Regulation for the testing and operation of automated driving vehicles on public roadways has been recently developed all over the world. For example, the licensing standards and the evaluation technology for automated driving vehicles have been proposed in California, Nevada and EU. Like M-city, the test bed has been developed worldwide to evaluate automated driving vehicles and K-city has been developed as the test bed in Korea. The K-city has a variety of facilities such as merge, ramp, tollgate, tunnel, intersection and so on according to five road conditions: motorway, suburban road, urban road, community road and valet parking zone. Therefore, it is necessary to have automated driving evaluation scenarios that would actually be implemented in K-city. For scenario implementation as vehicle tests, automated driving vehicles are needed. The safety evaluation scenarios and criteria for level 3 and level 4 automated driving vehicles were developed in consideration of the actual driving conditions, the real road driving data, and the existed automated driving evaluation methods (ISO, NHTHA, and so on). Then, the safety evaluation scenarios were determined by considering whether the test bed could be realized, repeatable, and safety performance could be well assessed. In order to evaluate automated driving on actual test bed, vehicles as well as evaluation scenarios are needed. First of all, it needs an automated driving vehicle that is evaluated. The evaluated vehicle is called as a subject vehicle. Besides the subject vehicle, vehicles that can help evaluation are also needed and are referred to as target vehicles. Target vehicles have the ability to driving autonomously in accordance with the scenario, maintaining safety performance with the surrounding vehicles, and recognizing subject vehicle to measure safety criteria. For example, the target vehicles are used to produce a variety of situations, such as cutting in front of the subject vehicle, decelerating in front of the subject vehicle, or driving on the main road when subject vehicle is on the merge road. In the study, a subject vehicle and four target vehicles have been developed and utilized. To verify the developed evaluation scenarios, vehicle tests were conducted using subject vehicle and target vehicles. The subject vehicle has level 3 and level 4 of automated driving that have diverse functions such as lane keeping and lane change. In accordance with the scenarios, all vehicles were self-driving and the subject vehicle was checked whether it meets the evaluation criteria. Through vehicle tests, the developed evaluation scenario was verified to be feasible on the test bed and to evaluate the performance of subject vehicle well. In this paper, the vehicle test result of merge scenario is presented among various motorway scenarios.