Abstract: To solve problem that full hydraulic braking system of trackless tyred vehicle was prone to failure in harsh environment of coal mine, a fault diagnosis method based on support vector machine was proposed. The support vector machine regression estimate algorithm was applied to establish fault prediction model of the full hydraulic braking system, while pattern classification algorithm was used to isolate fault, and fault diagnosis tests and isolation experiments were carried out in Matlab environment. The experiment results show that it is entirely feasible to use fault diagnosis method based on support vector machine to diagnose fault of full hydraulic braking system of trackless tyred vehicle, which can improve fault detection efficiency and diagnostic accuracy.