Abstract: Aiming at the maneuvering targets tracking problem for visual servoing robot，the “current” statistical model will be applied to the IMM algorithm，and it will be composed with Constant Velocity（CV） model.The Monte Carlo simulation is carried out using Matlab software，which is based on the two kinds of IMM algorithm.The results show that the capability of the “current” statistical IMM algorithm is better.It not only has the good astringency and the stability，but also improves the veracity and rapidity of the maneuvering targets tracking of visual servoing robot.

Key words: “current ”statistical model, Constant Velocity（CV） model, maneuvering targets, Interactive Multi-Model（IMM） algorithm

摘要： 针对视觉伺服机器人对机动目标的跟踪问题，将当前统计模型（CS）引入IMM算法，与匀速模型（CV）组成模型集。在Matlab上对当前统计IMM算法和基本IMM算法进行仿真比较，结果表明当前统计IMMS算法跟踪机动目标的性能好于基本IMM算法，具有很好的收敛性和稳定性，提高了视觉伺服机器人对目标跟踪的准确性和快速性。

关键词: “当前”统计模型, 匀速（CV）模型, 机动目标, 交互式多模型算法