Design and development of an acoustic sensor array for anomaly detection

Abstract

Anomaly detection is the identification of an anomaly with regards to misplaced objects or intruders in an indoor environment. It can be done by mapping a room along with its stationary objects with an ultrasonic acoustic frequency response. In this paper, we explore the feasibility of using active acoustic imagery in air and simulate phased array beam forming techniques to achieve a suitable acoustic sensor array design for a portable mobile robot which can be applied to detect the presence/absence of anomalous objects in a room. The selected room is insonified with a broadband acoustic signal and the multi-path reflection effects contained in the reflected signal time-series in enclosed rooms are analyzed to detect anomalies. The related hardware is designed with the same feasibility criterion that the developed system needs to be deployed on a portable mobile robot. There is a trade-off between image resolution and range with the array size, number of elements and the imaging frequency and has to be iteratively simulated to achieve the desired acoustic sensor array design. The designed acoustic imaging array system is targeted for use in surveillance missions for intruder alerts (anomaly detection) and imaging objects during dark and smoky scenarios where conventional optic based systems do not function very well.