| With the continuous development of aquaculture industry in China,Now China has become the world’s largest freshwater fish farming country,but aquaculture industry is still facing problems such as backward aquaculture equipment,low aquaculture efficiency,high labor intensity and high aquaculture risk.Therefore,the aquaculture industry is in urgent need of fresh technology.Underwater robot has mature technology and application in the fields of underwater exploration and fishing operation et al.The introduction of underwater robots into the aquaculture industry can replace manual to realize the real-time monitoring of aquaculture environment,real-time monitoring of water quality,underwater fishing and other unmanned operations.Compared with the traditional propeller propelled underwater robots,the bionic robotic fish has greater advantages in the aquaculture industry: Robotic fish has a bionic shape,in the breeding environment with a certain degree of concealability,fish are not easy to be disturbed;The bionic robotic fish relies on the bionic tail structure for propulsion,which produces less noise than the propeller.Compared with propeller propulsion,fish swimming mode is more efficient,more environmentally friendly and less power consumption.Based on the needs of aquaculture,this paper has designed a biomimetic aquaculture robot fish by simulating the swimming mode of carangidae.The research contents and conclusions are as follows:According to the task requirements of the robotic fish,the mechanical structure of the robotic fish was designed.:The double joint servos were used as the propulsion unit of the robotic fish tail to provide the movement power of the robotic fish.The middle pressure chamber was designed for control unit to be placed in,and the head was designed for adjusting the center of gravity and placing all kinds of sensors for data collection.The robotic fish control requirements were analyzed,and the robotic fish control system architecture was designed: As the main control,raspberry PI was equipped with depth sensor,UWB tag,gyroscope,and other sensors to realize the movement control of robotic fish.By analyzing the characteristics of fish swimming,the motion control system of the robotic fish tail was designed.The HOPF-oscillator-based CPG units were used as the signal generator to control the robotic fish model in Simulink.And then the algorithm was encapsulated as a control function to realize the control of the tail servos on the Linux platform.The RRT path planning algorithm was improved based on the movement data obtained by experiments,and the Minimum Snap algorithm was introduced to optimize the original uneven path.Based on the yawing phenomenon and gyroscope’s characteristics,anti-yawing LOS guidance controller and fuzzy-PID heading controller were designed,and the simulation test of the algorithm was realized on Webots platform.Robotic fish prototype was built and water tightness test was carried.Underwater tests were carried out on the movement functions of the robotic fish such as direct swimming and steering.By integrating all the control algorithms,the robotic fish control software was built,and the underwater test of the robotic fish path following was carried out,which proves that the robotic fish control algorithm can be realized well on the prototype. |
