3D Path Planning For Robotic Fish Based On Improved Ant Colony Algorithm

The ocean has a large number of biological resources,energy,etc.In the current situation of increasingly scarce land resources,it is of great significance to strengthen the exploration and utilization of marine resources.The development of underwater vehicle technology has provided great assistance to human exploration of the ocean.Autonomous underwater vehicle(AUV)is widely used in various underwater operations because of its excellent flexibility,strong intelligence and excellent individual endurance.In the field of underwater robot technology,dynamic modeling design and underwater path planning are the key technologies that affect the success of underwater operation.This thesis mainly studies the dynamic modeling verification,global static path planning and local dynamic obstacle avoidance of underwater robot fish.First of all,the underwater dynamic modeling of robot fish is carried out.First,the overall model is deduced by using Newton-Euler method,and then the theoretical verification is carried out.The complete dynamic model is obtained by analyzing the hydrodynamic damping force and moment,tail force and moment,and other forces and moments,and then the system parameter identification is completed.Finally,the accuracy of the dynamic model is verified by comparing with the underwater experiment of robot fish.Secondly,an improved ant colony algorithm is proposed for the robot fish underwater global static path planning problem.This algorithm designs a more efficient heuristic function,adjusts the pheromone update rules,and adds an elite ant system to adjust the pheromone carrying value of the considered elite ants separately.The simulation results show that the improved algorithm improves the convergence speed and reduces the length of the best individual path.For the local obstacle avoidance strategy of unknown dynamic obstacles,the speed obstacle method is selected to solve the problem.The speed and position information of the obstacles are calculated using the sensor data collected by the robot fish,the collision area is determined,and then the most reasonable way with the lowest energy consumption is selected to complete the obstacle avoidance.Simulation experiments have verified the effectiveness of this method in local dynamic obstacle avoidance for robotic fish.Finally,how to return to the preset global planning route after the obstacle avoidance is completed is explained.Finally,two underwater environments are designed by adjusting the layout of obstacle piles in the experimental pool.After three-dimensional modeling based on specific environmental data,the improved ant colony algorithm is used to iterate the best path,and the robot fish can reach the target point through the best path.On the basis of two kinds of pool environments,a number of dynamic obstacles with different directions and different speeds that are expected to collide with the robot fish are added.The robot fish have selected the most reasonable strategy to complete local dynamic obstacle avoidance.The above experimental results prove that the algorithm of this subject has certain engineering significance in the field of underwater robot path planning.