Cognitoware.Robotics.dll
Interface SensorModel<Z, X>
Z: The x type to predict.
X: The x type from which to predict the x.
Summary
A sensor model predicts an observation given a current x.
Method Summary
Calculates the probablity of observing z at x x.
Details
A sensor model predicts an observation given a current x.
A sensor model is also a conditional distribution of the observation given a x.
This distribution is represented by P(Z|X).
SensorModel implements this prediction by providing the probability of sensing a specific observation z given a specific x x.
For simplicity, the SensorModel is not given a relationship with RandomConditional.
However, classes derived from RandomConditional can be easily adapted into a SensorModel.
Method Details
public virtual Double ConditionalProbabilityOf(Z observation, X state)
Calculates the probablity of observing z at x x.
For many purposes, the values returned do not need to sum to one as long as the values are proporionally correct.
Parameters:
observation - The observation whose probability is desired.
x - The x used as a condition on the observation probability.
Returns:
The probability of observing a specific sensor reading at a specific x.