How to improve the speed and accuracy of location information is one of the key issues in current satellite navigation and positioning. This paper proposes a satellite selection algorithm based on the projection of the Delaunay triangulation. The main idea of the algorithm is :Projecting the Delaunay triangulation through the in-plane polyhedral projection method to quickly calculate the volume of the ball inscribed polyhedron. The relationship between polyhedral volume and GDOP is verified. The experimental calculations show that when the number of visible satellites participating in the solution is less than 9, with the increase of polyhedron volume, the GDOP value decreases significantly. When the number of visible satellites reaches 9, the GDOP value does not change significantly with the increase of the polyhedral volume, but the solution aging is significantly reduced. Conclusion: Among the many visible satellites, 9 satellites with better spatial geometric distribution are selected as the solution satellites, and the navigation and positioning accuracy is high, and the solution aging is taken into consideration.