In order to solve the problem of reduced registration accuracy due to various errors in the dual-station multi-target registration method, a minimum angle-based registration method is proposed. Firstly, the angle matrices corresponding to all the angle measurement lines in the dual stations are calculated, then the angle matrices are updated by scaling the trajectory information of each station, and finally the optimal combination of data associations is selected based on the principles of minimum angle and global optimum. The capability of the algorithm is evaluated by Monte Carlo simulation method, and the results show that the algorithm can not only enhance the accuracy of registration, but also drastically reduce computation time. When the distance between targets is 30 m, the correct rate exceeds 97%, and the algorithm's running time is approximately 0.016 ms. This algorithm outperforms other algorithms in terms of both accuracy and performance, providing an important theoretical foundation for improving the accuracy of multi-dimensional target positioning.