Configuration of a sparse network of LIDAR sensors to identify security-relevant behavior of people

摘要

Surveillance is an important application of sensor networks. In this paper it is demonstrated how a sparse network of stationary infrared (IR) sensors with highly directional, stationary beam patterns based on the LIDAR principle can be used to reliably track persons. Due to the small number of sensors and their narrow beam patterns a significant portion of the area to be surveilled is not directly assessed by the sensors. To nonetheless achieve reliable tracking of moving targets in the entire area to be monitored, we employ the most appropriate sensor network configuration and propose a probabilistic tracking approach. The behavior of a person moving through the area of observation is classified as "normal" or "abnormal" depending upon the trajectory and motion dynamics. The classification is based on a linear Kalman prediction.