Anonymous Functions - how to store the results of anonymous functions

rng(1) % important for technical reasons
U = [ 1 -1 3 4 ]; % an artificial input sequence
% applyPredictionUpdate returns the predicted state:
applyPredictionUpdate = @(bel, u_t) [bel '.PU(' num2str(u_t) ')'];
% readSensor returns the current sensor's reading:
readSensor = @() [ 'Z(' num2str(round(20*rand())) ')' ]; 
% applyMeasurementUpdate returns the state updated by the measurements:
applyMeasurementUpdate = @(sensorReading, belBar) [ belBar '.MU(' sensorReading ')' ];
bel = 'PRIOR'; 
%%%%%%%%%%% PLEASE DON'T CHANGE ANYTHING ABOVE THIS LINE %%%%%%%%%%%
for u_t = U
belBar=@(bel, u_t)[bel,u_t];
sensorReading=readSensor();
bel=applyMeasurementUpdate(sensorReading,belBar); 
end

Got the foll. error:

Error using horzcat

The following error occurred converting from char to function_handle:

Too many output arguments.

Error in solution>@(sensorReading,belBar)[belBar,'.MU(',sensorReading,')'] (line 8)

applyMeasurementUpdate = @(sensorReading, belBar) [ belBar '.MU(' sensorReading ')' ];

Error in solution (line 14)

bel=applyMeasurementUpdate(sensorReading,belBar);

" applyPredictionUpdate does not change bel in-place but returns the predicted state, so make sure to store it in a local variable that you can use in applyMeasurementUpdate."

[Stephen23]

As far as I can tell, defining belBar as a function handle is a red-herring.

rng(1) % important for technical reasons
U = [ 1 -1 3 4 ]; % an artificial input sequence
% applyPredictionUpdate returns the predicted state:
applyPredictionUpdate = @(bel, u_t) [bel '.PU(' num2str(u_t) ')'];
% readSensor returns the current sensor's reading:
readSensor = @() [ 'Z(' num2str(round(20*rand())) ')' ]; 
% applyMeasurementUpdate returns the state updated by the measurements:
applyMeasurementUpdate = @(sensorReading, belBar) [ belBar '.MU(' sensorReading ')' ];
bel = 'PRIOR'; 
%%%%%%%%%%% PLEASE DON'T CHANGE ANYTHING ABOVE THIS LINE %%%%%%%%%%%
for u_t = U
    sensorReading=readSensor();
    bel=applyMeasurementUpdate(sensorReading,[bel,u_t])
end

bel = 'PRIOR.MU(Z(8))'

bel = 'PRIOR.MU(Z(8)) .MU(Z(14))'

bel = 'PRIOR.MU(Z(8)) .MU(Z(14)).MU(Z(0))'

bel = 'PRIOR.MU(Z(8)) .MU(Z(14)).MU(Z(0)).MU(Z(6))'

But the best solution would be to not write code that writes code like that.

Answer 1

[David Hill]

for u_t = U
  belBar=applyPredictionUpdate(bel, u_t);
  sensorReading=readSensor();
  bel=applyMeasurementUpdate(sensorReading,belBar); 
end

[Ken]

Thanks. Works!

Answer 2

[Walter Roberson]

belBar=@(bel, u_t)[bel,u_t];

That is a function handle

applyMeasurementUpdate = @(sensorReading, belBar) [ belBar '.MU(' sensorReading ')' ];

You are requesting to concatenate the function handle and a character vector.

belBar is a function expecting two inputs, but you are trying to wrap it in a function that takes inputs that do not include the two inputs belBar needs, so you are going to have trouble.

I think it might make more sense to use

belBar = [bel,u_t];