Information Based Biochemical Source Localization
We want to develop and analyze information based control algorithms for single robots and groups of robots equipped with chemical concentration sensors. Information based control algorithms are ones in which the robot(s) move to the location(s) where the information gathered by the sensors is maximized. To this end we have achieved the following goals:
Developed an information based control algorithm for a distributed sensor network.
Developed computationally efficient numerical procedures to approximate maximum likelihood estimation.
Developed an information based motion algorithm, the modified Fisher Information Matrix (FIM) motion algorithm, and showed that for a certain family of concentration functions, it has desirable stability properties.
Applied the modified FIM motion algorithm to a circular biochemical source, and showed that the equilibrium point of the system is unique.
This research is supported by NSF grant CCR-0330342