The Cruise Strategy Of The MUUV Based On Artificial Fish School Algorithm

Unmanned Underwater Vehicle—a sunday punch of the future unmanned naval battle, isbecoming a new kind of underwater operation platform. As the key technology of UUV, thecruise strategy of multiple UUV has become the current research focus. School of fish hasaroused people’s universal interest from the organization behavior inspiration. There aremuch similarity between feeding behavior, formation cruise, intelligent obstacle avoidance offish school and target searching, cooperative engagement, obstacle avoidance of MUUV. Inthis paper, we introduce fish swarm intelligence into MUUV cruise strategy. It has greatsignificance for raising UUV’s econnaissance efficiency and maintaining military strength.At first, the biological origin of AFSA was described in detail. On that basis, themathematical and behavior description are presented, and parameter values in differentcircumstances, will solve the performance and efficiency of the algorithm have a significantimpact, that is analyzed by means of experiments. By further research, we find that AFSA hassome disadvantages, for example, the balance between the convergence rate and the solutionaccuracy is not reached. In view of questions about these disadvantages, this paper presentsan improved artificial fish-swarm algorithm. On the one hand, the method of dynamicadjustment of the step and vision of artificial fish to improve the abilities of seaching theglobal and local extremum has been adopted by the algorithm.On the other hand, thealgorithm introduces a differential evolution mutation, the ability of AFSA to break awayfrom artificial fish stochastic moving without a definite purpose or heavy getting togetherround the local optimum solution is greatly improve. Improved artificial fish-swarmalgorithm is used for optimizing a large set of numerical test functions and the resultsproduced by AFSA algorithm are compared with the results obtained by genetic algorithm,particle swarm optimization algorithm, differential evolution algorithm and artificialfish-swarm algorithm. Results show that the performance of the AFSA is better than those ofother population-based algorithms with the advantage of employing fewer control parametersand it can be efficiently used for solving multimodal and multidimensional optimizationproblems.Second, we introduce self-organization behaviors of fish school into MUUV cruisestrategy. Specifically, in the following aspects, through the swarm behavior in fish school wecan conclude following rules: flock centering, aligning, deconcentration, collision avoidance.In this paper, the realization rules were given. On that basis, the introduction of self-organization behaviors of fish school into MUUV cruise strategy by observation,perception, decision; Inspire of the way fish school was feeding, target searching strategy ofMUUV based on feeding behavior of fish school was proposed; Inspire of the way fish schoolavoid obstacle, in this paper, the sensor is used to simulate perceptive system of fish school, itcan achieve MUUV obstacle avoidance strategy.Third, in VC++6.0compile environment, by means of OpenGL function to realizevisualization of MUUV dynamic cruise in MFC framework. Its main contents compriserendering of sea, seafloor and sea surface, drawing of submarine obstacles, drawing, loadingand display of the model of the unmanned underwater vehicle, dynamic cruise of MUUV,target searching,dynamic display of blast effects, and so on.