Research On Local Path Planning For Unmanned Underwater Vehicles Based On Sonar Image Processing Of Whale Optimization Algorithm

Unmanned underwater vehicles(UUVs)are important carrying equipment that integrates simulation technology,intelligent control,information fusion,navigation and communication and have some characteristics of wide operation safety,strong environmental adaptability and strong autonomy.The efficiency of a UUV local path planning not only marks the level of UUV intelligence and autonomy but also guarantees the safety and reliability of a UUV effort.Under the premise of comprehensively considering factors,such as energy consumption,threat level,threat area,safety,and the purpose of a UUV path planning is to avoid all underwater target obstacles,and obtain the shortest and safest path from the start point to the target point.The precondition of realizing the UUV local path planning is to obtain the precise information of underwater target obstacles quickly and accurately.In this dissertation,a whale optimization algorithm(WOA)based on bubble-net attacking characteristics of humpback whales is the research object.WO A mainly imitates encircling prey,bubble-net attacking and random searching for prey to obtain the global optimal solution.The accomplishment of underwater target obstacles is mainly through light images and sonar images.Compared with the light image,the sonar image has advantages in terms of better imaging quality,higher resolution,longer action distance,stronger anti-interference ability and penetration,which is the main way for a UUV to obtain the underwater target information.Therefore,this dissertation focused on the threshold segmentation and template matching that based on the sonar image.The purpose is to identify and determine the shape outline and position information of underwater target obstacles according to the accuracy and stability of segmentation and matching,which has important theoretical guidance and practical value.The main research contents of the dissertation are as follows:Firstly,to enhance the exploration and exploitation of WOA,Levy flight strategy and the ranking-based mutation strategy are introduced,an enhanced whale optimization algorithm(EWOA)is proposed,which is applied to solve the function optimization.Levy flight strategy can increase the population diversity,broaden the search space and enhance the global search ability.The ranking-based mutation strategy can filter the optimal individual,avoid search stagnation and enhance the local search ability.The EWOA realizes the complementary advantages between the two optimization strategies to improve the overall search performance.The experimental results show that the EWOA has certain effectiveness and feasibility to obtain the higher calculation accuracy and faster convergence speed,which has laid a solid theoretical foundation for further research of underwater sonar image threshold segmentation and template matching.Secondly,to improve the optimization ability and segmentation accuracy of threshold segmentation algorithm,the EWOA based on Kapur’s entropy is proposed to solve the underwater sonar image threshold segmentation.Kapur’s entropy has the some advantages of the small calculation amount,easy operation,fast running speed,strong stability and high segmentation accuracy.The EWOA can avoid premature convergence to obtain the optimal solution.The purpose of image segmentation is to divide the given image into several specific and unique core areas,and present the principle and process of interested targets.That is to say,according to a certain criterion function to find the optimal threshold and the maximum fitness value,and segmentation quality directly affects the accuracy of underwater target obstacles recognition.The experimental results show that the EWOA has a higher calculation accuracy,better segmentation effect and stronger robustness.Underwater sonar image threshold segmentation improves the recognition accuracy and efficiency of target obstacles and directly affects the stability and reliability of a UUV local path planning.Thirdly,to improve the search ability and matching accuracy of template matching,the EWOA based on lateral inhibition is proposed to solve the underwater sonar image template matching.The lateral inhibition enhances the gray gradient and spatial resolution of the image by preprocessing the original image and the template image,and improves the optimization accuracy and precision.The EWOA has strong robustness and search ability to avoid falling into the local optimum.The purpose of the image matching is to accurately find the position that needs to be matched in the original image based on a template image,and to maximize the similarity between the two images to evaluate the matching precision.The experimental results show that the EWOA has a faster convergence speed,higher calculation accuracy and better matching effect.Underwater sonar image template matching provides target position information for a UUV local path planning,and improves the accuracy and robustness of a UUV matching assisted navigation,which enables a UUV to correct the course to avoid underwater target obstacles.Lastly,to avoid the underwater target obstacles with threat levels,WOA is applied to solve the UUV local path planning.The acquisition of the underwater target obstacles is very important for path planning.Therefore,the threshold segmentation method of Kapur’s entropy is used to obtain the segmented images with better segmentation effect and higher segmentation accuracy,and that utilizes the accuracy of the image segmentation to identify the outline information of the underwater target obstacles.The template matching method of lateral inhibition is used to obtain the matching images with better matching effect and higher calculation accuracy,and that utilizes stability of the image matching to determine the position information of the underwater target obstacles.The experiments have verified the effectiveness and feasibility of these methods,which provides detailed information of target obstacles for a UUV underwater operation.The purpose of the UUV local path planning is to avoid all underwater target obstacles and find an optimal path with lowest threat consumption and fuel consumption.The experimental results show that the WOA has higher search efficiency,stronger exploration and exploitation to obtain the global optimal solution.The WOA has certain superiority and stability to efficiently complete the UUV local path planning.