Intelligent security robot fire detection system using adaptive sensory fusion method

Abstract

Multisensor fire detection algorithm (MSFDA) was one of the important elements in the fire detection system for intelligent system. The paper proposed an adaptive fusion method for fire detection, and uses smoke sensor, flame sensor and temperature sensor to detect a fire. In reality, the phenomenon of the fire may have smoke, flame and high temperature. But these signals can't happen at the same time. First, we use computer simulation to improve the adaptive fusion method to be adequate. Next, we design the fire detection modular by using ionization smoke sensor (TG-135), temperature semiconductor sensor (AD590) and ultra-violet sensor (R2868). We use Taylor expression to modify the adaptive fusion method that was proposed by Ansari, and simulate to compare with results of one order, second order and third order expression. The modified adaptive fusion can provide adequate reliable fusion for fire detection. Finally, the experimental results of the fire detection module can provide exact fire detection results in a variety of conditions. The fire detection module was embedded in the security robot.