The paper presents an intelligent diagnosis method for electrical faults of in-wheel motor on the basis of an adaptive neuro-fuzzy inference system (ANFIS). Familiar electrical faults of in-wheel motor include inter-turn short circuit, phase short circuit, high impedance etc. two symptom parameters (SPs) of each phase current are defined to represent electrical faults, and ANFIS is used to fuse all SPs from different sensors, then to establish the intelligent diagnosis system. In order to demonstrate the effectiveness of the proposed method, experiments are carried out on the in-wheel motor test bench, on which each phase current is monitored to extract the features of electrical faults. The experiment results show that the proposed method is efficient.