Ultrasound-based active hearing techniques for tracking and identification of objects

The author describes the main characteristics of a sensor array configuration as a preliminary step in investigating the potential benefits of novel sonar configurations. He investigates, both theoretically and experimentally, a transducer array configuration capable of extracting features such as wells, corners, and edges in a 2-D world, by processing time-of-flights and echo peak amplitudes. The main objectives are to illustrate the methodologies and the techniques adopted to increase and validate the measuring capabilities of the system, and to outline some potential applications for the proposed sensing module. The virtual transducer formed by the three transducers is capable of providing accurate localization data, both in range and azimuth. A unique feature of the array residues in the implementation of active sensing techniques; once the azimuth estimates are available some of the transducers forming the array can be physically rotated toward the target for maximizing their sensitivities. To protect the system against spurious measurements, the measured data are finally filtered via a Kalman filter