A novel artificial lateral line sensing system of robotic fish based on BP neural network

Abstract

Lateral line is the typical sensory organ for aquatic vertebrates. And many scientists have done a lot of studies on combining artificial lateral line with robotic fish. However, the research in this filed is relatively immature and many involved approaches are backward. In this paper, we set up a system of bionic robotic fish virtual lateral line based on several pressure sensors and put up with a novel design for it to classify different types of flow field using BP neural network. To simplify the experiment, the system is set up in Computational Fluid Dynamics (CFD) softwares to obtain experiment data which is used for following research. After extracting some features from the raw data, we built up a flow field classifier. And it proves that the accuracy of the flow recognition is in a good performance. This paper proposes a new idea on the design of lateral line system and it's useful for the future works on it.