Smart Ultrasonic Vision System for Mobile Robots

For navigation of semi-autonomous robots various machine based vision systems are used. Ultrasonic systems are one of the cheapest, but some limitations reduce their performance. This paper is devoted to improvement of a data acquisition speed and noise robustness of such systems. These goals were achieved using simultaneous data acquisition along a few different directions using orthogonal complex coded series as probing signals, parallel correlation processing of the received signals by individual digital signal processors (DSP) and determination of the coordinates of the obstacles detected using the tri-aural approach. In order to increase the environment scan speed, electronically steered ultrasonic phased arrays are used with the different code sequences for each direction. The data exchange with the host computer is performed via the LON Works interface. Ultrasonic image formation and analysis algorithms