Altruistic approaches to cooperative estimation

We present a novel approach to cooperative estimation for two subsystems possessing the same information. Instead of the two systems ultimately yielding an identical, joint estimate - here, each of the two subsystems yields a different estimate than the other. The estimators are derived based on the underlying assumption that the global mission is accomplished if at least one of the estimators succeeds in providing a satisfactory estimate. A new notion of optimality is defined, that generalizes the common definition of minimum mean square error optimality. In the Gaussian case, explicit expressions for the optimal estimators are found, that only require calculation of the largest eigenvalue of the conditional covariance matrix and its corresponding eigenvector. A simulation study demonstrates that the performance improvement, resulting from using the new approach, depends on the problem's dimension and statistical distribution.