Vehicle Real-Time Attitude Estimation System

Period of Performance: 04/19/1995 - 10/19/1995

$69.9K

Phase 1 SBIR

Recipient Firm

Svs, Inc.
6207 Pan American Freeway NE
Albuquerque, NM 87109
Principal Investigator

Abstract

A Vehicle Real-time Attitude Estimation System (VRAES) is proposed which will use existing real-time tracking systems and proven software for object detection, outlining, and feature extraction. Using generalized shape parameters obtained during the detection and feature extraction processing, a subset of outline filters from a large bank of prestored object outlines will be selected. These will then be compared with the observed object outline to determine identity and orientation. A proven real-time tracking system will be used to provide complete closed loop tracking. The proposed system will perform primary data reduction and orientation estimation at each sensor. In this model, each sensor can operate independently, and if two or more sensors are available, the resulting data will be communicated over a low bandwidth link to a central processor which will resolve attitude ambiguities and triangulate the target position in 3-space. The proposed system is low risk because it draws heavily on existing and proven subsystems. The core processing software for object detection, outlining, feature extraction, and frame-to-frame correlation is the Optical Signature Extraction Pipeline (OSEP) developed at MIT Lincoln Laboratory and used to extract position and signature information from infra-red data collected at White Sands Missile Range. Missions analyzed include STORM BTTV 5, 6, and 7. If closed-loop tracking is used, the Opti-Trak 100 real-time tracker will be used. This system, originally developed as the Air Force Advanced Modular Tracker has been installed and used at a number of sites, including Malabar Tracking Station and AMOS. SVS has extensive experience in both real-time systems and advanced image processing applications.