controller after
  qualitative-to-quantitative transformation

8
Why numerical learning fails?

• 1. Linear regression alone insufficient -
• easy to see
• 2. But, why model trees also fail?
• 3. Why qualitative trees succeed when model
• trees fail?

9
Example from crane

• M5 model tree contains leaf
• - 0.0055 lt lX lt 0.0055
• Corresponding example set is very awkward for
  linear regression, see visualisation
• QUIN finds better split
• - 0.223 lt lX lt 0.163
• This example set exhibits clearer qualitative
  pattern

10
Example from crane, cont.
11
Example from crane, cont.
12
Tentative explanation

• QUIN can detect subtle qualitative patterns,
  although they are obscured by large quantitative
  changes
• Regression tree learning (M5) does not find good
  splits, resulting in regression-unfriendly leaves

13
Comparison

• Qualitative trees more successful
• suitability as explanation
• control performance
• Why regression trees fail?
• Important patterns overshadowed by numerically
  dominant attributes
• cVDes does not appear in regression trees!
• Still qualitatively visible