Data Calibration Algorithm for Artificial Lateral Line Sensor of Robotic Fish on Improved LSTM

All the time, the measurement of swimming speed has long been a major research challenge in the field of robotic fish. The artificial lateral line system can accurately measure the speed without modifying the structure of the robot fish too much. In this paper, we developed a cantilever artificial lateral line(ALL) sensor based on piezoresistive effect by imitating the “SNs” cell structure of real fish, and established an accurate physical model for it considering the complex underwater boundary layer effect, which is conducive to the selection of hardware parameters such as sensor size and stiffness. In addition, a flowmeter-based speed calibration experimental platform was developed, based on which we proposed an “end to end” speed calibration algorithm based LSTM without filtering. Then a potential mapping relationship between the electrical signal of the sensor and the speed value of the flowmeter was obtained, which laid a strong foundation for the subsequent measurement of free-swimming speed of robotic fish by artificial lateral line system composed of this sensor.

Recommended citation: Y. Ou, Z. Zhang, C. Zhou and B. Zhou, "Data Calibration Algorithm for Artificial Lateral Line Sensor of Robotic Fish on Improved LSTM," 2021 40th Chinese Control Conference (CCC), Shanghai, China, 2021, pp. 4308-4314, doi: 10.23919/CCC52363.2021.9549820. http://ouyaming.github.io/files/2021-07-26-CCC.pdf