In this paper, we propose an immune evolutionary algorithm (IEA)-based approach to determine the pose of a parallel manipulator from the knowledge of a given set of points in object space. We choose several sets of point in which the configuration and the numbers of point are different. We study the influence generated by the different numbers and configuration of points in object space, and study the performance of the algorithm when an image sequence is used as experiment data sets. Experiment results show that the IEA-based algorithm is free of the configuration of points, and its calculation accuracy will be improved when the number of points increases. It overcomes two problems: "ignore parameter of the geometric constraint" and "need better iteration of the initial value problem". It has fast convergence, high precision and good robustness.