Fuzzy reasoning for multisensor management

A multisensor management scheme is proposed in this contribution, intended to be applicable to a generalized suite of several multitask sensors, deployed for multitarget surveillance purposes. The objective of the multisensor management function is to optimize, in a coordinated fashion, the measurement process of each sensor using overall system performance information. The coordination manager will provide each sensor with a list of sensor-level tasks, ranked according to their priorities, which are in turn obtained from the inferred global system-level necessity of fulfilling the task. Each sensor-level task will be also accompanied with an indication of the sensing performance objective desired to achieve with it, expressed in different terms depending on the type of task. The management process will be performed by means of a symbolic reasoning process starting from the output information of the data fusion and situation assessment functions. Fuzzy reasoning strategies and possibility theory concepts, in conjunction with heuristic search techniques, are applied to the processes of system-level task necessity evaluation, sensor-level task priority computation, specification of the sensing performance objectives (desired to guarantee with each task) and sensor(s)-to-task(s) assignment.