摘要
微分扩张状态观测器作为自抗扰控制技术的核心之一,具有很好的滤波性能,在目标跟踪领域可以很好地实现对目标运动轨迹的预测。针对DESO参数多且难以整定的问题,提出了基于混沌遗传的参数整定算法。算法有效提升了遗传算法的寻优能力,提高了算法的收敛速度,在一定范围内能够求得的全局最优解,满足对多种运动模型的滤波精度要求。仿真实验结果证实了所提算法的可行性和有效性,并表明通过混沌遗传算法整定出的DESO参数具有很强的鲁棒性。
Differential extended state observer is one of the key techniques in Active Disturbance Rejection Controller techniques. It has strong filtering performance and can well predict target＇ s moving trajectory in object tracking field. In this paper, a parameter regulation algorithm of DESO based on Chaos Genetics is presented to deal with the difficulties in the parameter regulation of DESO based on chaos genetic algorithm. Compared with standard genetic al- gorithm, it shows that this method has stronger ability of looking for optimization solution and faster convergence. At the same time, the proposed method can obtain global optimization solution in a certain scope and meet the requirement of filter precision for many kinds of motion models. The experiments prove that the proposed algorithm is feasible and effective. At the same time , it shows strong robustness of DESO set by the chaos genetic algorithm.
出处
《计算机仿真》
CSCD
北大核心
2009年第7期199-203,共5页
Computer Simulation
基金
国家自然科学基金项目(60175010)
关键词
自抗扰控制
微分扩张状态观测器
混沌算法
遗传算法
Active disturbance rejection controller
Differential extended state observer （DESO）
Chaos algo- rithm
Genetic algorithm
