System Overview

THE RAPTOR SYSTEM: AN ANALOGUE OF HUMAN (or Owl) VISION

Predators have evolved a highly sophisticated vision system for both imaging and change detection. The human eye has a wide-field, low-resolution, imager (rod cells of the retina) as well as a narrow-field, high-resolution imager (cone cells of the fovea) [6]. Both eyes send image information to a powerful real-time processor, the brain, running "software" for the detection of interesting targets. If a target is identified, both eyes are rapidly slewed to place the target on the central fovea imager for detailed "follow-up" observations with color sensitivity and higher spatial resolution.

Human Vision also employs two spatially separated eyes viewing the same scene both to eliminate image faults like "floaters" and to extract distance information about objects in the scene. The system concept for RAPTOR is best understood as an analogue of Human Vision. RAPTOR employs two primary telescope arrays (RAPTOR A and B) that are separated by a distance of 20 miles to provide stereoscopic imaging. Each telescope array simultaneously images the same 1500 square-degree field with a wide-field imager and a central 16 square-degree with a narrow-field "fovea" imager. Real-time processors instantly analyze images from RAPTOR A and B and the positions of interesting transients are fed back to the mount controllers with instructions to point the fovea telescopes at the transient. The two fovea cameras then image the transient with higher spatial resolution and at a faster cadence to gather light curve information. Each fovea camera also images the transient through a different filter to provide color information. Altogether, the RAPTOR system therefore acts as a closed loop system that autonomously identifies and makes detailed follow-up observations of optical transients in real-time.

   low resolution (85mm)         high resolution (200mm)                                        

Each observation image will have could have hundreds or thousands of false positives within the frame. Categorizing and eliminating these is a daunting task. We have created software and hardware techniques to eliminate these false positives from our transient data. Some of the possible false positives are:

• Hot Pixels

• Satellites

• Airplanes

• Cosmic Rays

• Glints form Space Junk

• Image Defects

• Meteors

• Asteroids

• Flaring Stars

• Comets

• Flaring Active Galactic Nuclei (AGN)

Some of these are interesting astronomical data and while it may not be of interest to optical transient discovery, it could be important and vital information. In order to keep this possible useful data we are constructing an adaptive archival catalog that monitors the current state of the sky and provides an “ideal reference image”, not just for transients, but for “important” nightly variations in more than 50,000,000 stars. Not only will the reference frame be available but also the raw image data can be acquired so that studies of comets, meteors and asteroids can be gleaned from the information.