Research-stage implementation plan
Aperture Implementation Details
Purpose
Project Scope
Estimate optical misalignment from images and command corrective stage motion while accounting for camera noise, settling time, and manufacturing tolerances.
- Reproduce a relign alignment task
- Compare classical and learned controllers
- Define a safe sim-to-real interface
Stack
Tools & Technologies
PythonExperiment harness
relignOptical simulation
GymnasiumControl environment
OpenCVImage metrics
Architecture
System Design
A camera observation feeds an error estimator and controller; a constrained stage model applies the motion and returns a new optical image after settling.
- Optical sensor simulator
- Image-quality objective
- Controller benchmark suite
- Calibration and safety envelope
Validation
Evidence Plan
Results remain research claims until these checks produce reviewable artifacts.
- Known-offset recovery tests
- Noise and tolerance sweeps
- Unseen-start-position evaluation
- Movement-count and convergence reports