A simulation robot fish control method and device based on extreme learning machine

Abstract

An embodiment of the invention discloses a simulated robotic fish controlling method and a simulated robotic fish controlling device based on an extreme learning machine. By the aid of the extreme learning machine, the position of a simulated robotic fish is determined, a hitting point is determined according to the positions of the simulated robotic fish and a water polo, and then a corresponding action strategy is taken. The simulated robotic fish controlling method and the simulated robotic fish controlling device based on the extreme learning machine solve the technical problem that an existing point-to-point control algorithm is non-ideal in controlling effect since environmental factors around the simulated robotic fish are uncertain and the simulated robotic fish is interfered by waving of bionic water during swinging of the simulated robotic fish.

Description

technical field [0001] The invention relates to the field of robotic fish in water, in particular to a method and device for controlling a simulated robotic fish based on an extreme learning machine. Background technique [0002] With the rapid development of marine technology, the application of underwater robotic fish has become more and more extensive. As an important part of scientific research, the simulation research on underwater robotic fish has gradually become one of the research hotspots in this field. In recent years, various types of robotic fish competitions have been held at home and abroad, aiming to exchange new ideas and new progress in the field of robotic fish research through competitions. Based on this background, Peking University has joined forces with many universities and research institutes to build the URWPGSim2D simulation platform, which is a good research platform for simulating robotic fish in water. The platform takes fish as the simulation ...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides a control method and device for a simulated robotic fish based on an extreme learning machine. The position of the robotic fish is determined by the extreme learning machine and the hitting point is determined according to the positions of the robotic fish and the water polo, so that a corresponding action strategy is adopted. It solves the problem that the existing point-to-point control algorithm has an unsatisfactory control effect on the robot fish caused by the uncertainty of the environmental factors around the robot fish and the fluctuation of the bionic water during the swimming process of the robot fish. question

Images