We designed an integrated system of an UGV and an UAV, and propose a sensor-based relative pose estimation method for this system. Most of service robots are wheeled robots which can be easily controlled. However, due to the development of aerial vehicle technologies including sensors and wireless communication, recent UAVs can have lower weight, longer flight time and stabilized control system. Emerging an UGV and an UAV for certain service can be a better choice for personal or public services, especially for effective surveillance. In this paper, a commercial micro-UAV and a low-price, sensor-equipped mobile base are integrated to provide surveillance services. A fish-eye camera and an ultrasonic range finder sensor are utilized to estimate relative translation between two robots. Attitude and heading reference systems in both robots are utilized to estimate relative rotation between them. The probabilistic filter is applied to compensate sensor or camera measurement noises. The robustness of the proposed system is verified by a quad-camera stereo capture system. Also, practical application is provided with a scenario which performs the homing of the UAV from arbitrary location.