A collaborative control method based on PSO-BP neural network was proposed to solve the problem of uneven distribution of the composite sealing system due to the wide difference in size between fiber and gravel, which realized the high precision control of the composite sealing system. In the composite sealing system, proportional-integral-derivative（PID）controller is used to control the amount of fiber, asphalt and gravel. BP neural network is introduced to optimize the PID parameters in real time, and particle swarm optimization（PSO）algorithm is used to adjust the initial weight and threshold of BP neural network, which effectively improves the overall control performance of the system. Through Matlab/Simulink simulation experiments, the comprehensive performance of the proposed method is compared with that of traditional PID controller and BP-PID controller. The results show that the proposed controller has obvious advantages in response speed, overshoot and steady-state error during the spreading process. The real vehicle test verifies the spreading accuracy of the proposed control method, and successfully realizes the high-precision cooperative control of fiber, asphalt and gravel in the working process of the composite sealing machine.