仿箱鲀机器鱼人工侧线系统的仿生设计与实验研究

Fish possess unique and important sensory organs named as the lateral line system. The lateral line systems provide fish with flow-related information of the environment. Learning from fish sensing, developing an artificial lateral line system will greatly improve the environment perception ability of underwater robots. In this proposal, based on the biological research on the lateral line system of fish and the fluid dynamics, an artificial lateral line system that is constituted of pressure sensors and possesses biological distribution, will be developed and integrated into the boxfish-inspired robot fish. By using suitable filtering algorithms, every measurement position on the surface of the robot will be accurately evaluated by the pressure sensors. Based on the theory of fluid dynamics and the robot kinematics, the relation between the artificial lateral line system and the swimming speed of the robot will be derived. By the use of artificial lateral line system, the swimming mode recognition algorithm will be established for the robot in a three-dimensional underwater environment. The characteristics of the Reverse Karman Vortex Street generated by the robotic fish will be studied. By applying the vortex dynamics and data fusion algorithms, the underlying relation between the vortex characteristics acquired by the artificial lateral line and the states of the neighboring robot fish will be analyzed. Finally, a perceptual model for the state estimation of the neighboring robot fish will be established by using the artificial lateral line system, providing an effective information interaction method for multiple robot fish collaboration control. This proposal will extend the research field of mechanical bionics. It has important significance to improve the sensing ability of future underwater robots.

侧线系统是鱼类重要感知器官，可感知周围水流动力学特征，获取环境信息。设计和实现人工侧线系统，将极大提高机器人的水环境感知能力。本项目拟研制具有仿生分布和水流压力感知能力的人工侧线系统并集成到仿生机器鱼上；设计相应的滤波算法，准确估计各个压力测量点的数据，提高测量精度；结合流体动力学理论，建立基于人工侧线系统的机器鱼游速评估算法；设计适当模式识别算法，建立三维环境下基于人工侧线系统的机器鱼游动模态识别算法；研究侧线系统对邻近机器鱼产生的反卡门涡街的特征识别，应用流体涡动力学理论和数据融合算法等方法综合地分析涡街特征与邻近机器鱼的相对状态信息之间的关系，建立基于人工侧线系统的邻近机器鱼状态的感知模型，为多机器鱼协作控制研究提供有效的信息交互方法。本项目拓展了机械仿生的研究领域，对提高水下机器人的感知能力有重要意义。

侧线系统是鱼类重要感知器官，可感知周围水流特征来获取环境信息。设计和实现人工侧线系统，将极大提高水下机器人的环境感知能力。本项目设计了一套具有仿生分布和流感知能力的人工侧线系统，并集成到仿箱鲀机器鱼上；结合流体力学理论，建立基于人工侧线系统的机器鱼游速评估算法；基于势流理论建立人工侧线系统感知模型，结合数据驱动的方法估计模型参数，建立二维和三维情况下基于人工侧线系统的机器鱼轨迹评估算法；研究人工侧线系统对邻近机器鱼产生的尾涡的特征提取算法，探究基于侧线系统的邻近机器鱼多状态信息的感知规律，为多机器鱼协作控制研究提供有效的信息交互方法。本项目成果有助于拓展仿生机械的研究领域，对提高水下机器人的感知能力提供新思路和新方法。