Sensor selection based on fuzzy inference for sensor fusion

Sensor fusion is received much attention because a robot has various and numerous sensors in order to recognize an environment. By sensor fusion, the robot can get immeasurable information and more accurate information. We have proposed a sensor fusion method with sensor selection based on the reliability of sensor value in order to select sensor values. This paper proposes a novel sensor selection method with fuzzy inference. In this method, selection rules for sensor values are constructed fuzzy rules. Here, selection rules are generated by the genetic algorithm. Then, the sensor values for sensor fusion process are selected according to selection rules. Finally, selected sensor values are fused in a neural network based sensor fusion module.