Based on the wolf pack algorithm, the update rule of scout wolf is proposed, and a phase factor is added to its formula in order to improve the flexibility of scout wolf. To enhance the circumvent ability of ferocious wolf, the concept of siege radius is defined, and its calculation method is given. The change of siege radius can evaluate the ability of escaping from the local optima. The attack step kind is optimized, and a formula new way of position update is designed. On the basis of the above improvements and chaos optimization theory, a framework of the wolf pack algorithm is constructed. The comparing experiment with the traditional wolf pack algorithm on the benchmark functions and path planning problems verifes the effectiveness of the improved algorithm.