论文信息 - Learning flight control and LoFLYTE

Learning flight control and LoFLYTE

Accurate Automation Corporation is leading the LoFLYTE advanced development program demonstrating the benefits of neural networks. This is a multi-phase effort to develop the technologies necessary to design, fabricate, and flight test a Mach 5 waverider aircraft. Phase I of the program consisted of a feasibility study. Phase II is the critical design phase to construct a subsonic remotely piloted aircraft that will demonstrate the low speed characteristics of a waverider. During Phase II, a number of innovative technologies will be validated. Included in this phase is the neurocontroller system test for this aircraft. Prototype actuator and sensor hardware has been developed and tested with our MlMD neural network hardware.

[1] Homayoun Seraji,et al. Decentralized adaptive control of manipulators: theory, simulation, and experimentation , 1989, IEEE Trans. Robotics Autom..

[2] Richard S. Sutton,et al. Neuronlike adaptive elements that can solve difficult learning control problems , 1983, IEEE Transactions on Systems, Man, and Cybernetics.

[3] Mitsuo Kawato,et al. Feedback-Error-Learning Neural Network for Supervised Motor Learning , 1990 .

[4] Donald A. Sofge,et al. Handbook of Intelligent Control: Neural, Fuzzy, and Adaptive Approaches , 1992 .

[5] Mitsuo Kawato,et al. Feedback error learning of movement by multi-layer neural network , 1988, Neural Networks.