Abstract: To solve the technical problem of low efficiency and difficulty in designing wellbore trajectories for cluster well platforms, and effectively improve the design efficiency of unconventional oil and gas cluster well platforms, the matching problem of cluster well targets are abstracted as a task allocation problem. Based on the KM algorithm, a target matching objective function is established, and the matching order is determined according to the horizontal displacement weighting value from the wellhead to the target, forming an automatic matching method between the wellhead and the target. A guided heuristic function is created based on the A* algorithm, a fence search environment is constructed, and the search direction is optimized to form a guided intelligent obstacle avoidance method used in cluster wells. The design strategy for cluster wells is generalized and the relevant design software is developed. With this software, the average time for designing the track of a 4-wellole cluster well platform on site is 8.9 seconds, the average time for designing the track of a 6-well platform is 9.5 seconds, the average time for designing the track of an 8-well platform is 36.6 seconds, and the average time for designing the track of a platform with 10 or more wells is 52.5 seconds. Compared to conventional design techniques, the time for design of platform with 5-8 wells has been reduced from approximately 3 days to 1.2 days, and the time for design of platform with 9-13 wells has been reduced from approximately 6 days to 2.4 days, showing 60% of design time saved. The conclusion and suggestion have solved the problem of long drilling design time and low efficiency in cluster well platforms, providing reference for the application of intelligent algorithms in other professional fields in the petroleum industry.