First, established submersible vehicle movement mathematic model; then analyzed disadvantage of PID autopilot effect. Mainly, a multi-controller method with a neural network control and PID control was adopted, and researched the submersible vehicle autopilot control technology, then a neural network identification(NNI) was designed, and to identify the submersible vehicles space mathematical models on line, though the submersible vehicle mathematical models can not fully determine, the submersible vehicles approaching mathematical models also can be identified on line by NNI through the real time input and output of the submersible vehicle under large interference. At the same time, the multi-layer prior neural network as neural PID controller (NNC) was adopted, and it improved accuracy. The simulation results show that neural network PID control autopilot has very good performance and is better than traditional PID autopilot in robustness and practicability. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Yang, J., Li, B., & Tao, H. (2010). Research of some autopilot controller based on neural network PID. In Lecture Notes in Electrical Engineering (Vol. 72 LNEE, pp. 161–170). https://doi.org/10.1007/978-3-642-14350-2_20
Mendeley helps you to discover research relevant for your work.