To solve problems of heavy calculation burden and low solution accuracy for redundant robot in inverse kinematics problem,a solution method based on improved fruit fly optimization algorithm（IFOA） was proposed. On the basis of FOA,the evolutionary equation is optimized by adding an improved strategy of learning worst individuals. The ability of the IFOA to break away from the local optimum and to find the global optimum is greatly enhanced. Experimental results of several typical functions show that IFOA has better global search ability,faster speed,higher accuracy and reliability,compared with FOA. In the application of inverse kinematics problem of the redundant robot,the accuracy,speed and stability were improved effectively,and is thus applicable to solve the inverse kinematics problem of redundant robot.