Research On The Technology Of Sea Cucumber Fishing Robot

In recent years,with the improvement of people's living standards,the large demand for sea cucumber promoted the rapid development of sea cucumber farming industry,so that it has become a major fisherman in northern China to achieve fishery efficiency,fishermen income an important industry.However,the survey found that the current way of sea cucumber fishing is still the traditional way of artificial fishing,there are several semi-automatic sea cucumber fishing devices,there are complex operation,high cost,in the current,ocean currents and waves in the marine environment is difficult Maintain stance stability,fishing efficiency is too low and so on.In this context,this paper is devoted to the study of a set of automated sea cucumber fishing device,through the coordination of various parts of the completion of underwater sea cucumber identification,positioning and fishing operations.Firstly,the design scheme of the two-stage structure of sea cucumber fishing robot is given.The sea cucumber fishing device and the sea cucumber collection device are designed and realized separately.Through the connection between the straw and the umbilical cord,the analysis of the sea cucumber Has the superiority.In addition,the structure of the main devices such as the underwater positioning device,fishing head and sea cucumber collection box is designed.The selection of marine diesel engine,water pump,camera and lighting equipment is made to ensure the rationality and reliability of the system structure.Secondly,the recognition algorithm of underwater sea cucumber target is studied.It consists of four parts: underwater image preprocessing,image segmentation,shape feature extraction and target sea cucumber identification based on BP network.The median filtering algorithm in the image preprocessing may give the gray value of the noise point directly to the current pixel value,resulting in a certain degree of distortion of the image.The edge detection algorithm in the image segmentation may lead to the shortcoming of the pseudo edge The Hu's invariant moment group in the shape feature extraction can keep the translation constant under the discrete condition,but it can not keep the proportion of the same.The BP algorithm in the sea cucumber target recognition is slow and easy to fall into the local minimum The corresponding improved algorithm is used to verify the effectiveness of the improved algorithm.Then,the underwater target sea cucumber positioning method based on binocular vision is studied,including the calibration of binocular stereoscopic vision system based on thecamera imaging model and calibration algorithm.According to the extracted target sea cucumber,the three-dimensional coordinates of the center of the target sea cucumber are restored by the three-dimensional reconstruction principle.The three-dimensional coordinates of the center of the target sea cucumber are restored by the three-dimensional reconstruction principle.Finally,based on the dynamic positioning control of the underwater positioning device of sea cucumber fishing robot,the space motion model of the underwater positioning device is established,and the motion equation is simplified according to the dynamic positioning requirement.According to the motion model,the fuzzy controller with dynamic positioning control is designed,and the basic fuzzy controller is optimized and improved by combining the variable universe theory.At the same time,the successive distribution rules of the propeller thrust are given.The simulation results verify that the fuzzy The effectiveness of the controller and the thrust distribution algorithm.