The Study Of The Fuzzy Control Algorithm Of Autonomous Navigator Sequence Learning

Autonomous aircraft can be used in important areas such as water quality monitoring,ship inspection and marine environmental exploration,and has a very broad application prospect.It has attracted widespread attention from environmental protection agencies,defense agencies,and mining authorities in various countries around the world.Because the inertia of the water flow makes the autonomous navigation of the aircraft on the water surface different from the autonomous driving of the vehicles on the land,the aircraft will easily produce a serious overshoot in the process of autonomous navigation on the water surface,and the navigation route of the aircraft will not be ideal.The operating state is cruising;in addition,the control response delay of the aircraft also has an important influence on the operational effect of the aircraft.In a complex water environment,the aircraft is relatively affected by the water environment.At present,the research results of the aircraft control algorithm are few and simple,and it is very difficult for the aircraft to achieve the desired navigation effect.Therefore,designing an autonomous navigation control algorithm adapted to a complex surface environment has important theoretical and engineering implications.The main research contents and innovations of the paper are as follows:(1)This thesis studies the problem of aircraft state prediction and proposes an aircraft state prediction algorithm based on sequence learning.The method uses the state values of n aircraft as a sequence to predict the state value of the next time:firstly,the operating state of the aircraft is collected for the first t-I time,and the online sequence limit learning machine is initialized;The unit enters the aircraft’s state value as a sequence into the online sequence extreme learning machine for predictive learning,and outputs the prediction result of the aircraft state value at the next moment;finally,the on-line sequence extreme learning is continued with the continuous updating of the aircraft state value.The machine’s prediction model is also updated accordingly.(2)In this dissertation,an aircraft variable-domain fuzzy control algorithm is proposed based on the aircraft state prediction value obtained by the above aircraft state prediction algorithm.Combining the aircraft state prediction value with the variable universe coefficient to carry out the fuzzy control algorithm,the control output of the aircraft at the next moment is obtained,which realizes the autonomous navigation of the aircraft without deviation and improves the control accuracy of the aircraft.(3)In this dissertation,a water environment model suitable for aircraft movement was designed.A large number of simulation experiments were conducted to obtain the optimal parameters of the algorithm under the water environment model,and the optimal parameters in the simulation experiment were used in the actual environment.Comparing with the traditional aircraft autonomous navigation control algorithm performance experiments,the effectiveness,low complexity and high accuracy of the proposed algorithm are verified.