Intelligent Techniques for Star-Pattern Recognition

2003 ◽  
Author(s):  
Lalitha Paladugu ◽  
Marco P. Schoen ◽  
Brian G. Williams
Keyword(s):  
Good Accuracy ◽  
Attitude Determination ◽  
Space Vehicle ◽  
Field Of View ◽  
Space Vehicles ◽  
Neural Network Approach ◽  
Star Pattern ◽  
The Neural Network ◽  
Star Pattern Recognition

This work presents the study of two different approaches for the attitude determination of space vehicles. The Neural Network approach is based on a simple Kohonen network, where the characteristics of a star distribution within the Field Of View (FOV) are matched against an on-board stored star map. The second approach utilizes a Genetic Algorithm (GA). The GA minimizes the discrepancy between the characteristics of the stars inside the FOV and a candidate FOV selected from the star map. Preliminary simulations indicate that both approaches work well and deliver good accuracy in determining the bore sight direction of the space vehicle with respect to the star map.

Optimization of a Star Pattern Recognition Algorithm for Attitude Determination Using a Multi-Objective Genetic Algorithm

2005 ◽  
Author(s):  
Ebenezer Seisie-Amoasi ◽  
Brian G. Williams ◽  
Marco P. Schoen
Keyword(s):  
Genetic Algorithm ◽  
Pattern Recognition ◽  
Attitude Determination ◽  
Recognition Algorithm ◽  
Control Parameters ◽  
Pattern Recognition Algorithm ◽  
Multi Objective ◽  
Star Pattern ◽  
Multi Objective Genetic Algorithm ◽  
Star Pattern Recognition

Attitude determination for unmanned spacecrafts usually employs star trackers. The specifications for these devices dictate fast, reliable, robust, and autonomous algorithms to satisfy various mission constraints. This results into simple algorithms for reduced power consumption and reduced overall weight. Optimizing a Star Pattern Recognition Algorithm (SPRA), using an imbedded star map, requires the optimization of the genetic operators that constitute the SPRA and the control parameters within the SPRA. Simultaneous optimization of the control parameters of the SPRA results into a multi-objective and multi-parameter constrained optimization problem. The optimizing of genetic algorithms is often time consuming and rather tedious by nature. In this work, a Multi-Objective Genetic Algorithm (MOGA) acting as a meta-level GA is applied together with a double objective transition selection scheme to achieve the optimization. This approach results in significantly expediting the cost assignment process. By evolving a pareto set, an optimization population element rule is determined to exist between the control parameters of the SPRA. The existence of this rule ensures effective balance between population exploitation and exploration in the algorithm estimation process. This leads to effective solutions for finding the optimum with multiple concurrent objectives while taking the constraints into consideration. Simulation results using the optimized parameters for the SPRA indicate an improvement of the recognition accuracy from less than 60% to 100% as well as a reduction of the processing time of over 2000 generations to under 250 generations at 99% precision.

Intelligent Star Pattern Recognition for Attitude Determination: the "Lost in Space" Problem

2006 ◽  
Vol 3 (11) ◽  
pp. 538-549 ◽  
Author(s):  
Lalitha Paladugu ◽  
Ebenezer Seisie-Amoasi ◽  
Brian G. Williams ◽  
Marco P. Schoen
Keyword(s):  
Pattern Recognition ◽  
Attitude Determination ◽  
Space Problem ◽  
Star Pattern ◽  
Star Pattern Recognition
Sign in / Sign up

Export Citation Format

Share Document