As a challenging high dimension optimization problem, three-dimensional path planning for Uninhabited Combat Air Vehicles (UCAV) mainly centralizes on optimizing the flight route with different types of constrains under complicated combating environments. An improved version of Bat Algorithm (BA) in combination with a Differential Evolution (DE), namely IBA, is proposed to optimize the UCAV three-dimensional path planning problem for the first time. In IBA, DE is required to select the most suitable individual in the bat population. By connecting the selected nodes using the proposed IBA, a safe path is successfully obtained. In addition, B-Spline curves are employed to smoothen the path obtained further and make it practically more feasible for UCAV. The performance of IBA is compared to that of the basic BA on a 3-D UCAV path planning problem. The experimental results demonstrate that IBA is a better technique for UCAV three-dimensional path planning problems compared to the basic BA model.