When the intelligent weeding robot was working in the lawn, it was easy to be affected by external disturbance and system uncertainty, which led to long tracking convergence time and poor tracking effect. Therefore, an adaptive fast integrating terminal sliding mode control algorithm for trajectory tracking was designed. Firstly, the dynamics model of the weeding robot was established by considering the dynamic characteristics of the driving wheel and uncertainties such as unmodeled errors, external interference and dynamic and static friction. Then based on the established dynamic model, an adaptive fast integrating terminal sliding mode controller was designed. The proposed controller combined the advantages of fast terminal sliding mode, integral sliding mode and adaptive estimation technology to achieve the desired tracking performance and suppress control signal jitter. At the same time, without specifying the upper bound of the system uncertainty and external interference, the designed adaptive estimation can be used for realtime compensation to improve the robustness of the system. Finally, the effectiveness of the proposed method was verified by simulation and experiment. The experimental results showed that the designed controller can make the tracking error converge quickly in a limited time, and the absolute value of the lateral error was no more than 0.0979m, the absolute value of the longitudinal error was no more than 0.1026m, and the absolute value of the heading angle error was no more than 0.0578rad, which can ensure the robot to track the working path accurately and have strong robustness.