In order to solve the problem of poor control robustness of multi degree of freedom packaging robot control under the background of high complexity of human-computer interaction content and form and difficult to obtain depth information, a new multi degree of freedom packaging robot human-computer interaction control method is proposed. The Kinect sensor is used to obtain the information of multi degree of freedom packaging robot. On this basis, the complexity of human-computer interaction is calculated by information entropy method, and the corresponding human-computer interaction control system is constructed. The dynamic model of human-computer interaction is established based on Lagrange equation, and the depth of human-computer interaction information of multi degree of freedom packaging robot is obtained; An adaptive fuzzy controller is used to control the human-computer interaction process and interface of multi degree of freedom packaging robot. The six degree of freedom packaging robot is selected as the experimental object to determine the optimal free parameters and carry out the simulation experiment of human interaction control. The experimental results show that the average action completion rate of the multi degree of freedom packaging robot is 89.6%, the average packaging time is 1.542 min, and the control robustness is better. The control method greatly improves the robustness of the robot control, and effectively improves the packaging efficiency of the packaging robot.