An Optimal Sonar Array for Target Localization and Classification Lindsay Kleeman Intelligent Robotics Research Centre Dept of Electrical & Computer Systems Eng. Monash University, Australia Roman Kuc Intelligent Sensors Laboratory Department of Electrical Engineering Yale University, USA Abstract A novel sonar array for mobile robots is presented with applications to localization and mapping of indoor environments. The ultrasonic sensor localizes and classifies multiple targets in two dimensions to ranges of up to 8 meters. By accounting for effects of temperature and humidity, the system is accurate to within 1 mm and 0.1 degrees in still air. Targets separated by 10 mm can be discriminated. Targets are classified into planes, corners, edges and unknown, with the minimum of two transmitters and two receivers. A novel approach is that receivers are closely spaced to minimize the correspondence problem of associating echoes from multiple targets. A set of templates is generated for echoes to allow the optimal arrival time to be estimated, and overlapping echoes and disturbances to be rejected. 1. Introduction Ultrasonic sensors provide a cheap and reliable means for robot localization and environmental sensing when the physical principles and limitations of their operation are well understood. A sensor design is presented that approaches the fundamental physical limitations of sonar in terms of accuracy and discrimination. The properties of air, reflectors and noise are the limiting factors. We concentrate on environments composed of specular surfaces, such as smooth walls, bookcases, tables and chairs that reflect acoustic energy analogously to a mirror reflecting light. Rough surfaces can be treated with other techniques [1]. The applications of primary interest are robot localization from sensing known environmental features, such as wall and corner positions [2, 3, 4], and conversely, mapping of unknown environments for localization and navigation [5, 6, 7, 8]. Obstacle avoidance [9, 10] is another application of the sensor. X (m) -> -0.15 -0.1 -0.05 0 0.05 0 .1 0.15 0 .2 1 1 .1 1 .2 1 .3 1 .4 1 .5 Edge reported P lane reported C o rne r re p o rte d Figure 1 - Sensor reporting a 25 mm diameter table leg as an edge, and a corner and plane. The sensor is at the origin and points along the X-axis. An emerging classification standard for two dimensional indoor target types is that of planes, corners and edges [2, 5, 11, 12, 13]. A plane is assumed to be a vertical smooth flat surface that reflects ultrasound specularly. A corner is a concave intersection of two planes at right angles, and an edge is assumed to reflect ultrasound from a point that is approximately independent of the transmitter and receiver positions. The sensor approach presented here is novel in the sense that it classifies all three target types with the one stationary sensor, simultaneously in some cases, with high accuracy and discrimination. For example in Figure 1, the sensor views a table leg in front of a corner. In one measurement, the sensor localizes and classifies the corner, the table leg (as an edge) and a plane. Our approach has higher speed and accuracy, particularly in bearing, compared to single transducer systems that rely on multiple displaced readings and wheel odometery for L.Kleeman and R.Kuc, "An optimal sonar array for target localization and classification", IEEE International Conference on Robotics and Automation, San Diego USA, May 1994 pp 3130-3135.