Development of Mobile Robot with Sensor Fusion Fire Detection Unit

In this study, a mobile robot platform equipped with early fire detection unit was designed, manufactured and tested. The robot was especially developed to track virtually prescribed paths with obstacle avoidance function through obstacle avoidance and motion planning units and to scan the environment in order to detect the fire source via fire detection unit. Smoke, flame and temperature measurement sensors were used in the system to detect the fire event. A sensor based fusion system was also developed and proposed to get more reliable and highly accurate results from the fire detection unit. Partial Least Square (PLS) was used to derive the fire detection equation from the temperature, smoke and flame measurement values based on calibration and validation data sets with 200 samples. SAS/STAT (Statistical Analysis System) software was used for the development of the calibration and validation data sets. Five different distances at 100, 500, 600, 800 and 1000 mm were selected to test the fire detection capability of the system. Test results showed that the developed robot can detect the fire source up to 1000 mm distance. It can also be understood from the test results that the proposed sensor fusion system provides more reliable detection than one sensor based system with the accuracy value of 92%. (C) 2018, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.