Aiming at the problems of multiple solutions and collisions in the inverse kinematics solution of multi-degree of freedom manipulators in the obstacle environment, an optimal solution algorithm of inverse kinematics of manipulators with obstacle avoidance ability is studied, the method is composed of collision detection algorithm, shortest travel algorithm and differential evolution algorithm. Firstly, taking the six-degree of freedom manipulator as the research object, the models of a six degree of freedom manipulator and obstacles are built. The objective function for inverse kinematics solution is established, the objective function is formed by the weighted summation of the pose error function of end effector, the variation function between target angle and initial angle and the collision detection function. Secondly, the differential evolution algorithm is applied to optimize the objective function. an adaptive weight adaptive algorithm is designed to reduce the influence of the function weight on the solution speed and accuracy. Based on the adaptive algorithm, the joint angle can quickly reach the shortest travel position in the early stage of the optimization solution, and higher solution accuracy can be achieved in the latter stage of the optimization solution. So the optimal inverse kinematics solution with obstacle avoidance ability, shortest travel and high precision can be obtained. Finally, taking the UR5 manipulator as an example, the algorithm is simulated by the Robotics Toolbox in MATLAB software, and the effectiveness of the algorithm is verified.