This paper proposes the concept of automated driving vehicle failsafe system structure. It contains vehicle hardware and software structure design for automated driving vehicle failsafe system. Moreover, it handles the contents fail detection, fault-tolerant control, and emergency braking strategy in case there is no driver intervention in the fail condition of automated driving vehicle. According to the 2017 'AUTOMATED DRIVE SYSTEM 2.0:​ a vision for safety' report released by the NHTSA, it states that deployment of the crash avoidance system is essential to switch to a minimum hazardous condition in the event of a problem with the self-driving vehicle, or the system cannot operate safely. First, the method used to build the hardware & software of the vehicle was based on the guideline of ‘AUTOMATED DRIVE SYSTEM 2.0:​ Section 1 fallback (Minimal Risk condition)’ report released by NHTSA. Second, a method of an algorithm is sliding mode control based fault tolerant control and emergency deceleration control which designed to target SAE International standard J3016 autonomous driving phase 4:​ automated driving system perform ass aspects of the dynamic driving task, even if a human driver does not respond appropriately to a request to intervene. In this paper, to meet the requirements of autonomous driving phase proposed by SAE International standard J3016 phase 4 and NHTSA safety standard, the hardware configuration was created to ensure that the automated driving vehicle could perform the given task without proper driver intervention. In detection part, hardware (Actuator, Sensor, CAN signal, Upper&Lower controller) and module based failsafe diagnosis method and algorithm were proposed to detect fail condition. In decision and control part, when a failure of an automated driving vehicle is diagnosed, and no driver intervention was detected, the automated driving vehicle failsafe phase is a move to the system error. In the phase of the system error (lower controller), proposed methodologies are utilized. Automated driving vehicle experiments have demonstrated the algorithms as mentioned earlier and failsafe structure. First of all, it is true that not many papers and studies have been done on the failsafe system of an automated driving vehicle. NHTSA's safety report of an autonomous vehicle only contains a "suggestion" that says, "It is a good thing to do this," and has not yet created a rule. However, this paper proposes an automated driving vehicle failsafe system that is not commercialized but has been configured to meet NHTSA's requirements to take into account safety. The proposed failsafe system is applied to the automated driving vehicle, and the vehicle experiment was completed with the proposed algorithm. The proposed system is considered to be very compatible with the subject of the technical session by suggesting the system that meets the NHTSA standards as well as testing control and emergency systems targeted automated driving vehicle phase 4.