Black-box optimizer that uses JCobyla as the solver
<p> Performs mutual information-based optimization using a zeroth-order Gaussian approximation
to the entropy expression that is approximated over a subset of particles. See references for
further details on the theory.
<b>References</b>
<p>[1]
B. Charrow, V. Kumar, and N. Michael <i>Approximate Representations for Multi-Robot
Control Policies that Maximize Mutual Information</i>, In Proc. Robotics: Science and Systems Conference (RSS), 2013.
Ilge Akkaya, Shuhei Emoto
$Id: MutualInformationCalculator.java 70402 2014-10-23 00:52:20Z cxh $
Ptolemy II 10.0
Red (ilgea)
The computed mutual information between particle sets.
Particles input that accepts an array of record tokens. One field of the record must be labeled as "weight".
Other fields will be resolved to state variables.
The locations of the pursue robots that are producing the particle estimate.
The control input that determines the update in robot position.
Measurement noise covariance.
Time horizon over which the mutual information will be calculated.
The index of the robot, over which the mutual information is being optimized.