Realtime Control of Multiple Sensor Systems.

摘要

The technical objectives of this project are the design and demonstration of methods for the optimal management of sensor systems. As a specific application, we consider the nonlinear filtering of a vector diffusion process, with several noisy vector observations. Any number of sensors can be utilized in the signal processing performed by the nonlinear filter. The problem considered is the optimal selection of a schedule of these sensors from the available set, so as to optimally estimate a function of the state at some given time. Solution of the optimal schedule is derived from solving a system of quasi-linear inequalities (QVIs). The schedule derived from these methods is optimal with respect to predetermined, but possibly varying, running and switching costs for the sensors, as well as some cost functional of the estimate of the dynamics. Methods include precise strategies for adapting the sensor configuration to the engagement at hand, including optimal timing of decisions to communicate or activate active sensors which reveal position. These strategies have been developed by TSI in the form of feedback control laws which are computationally efficient and simple to implement.... Sensor systems, scheduling, realtime control.