Abstract: Geometric constraint solving can be transformed into optimization problem which is non-linear and multi-variable. Geometric constraint solving based on artificial immune algorithm and improved chaos search strategy is proposed in this study. The local optimal solutions obtained by artificial immune algorithm are used as the heuristic information and the global best solution is searched by improved chaos search strategy in the neighborhood of local optimal solutions. In order to enhance precision and searching speed, chaos search area is controlled in the neighborhood of local optimal solutions by reducing search area of variables. This algorithm differs from current optimization methods in that it gets the global best solution by excluding bad solutions. Experiment results show that the proposed method is better than artificial immune algorithm and can deal with geometric constraint solving efficiently.