Search

Analyzing objects concerning their periodic behavior is mostly performed with inertial measurement units (IMUs) or global navigation satellite system (GNSS) sensors fixed to its surface. For connecting observations, sensors have to be assigned to the same reference frame in space and time as a prerequisite. Using light detection and ranging (LiDAR) observations enables contactless, time-synchronized, and spatially connected data points within a single sensor. Therefore, common signal properties are further analyzed in the spectrum to find connections and similarities between observations. Since observations are spatially continuous we can discretize them and traditionally process them. However, the time domain offers a diversity of ways to simultaneously estimate frequencies and continuously model properties at different spatial locations. Within this work, we exploit the potential of processing LiDAR data in the time domain to make use of the sensor’s contactless observations and its sampling rate in space and time. Consecutive points and their spatial neighborhoods are used to implement temporal as well as spatiotemporal connections to directly model oscillations in 2-D space. Moreover, we compute an uncertainty of estimated variables to qualify our solution. Consequently, our approach offers the opportunity to describe as well as evaluate movements and vibrations of spatially connected areas.