Tire-road adhesion coefficient online estimation method and device

The invention provides an online estimation method and device for a tire-road adhesion coefficient. The method comprises the following steps: acquiring tire pressure, tire speed, tire load and a voltage signal; based on the voltage signal of the piezoelectric film sensor, using a pre-established tire grounding print longitudinal length numerical calculation model to obtain the grounding print longitudinal length of the corresponding tire; based on the voltage signal of the piezoelectric film sensor, the tire pressure and the tire speed, using a first neural network model, a second neural network model and a third neural network model which are established in advance to predict the vertical force, the slip angle, the lateral force and the aligning torque of the tire; and pre-estimating the tire-road adhesion coefficient by using a pre-established brush tire model. Based on a BP neural network technology, a numerical method and a brush tire model, the invention provides a low-cost and high-efficiency tire-road