Abstract: | Aiming at the problem of target identification with unknown antecedent reasoning rules, and considering the multi-source and heterogeneous measurement of target information, fuzzy sets are used to describe the multi-dimensional feature information such as speed and image information in a unified abstract way. On this basis, Genetic algorithm is used to optimize fuzzy inference rules. On the premise of determining the fuzzy partition interval, the optimal membership degree is obtained by training and the optimal reasoning rule base is established. In addition, aiming at the measurement uncertainty caused by the limited detection accuracy of sensors, and drawing lessons from the fuzzy classification of interval sample data, a genetic optimization algorithm based on type-2 fuzzy inference system is proposed after the construction of type-1 fuzzy inference system. The derivation process of constructing the corresponding type-2 fuzzy set by embedding trapmf membership function is given and a type-2 fuzzy inference system satisfying the compound input of point value and interval value is designed. At last, the feasibility of the inference system is verified by the simulation. |