COMPARISON OF KALMAN-TYPE ALGORITHMS IN NONLINEAR NAVIGATION PROBLEMS FOR AUTONOMOUS VEHICLES

The efficiency of using a linearized, an iterated and an optimal linear algorithms for the solution of nonlinear navigation problems is analyzed. The emphasis is on the problem of deriving an adequate accuracy characteristic. The comparison of is carried out by the example of solving problems of navigation by beacons and terrain-aided navigation with the use of a map. The results are given, illustrating the advantages of the linear optimal algorithm both from the accuracy standpoint and the standpoint of adequacy of the covariance matrix being generated.