eduzhai > Applied Sciences > Engineering >

Artificial Lateral Line Based Relative State Estimation for Two Adjacent Robotic Fish

  • Save

... pages left unread,continue reading

Document pages: 11 pages

Abstract: The lateral line enables fish to efficiently sense the surroundingenvironment, thus assisting flow-related fish behaviours. Inspired by thisphenomenon, varieties of artificial lateral line systems (ALLSs) have beendeveloped and applied to underwater robots. This article focuses on using thepressure sensor arrays based on ALLS-measured hydrodynamic pressure variations(HPVs) for estimating the relative state between two adjacent robotic fish withleader-follower formation. The relative states include the relative oscillatingfrequency, amplitude, and offset of the upstream robotic fish to the downstreamrobotic fish, the relative vertical distance, the relative yaw angle, therelative pitch angle, and the relative roll angle between the two adjacentrobotic fish. Regression model between the ALLS-measured and the mentionedrelative states is investigated, and regression model-based relative stateestimation is conducted. Specifically, two criteria are proposed firstly toinvestigate not only the sensitivity of each pressure sensor to the variationsof relative state but also the insufficiency and redundancy of the pressuresensors. And thus the pressure sensors used for regression analysis aredetermined. Then four typical regression methods, including random forestalgorithm, support vector regression, back propagation neural network, andmultiple linear regression method are used for establishing regression modelsbetween the ALLS-measured HPVs and the relative states. Then regression effectsof the four methods are compared and discussed. Finally, random forest-basedmethod, which has the best regression effect, is used to estimate relative yawangle and oscillating amplitude using the ALLS-measured HPVs and exhibitsexcellent estimation performance. This work contributes to local relativeestimation for a group of underwater robots, which has always been a challenge.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...