In this paper, a real-time, contactless pulse-based ultra-wideband (UWB) radar sensor was used to precisely detect the movement of the robotic arm, heartbeats, and respiration activities. To increase the object tracking resolution, an algorithm using adaptive and digitalized post-calibration techniques were proposed and tested. Both the radiated and reflected analog pulses were converted to digital signals by a low-noise comparator and its buffer for the time interval and distance detection. Moreover, the dynamic and calibrated tracking mechanism was incorporated into the algorithm for improving tracking accuracy and to continuously localize an object's vibration. The measured performance revealed that the proposed method successfully achieved μm movement precision, thus providing its excellent tracking ability. We performed three different experimental tests to demonstrate the UWB radar's viability. First, the radar was employed to detect the undesired structural vibration of a delta robotic arm, which was subjected to high acceleration and structural flexibility. The measured results indicated that the accuracy of 100 μm can be achieved at a 10 Hz vibration frequency. Next, the apnea alarm was tested, which included an embedded sensitive sensor in the smart mattress in order to execute longterm physiological monitoring. The apnea was mitigated by autoadjusting the patient's sleeping posture on the mattress when the UWB sensor detected the apnea condition. The final experiment using the radar sensor was to monitor human vital signs, such as heart and breathing rates, simultaneously. In comparison with a commercial electrocardiography instrument, this radar sensor measured 73.2 bpm with only 1.2% deviation. Our results shown that the contactless UWB sensor performs superbly and is wellsuited for monitoring physiological parameters.