Flight reliability of multi rotor UAV Based on genetic algorithm
暂无分享,去创建一个
In the disturbed case, since the parameter tuning method is imperfect, there have been issues of excessive overshoot and adjusting time in MFAC. To optimize MFAC controller, a MFAC-ABC is proposed, with ITAE criterion as the fitness function, and key parameters of ηkand μ are optimized and tuned using ABC, and MFAC simulation model is established using Matlab Simulink platform, with written ABC control algorithm. Simulation results show that, compared with traditional MFAC and MFAC-PSO, MFAC-ABC control method is effective in reducing the response time and the overshoot, with good robustness, which has improved the adaptability and effectiveness of MFAC in the disturbed case. Introduction MFAC is a new method of data-driven control established based on the new non-parametric time-varying dynamic linearization, and its essence is to replace the general nonlinear system with a series of dynamic linear time-varying model nearby trajectory of controlled system by using the newly introduced concept of pseudo-partial-derivative and pseudo order, and only the input and output data of the controlled system is used to estimate pseudo-partial-derivative of system online, and on this basis, self-adaptive control is re-designed to meet the performance indicators. MFAC We assuming that there are N cities in total, M is the total number of ants, let us suppose that M ants are put into N randomly selected cities, ( ) , 1,2,3 , i j n ij d = represents the distance between city i and city j , ( ) t ij t represents the amount of remaining information on the t time between the city of i and the city of j , at the initial time, the amount of information on each path is equal, assuming that ( ) 0 c ij t = (c is the constant), when ant ( ) 1,2,3 , k k m = is in the process of movement, they will choose the next city that has not yet been visited on the basis of the amount of information on each path, at the same time, after they complete a step or a cycle, in other words, from a city to another city or complete the access to all cities, they will update the residual information on all paths. The basis for choosing the next city is mainly as follows [15]: residual amount of information was left between city of i and city of j on the time of ( ) t ij t , that is, the information was provided by ant colony algorithm, as to the aspirational information the ij η ants transfer form city of I to city of J, this heuristic information can be given by the problem which were waiting to be solved, and it can be realized by a certain algorithm, in the problem of TSP, generally we get the value 1 ij dij η = , ij η here can be called a priori knowledge, and then we can get the probability that ( ) t t ij p on the t time ants k transfer from city i to the target city j . 2nd Workshop on Advanced Research and Technology in Industry Applications (WARTIA 2016) © 2016. The authors Published by Atlantis Press 1253
[1] Peng Shi,et al. A Novel Model-Free Adaptive Control Design for Multivariable Industrial Processes , 2014, IEEE Transactions on Industrial Electronics.
[2] Zhongsheng Hou,et al. Controller-Dynamic-Linearization-Based Model Free Adaptive Control for Discrete-Time Nonlinear Systems , 2013, IEEE Transactions on Industrial Informatics.