Unmanned system performance boundary adaptive identification method based on statistical characteristics

Abstract

The invention provides an unmanned system performance boundary self-adaptive identification method based on statistical characteristics, belongs to the technical field of unmanned system performance analysis and optimization, and aims at solving the problem that in an existing performance boundary identification method, due to the complexity of real system performance, the system performance cannot be identified easily. The method solves the problems that the performance boundary calculation process is complex and the efficiency is reduced due to the fact that k searched nearest sample sets cannot accurately represent the actual situation of an accumulation area, and comprises the steps that an unmanned system is initialized, and the same performance boundary under an unmanned condition is defined; setting a two-dimensional position space range, randomly moving the position of the obstacle, initializing a sample point, adaptively adjusting the radius taking the corresponding point as the circle center until the sample density in the circle is not increased before and after the radius is adaptively adjusted, and outputting a sample set and a performance space; and for each successful sample point in the sample set, determining a performance boundary and a performance boundary area based on a mathematical statistical method.

Images