機器人系統之聲場感知與聲音通訊介面研究(I)

Abstract

The research on sound was not emphasized in the area of Robotics when the applications of robots are mainly on industrial automation. Most researches of the environmental sensing for robots were based on vision, ultrasound, temperature or touch. However, sound (or speech) is a very important communication mean for human being. That is to say, if we wish to make robots that can enter household or become a personal assistant, the sound interface has to be a necessary part of the system. The communication of sound can be separated into two categories. The first one is the content of the sound such as speech or music melody. The second one is the information created by sound wave interaction with the environment. For example, the interaction can reveal the sound source location or even the acoustic field that is related to the space where the sound is generated. This is a 3-year proposal to study the second category mentioned above. The topics include making a microphone array module to facilitate the research experiment and integration on robots. Secondly, due to the movement of robots, the microphone array on the robot has the property of infinite sampling in space. We will study the advantage it brings as well as the shortcomings. Meanwhile, we will also study the classical sound source separation and speech enhancement problem as they are very important for the hearing ability of a robot. This proposal is part of an integrated effort to make personal robots. And this work will play the role of hearing systems for the robot as we will collaborate closely with other research groups participated in this project.