Skip to content
← Projects

Project RIPRA

Real-time wavefront reconstruction from Shack-Hartmann sensor data

Domain
Optics
Status
Active
Year
2026
Stars
★ 16
Built for
ISRO Bharatiya Antariksh Hackathon 2026

RIPRA reconstructs atmospheric-turbulence-distorted wavefronts from Shack-Hartmann sensor frames and computes deformable-mirror actuator commands for adaptive optics. It combines a C99 real-time core with ML reconstructors and an LSTM predictor for loop-lag compensation. Built for the ISRO Bharatiya Antariksh Hackathon 2026.

The problem

Atmospheric turbulence distorts wavefronts, and correcting it requires fast, accurate reconstruction and turbulence characterization within a real-time control budget.

How it works

  1. 01

    Thresholded centre-of-gravity spot centroiding

  2. 02

    Zonal (Fried geometry) and modal (Zernike) wavefront reconstruction

  3. 03

    Deformable mirror actuator mapping and turbulence characterization (r0, τ0)

  4. 04

    CNN/MLP reconstructors and LSTM predictive control for latency compensation

Results

As reported in the repository

761 µsHot-path latency per frame
2.26 msEnd-to-end (first frame)
0.001957 (mean correlation 99.97%)CNN reconstructor test MSE
4.6× gainCNN vs MLP accuracy
6.6%LSTM predictor residual RMS reduction

Figures

Project RIPRA figureProject RIPRA figureProject RIPRA figureProject RIPRA figure