Audio sensors fusion based on vote for robot navigation

Rescue robots usually work in unknown complex environment with mess obstacles and dense smoke caused by the disaster. So, in most cases, the rescue targets are unvisualable. When there is a call for help from a survival wounded man blocked behind a wall in an earthquake, the rescue robot can not find the target via the visual, infrared, laser or ultrasonic senors. According to the diffraction property of audio signal, sound can bypass obstacles, and propagate through the smoke. An audio-based robot navigation system can guide the rescue robot run to the target according to the sound for help. However, the echo caused by the walls and obstacles may mislead the robot. So a sensor fusion method based on vote for robot navigation is developed to distinguish the real sound source from the psudo ones. And ideal effects are obtained in experiments.earthquake,