Wind and Payload Disturbance Rejection Control Based on Adaptive Neural Estimators: Application on Quadrotors

Pseudocode 1

 Net=configureNet()

 Dataset=

 For i=1 totoff   #Generating the dataset for off-line learning

   inputNet=[out1(i-1),… out1(i-n1), out2(i-1),…, outM(i-1)…. outM(i-nM)]

   ParM(i)=model(out1(i-1),….., outM(i-1)])

   Element=parM(i), inputNet

   Dataset=Dataset Element

 endFor

 Net=offlineTraining(Net, Dataset)

 For i=toffto tend  #on-line learning

   inputNet=[out1(i-1),… out1(i-n1), out2(i-1),…, outM(i-1)…. outM(i-nM)]

   ParS(i)=simulate(Net, inputNet)

   [out1(i),….., outM(i)]=executeUAV(ParS(i))    #Execute UAV + Controller

   ParM(i)=model(out1(i-1),….., outM(i-1)])

   IfParM(i) then

     Element=parM(i), inputNet

     Net=onlineTraining(Net, Element)

   endIf

 endFor