Abstract: Researching on tribological behavior is of great significance for understanding the interfacial friction and wear characteristics, as well as revealing the generation mechanism and evolution manner of friction-induced vibration and ensuring the reliable operation of the friction system. Due to the disturbance of interfacial friction vibration response caused by the rigid connection of tribological behavior simulation experimental equipment, it is generally difficult to investigate the tribological behavior and evaluate material wear performance purely at the friction interface. Therefore, it is necessary to eliminate the coupled vibration between the frictional interface and the mechanical connecting component. For this purpose, a tribological behavior simulation experiment device which achieves vibration decoupling using air bearings was developed in this study. Hammer tests were carried out on the T-shaped air-bearing configuration of the loading system and the air-bearing stage of the drive system, respectively. A complete modal analysis of the vibration decoupled experimental device was conducted using finite element software, and the results were compared with hammer tests to fully illustrate the effectiveness of the vibration decoupled function of the experiment device. Finally, tribological comparative tests of “plane-to-plane” contact were carried out on the vibration decoupled experimental device and a vibration coupled experimental device (CETR UMT-3) to further validate the vibration decoupled capabilities of the experimental device and to demonstrate the differences between vibration decoupled/coupled tribological test equipment. The hammer test showed that several natural frequencies were detected at the bushing of the air bearing in the deflated state, while only a low single frequency was found in the inflated state, being the natural frequency of the air-bearing bushing. The results of the finite element modal analysis revealed that the first mode of the experimental device was dominated by torsion in the vertical direction of the T-shaped air-bearing configuration. The maximum error between the modal frequency and the hammering test was 8.6%. Due to the small difference in the absolute values of the frequencies, the modal analysis corresponded well to the hammer test results. This demonstrated that the air-bearing could effectively isolate the vibration transfer from the connected components and achieve vibration decoupling. The hammer test and modal analysis clearly illustrated the effectiveness of the air bearings in isolating the transmission of vibration from the connected components, and achieved vibration decoupled. The tribological comparative tests found that with the increase of the normal loading force or the reciprocating frequency, the root-mean-square value (RMS) rate of change of the vibration acceleration increased linearly for the vibration decoupled device, while the RMS rate of change increased first and then decreased for the vibration coupled device, which showed a significant difference between the two kinds of experiment devices with respect to vibration response. Therefore, the tribological behavior simulation experiment device had successfully decoupled the frictional interface from the other mechanical connecting component, providing a new idea and an effective means to investigate the influencing factors and the evolution manners of the interfacial tribological behavior further precisely, revealing the generation mechanism of friction-induced vibration, as well as avoiding the influence of experimental equipment when comparing and evaluating the friction and wear performance of different materials.