Abstract: Aiming at the deficiency of butterfly optimization algorithm (BOA), an improved butterfly optimization algorithm (IBOA) was proposed and applied to the inverse kinematics of redundant manipulators. The IBOA algorithm integrates three strategies, namely dynamic transformation probability strategy, optimal neighborhood disturbance strategy and random inertia weight strategy, to improve the convergence speed and convergence accuracy of the algorithm, and overcomes the problems that BOA algorithm is easy to fall into local optimal and the late convergence accuracy is not high. The test results of the three benchmark functions show that IBOA is superior in solving accuracy, solving speed and computational stability. The application results of kinematic solution of redundant manipulator obtained by IBOA show that the manipulator has smaller pose error, less calculation time and better solution stability.