There exists track swap when tracking closely spaced targets,which may decrease the reliability of traditional track association and fusion algorithms greatly.Thus we proposed a robust track association and fusion algorithm with tracking state monitoring.Firstly,a sliding window global optimum association was adopted to ascertain associated track pair and establish system tracks with multiple frame track data.Then the real-time association relation of associated track pair was used to detect track swap.Lastly,track attenuated residual was used to identify maneuver in order to select the fused measurements or fused state estimation adaptively in updating state of system tracks.Simulation result shows that the proposed algorithm can improve accuracy of fusion tracks and realize robust track association and fusion.