Aiming at the problem of poor control precision of excavator tool track, an excavator tool trajectory control method based on improved RBF neural network was proposed. The output response characteristics of excavator working device under pure mechanical control strategy and RBF neural network control strategy based on S-Function were analyzed. RBF neural network tool trajectory control strategy based on genetic algorithm was proposed. Through the horizontal mining simulation experiment, the control precision of tool trajectory was analyzed experimentally. The results show that the control method of excavator tool trajectory based on RBF neural network proposed in this study is about 10 mm more accurate than conventional PID control, which provides reference for further realization of excavator automation.