There are many types of equipment and complex failure forms in the fracturing well field, it is difficult and inefficient to inspect manually, so unmanned aerial vehicle(UAV) was applied to inspect the fracturing well field instead of manually. In order to enable UAV to perform effective path planning and achieve inspection purposes in the fracturing wellsite environment, a comprehensive and improved rapidly-exploring random trees(RRT) algorithm was proposed for UAV path trajectory planning. Firstly, an improved artificial potential field method was added to improve the extension direction of the new node, and the random tree was extended more purposefully by the combined force of the target point and the obstacle, thus improving the efficiency of UAV path planning. Secondly, a pruning strategy was added to shorten the length of the planning path by deleting the redundant nodes and folded segments of the random tree during the expansion process. Thirdly, Bezier curve optimisation was done on the planned trajectory so that the planned trajectory was more in line with the motion characteristics of the UAV, and in order to prevent the optimised path trajectory from colliding with the obstacles, an inflated obstacle processing was done. Finally, based on robot operating system (ROS), simulation experiments were carried out for UAV inspection of fracking well sites. Simulation experiments show that the path generated by the integrated and improved RRT algorithm is more in line with the motion characteristics of UAVs, has the advantages of short paths and less time-consuming, and can inspect the equipment in the fracturing well site while avoiding obstacles.