提出了一种基于模糊竞争网络的主动声纳目标分类器。该分类器结构简单，学习速度快，仅用少量样本进行训练便可获得良好的推广性能．便于实时学习和应用。通过对模式属于各类别的隶属度的分析，可以知道判决的可靠性，具有较好的应用价值。 An active sonar target classifier based on fuzzy competition network is proposed. The classifier has the advantages of simple structure and fast learning speed, and good promotion performance can be obtained only by training with a small amount of samples. Easy to learn and apply in real time. By analyzing the degree of membership belonging to each category, we can know the reliability of the judgment and have good application value.