基于改进GA优化的水下机器人航向鲁棒控制

基于某型号水下机器人水平面线性化模型,针对水下机器人系统存在的多种不确定因素,工作环境复杂多变的问题,采用混合灵敏度法设计水下机器人航向鲁棒控制器。在设计中,通过罚函数小生境遗传适应度值作为桥梁建立起遗传寻优与混合灵敏度参数之间的关系,构成多目标优化问题寻优加权函数。实验结果表明,所用方法可在保证系统较强鲁棒性的同时,又能兼顾时域指标和混合灵敏度加权函数频域要求,获得综合性能指标最优的鲁棒控制器,控制效果较好。

The Improved GA Optimization Based Robust Control for Course of Underwater Robot

Investigation is made on some models of the level of the underwater robot linear equation. In view of the underwater robot system which has many uncertainty factors and the work environment are both complicated and varied due to some reasons such as ocean currents,Mixed sensitivity method is employed to design robust controller of heading.Through the fitness value of niche genetic algorithm based on penalty function established the relationship between genetic optimization and mixed sensitivity parameters which can constitute the weighted function multi-objective optimization problem.The simulation results shows that this approach allows the system has strong robustness ,which can also meet the time and frequency of mixed sensitivity weighting function domain indexes at the same time,the effect of control is encouraging.