Diagnostic robot for vibro-acoustic measurements in accelerated life tests

The present paper outlines the application scenario which motivates the development of an autonomous mobile robot for mechanical diagnostics based on vibro-acoustic testing of electromechanical systems undergoing Accelerated Life Testing (ALT). The application for ALT of Washing Machines (WM) is demonstrated. Usually in ALTs laboratories hundreds of samples are tested in parallel, and the instrumentation of a huge number of stations with sophisticated noise and vibration test equipments would be too expensive. A mobile robot platform equipped with sensors for noise and vibration diagnostics (14 microphones array, Scanning Laser Doppler Vibrometer (SLDV), vision system and robot arm) opens the possibility to monitor in sequence all appliances under test. Data from microphones have been processed with a beam forming algorithm in order to increase the signal-to-noise ratio and to minimize the effects on the measurement of the appliances close the one under test, while an algorithm based on images pattern matching has been developed to enhance the repeatability of laser spot positioning on the products. In this paper, the main issues and problems related to the implementation of a vibro-acoustic analysis system on a mobile robot are discussed and its performances in terms of repeatability of measures, defects recognition through signal features extraction (e.g. faulty motors, bearings, etc…), are presented and discussed with reference to house-hold appliance testing. Tests have been carried out both in laboratory, with few WMs in controlled operating conditions and in a real ALT area in a washing machine factory.