Information Fusion for Hazard Analysis in Multi-Sensor Systems based on Bayesian Networks

—Modern sliding table saws are equipped with several passive safety devices. However, alone in Germany several hundred severe injuries or limb amputations occur each year due disregarding safety regulations and demounting of passive safety devices. Thus an active safety system is developed in order to protect the machine user. For this purpose, a multi sensor system for hazard analysis based on infrared and capacitive sensors is applied. When a dangerous situation is recognized by the developed sensors, the system triggers a rapid saw blade braking device. The hazard analysis is carried out by means of an information fusion of the sensor signals based on a Bayesian network. For the fusion process the measurement values of the sensors have to preprocessed appropriately. The system is implemented in C and LabView. Therefore the system runs in real-time. Validation tests have shown the very reliable function of the developed hazard analysis. Bayesian network, measurement fusion, infrare...

• Type: Conference
• Year: 2007
• Where: WCE
• Authors: Jörg Barrho, Johannes Hauger, Uwe Kiencke
• Keywords: Capacitive Sensor, Hazard Analysis, Passive Safety Devices, WCE 2007, WCE 2008