<p>At present, there are some problems in the construction of road marking, such as low efficiency and poor precision, so using the guidance line drawing robot is an effective solution. However, satellite signals are easily disturbed by the environment, navigation accuracy and stability cannot meet the needs of robot construction positioning, and the high cost of optical fiber inertial navigation is not conducive to the popularity of the robot. Therefore, a navigation control algorithm based on low-cost GNSS/INS for guidance line drawing robots is proposed in this paper. In order to solve the problem that the heading error of the guidance line drawing robot is weak in the low speed and uniform linear motion, which leads to the divergence of heading angle, the heading information of the dual-antenna GNSS receiver is introduced as the observation quantity. In order to solve the problem that the measurement error of MEMS IMU increases when the carrier is moving, the IMU errors divergence is suppressed by non-holonomic constraint. The model prediction controller is designed to realize the smooth control of the guidance line drawing robot by predicting the future system state and the incremental constraints of the controller. The experimental results show that the improved GNSS/INS navigation control algorithm proposed in this paper shows good performance on the built platform of the guidance line drawing robot. The heading accuracy and position accuracy of the robot in the process of drawing the guidance line are improved by 46% and 40% respectively, and the guidance line drawing accuracy and smoothness both meet the construction requirements.</p>