High Contrast Imaging Laboratory

Testbed name Managing institution
HCIL Princeton University
Contact person People willing to give talks
N. Jeremy Kasdin
  • N. Jeremy Kasdin
  • He Sun
  • Christian Delacroix
    		•
    Main scientific focus Testbed environment
    HCIL is designed to demonstrate cutting-edge technologies for exoplanet direct imaging and characterization from space-based platforms. The laboratory simulates an integrated telescope and coronagraph instrument, operating in the visible to near-infrared, similar to that baselined for the WFIRST mission. Our specific focus is the development and validation of the shaped pupil coronagraph along with various model-based wavefront control and estimation techniques. On-going developments of the HCIL include the addition of low-order wavefront sensing (LOWFS) and an integral field spectrograph (IFS). The testbed is located in a 900 sq. ft. clean room with temperature and humidity control. The testbed is equipped with vibration isolation and a clean air system designed for optical research.
    Key hardware items Current status
  • A star-planet simulator using two fiber sources with different wavelengths is used as the source. An off-axis parabola (OAP) is used to create the collimated beam.
  • Two Boston MicroMachine kilo-DMs with 952 active actuators on each are used for wavefront control and estimation.
  • The starlight diffraction pattern is modified by a rippled-shaped pupil coronagraph (SPC) and the energy outside of the dark holes is blocked using a bowtie-shaped focal plane mask (FPM).
  • A science camera (QSI model RS 6.1s) is placed on a 300mm motorized stage for focal plane imaging and phase retrieval.
    		•

    Using the shaped pupil coronagraph only, the testbed can reach a contrast of 1x10^(-4). The addition of the 2 -DM focal plane wavefront control improves the contrast to 1x10^(-7). Several new wavefront control and estimation algorithms, including EFC, stroke minimization, Kalman filtering, and extended Kalman filtering, have been demonstrated in this layout. Current research is directed at achieving end-to-end system identification and reinforcement learning control.

    Future Developments: The testbed will be equipped with low-order wavefront sensing based on the reflected light from the FPM in the near future. A lenslet-based integral field spectrograph (IFS) is under development as a demonstration for the WFIRST coronagraph instrument (CGI). It features an 18% band around 660nm with a spectral resolution of 50 and will be used to demonstrate IFS-based closed-loop broadband wavefront control.
    Software, languages Is our software shared?
  • Matlab
  • Some Python
    		•  Currently private, our plan is to translate our Matlab codes into an open source Python package.

    Reference papers:
  • The shaped pupil coronagraph for planet finding coronagraphy: optimization, sensitivity, and laboratory testing, Kasdin et al. 2004
  • Optimal dark hole generation via two deformable mirrors with stroke minimization, Pueyo et al. 2009
  • Kalman filtering techniques for focal plane electric field estimation, Groff and Kasdin et al. 2013
  • Recursive starlight and bias estimation for high-contrast imaging with an extended Kalman filter, Riggs, Kasdin, and Groff et al. 2016
  • Methods and limitations of focal plane sensing, estimation, and control in high-contrast imaging, Groff et al. 2015
  • Identification of the focal plane wavefront control system using EM algorithm, Sun, Kasdin, and Vanderbei et al. 2017