A collaborative collision avoidance planning method for multi-usv groups based on genetic algorithm

Abstract

A multi-USV group collaborative collision avoidance planning method based on genetic algorithm belongs to the technical field of USV control. The present invention firstly uses the floating-point coding method to initialize the speed adjustment and heading adjustment of the USV and set other control parameters; and then constructs an evaluation function, and calculates the evaluation function value of each generation of individuals of the population, so as to conduct rounds of individual populations. The genetic operations of betting selection, discrete crossover and Gaussian mutation are used to establish an iterative process to obtain the optimal solution; finally, QT software is used to build a simulation software platform for USV collision avoidance planning, adding radar detection module and genetic algorithm, and designing a typical simulation case verification algorithm effectiveness. The invention solves the problems of poor timeliness of the genetic algorithm, falling into local optimum, premature convergence, and the optimal child generation being inferior to the parent generation optimum, as well as the problem of bad navigation such as large-angle steering and wide-scale acceleration and deceleration in the navigation process.

Description

technical field [0001] The invention belongs to the technical field of USV control, and in particular relates to a multi-USV group collaborative collision avoidance planning method based on a genetic algorithm. Background technique [0002] With the development of modern science and technology, human beings develop and utilize marine resources in a variety of ways. In order to adapt to operations in offshore shallow waters and carry out non-combat naval activities such as collecting meteorological and oceanographic data, the USV (unmanned aircraft) is designed surface craft) stand out. USVs are often used to perform various special tasks such as military activities, maritime supervision, and maritime safety cruises that are not suitable for manned platforms in severe and complex marine environments. However, for some complex, dynamic and difficult tasks , a single USV is incompetent. In this case, multi-USV is gradually being valued for its overall advantages. The realiza...

AI Technical Summary

Problems solved by technology

[0003] Defects of the existing technology: Genetic algorithm, as an intelligent global search algorithm, is applied to USV objects for collision avoidance planning, and there are problems such as falling into local optimum, premature convergence, and the optimum of the offspring is inferior to the optimum of the parent.

On the other hand, the number of algorithm iterations and the population capacity determine the duration of each optimization, which generally has the defect of poor effectiveness. In literature [2], the fitness function of the genetic algorithm is embedded with the simulated annealing algorithm, which is difficult to solve and will definitely increase significantly. At the time of each iteration, the inter-population immigration operator used in the multi-population genetic algorithm in [3] has little effect, and does not consider the requirements of smoothness and stability in actual USV navigation to set limits on steering and speed

The heuristic binary initialization coding process adopted in literature [1] is complicated and redundant, adding, deleting, and repairing operations have little effect, and it is difficult to model and measure economy, smoothness, and security as evaluation factors, and the simple simulation case designed cannot prove Algorithm effectiveness

Images