Title:
SYSTEM AND METHOD FOR CONTROLLING MODEL AIRCRAFT
Document Type and Number:
WIPO Patent Application WO2004108524
Kind Code:
A3
Abstract:
In one embodiment, a method for controlling an aircraft (10) comprises providing an attitude error (214) as a first input into a neural controller (202) and an attitude rate (216) as a second input into the neural controller (202). The attitude error (214) is calculated from a commanded attitude (218) and a current measured attitude (220), and the attitude rate (216) is derived from the current measured attitude (220). The method also comprises processing the first input (214) and the second input (216) to generate a commanded servo actuator rate (222) as an output of the neural controller (202). The method further comprises generating a commanded actuator position (226) from the commanded servo actuator rate (222) and a current servo position (224), and inputting the commanded actuator position (226) to a servo motor (204) configured to drive an attitude actuator (206) to the commanded actuator position (226). The neural controller (202) is developed from a neural network, wherein the neural network is designed without using conventional control laws, and the neural network is trained to eliminate the attitude error (214).
More Like This:
Inventors:
FOUCHE MICHAEL J
Application Number:
PCT/US2004/012812
Publication Date:
February 10, 2005
Filing Date:
April 23, 2004
Export Citation:
Assignee:
NEURAL ROBOTICS INC (US)
International Classes:
B64C13/16; B64C39/02; G05B13/02; G05D1/08; (IPC1-7): G05D1/00
Foreign References:
| US6473676B2 | 2002-10-29 | |||
| US5553812A | 1996-09-10 | |||
| US5841537A | 1998-11-24 |
Other References:
See also references of EP 1642181A4
Download PDF: