摘要
针对粒子群优化算法(PSO)和遗传算法(GA)在加速度计标定优化后期出现早熟、陷入局部最优,以及在设计与应用过程中存在的缺陷,将自适应协方差矩阵进化策略(CMAES)算法应用于加速度计的快速标定:采用具有不同函数特征的Sphere、Rastrigin和Rosen三个基准函数对比测试CMAES算法的总体性能;以模观测标定方法为基础建立加速度计标定模型,选取加速度计的24个位置进行仿真观测。实验结果表明:CMAES算法在收敛速度、收敛精度、全局搜索等方面性能优异,将加速度计的标定精度提升了12个数量级,为其它算法标定加速度计奠定了良好基础,对位移系统测量研究具有重要意义。
In view of premature insufficiency,being trapped in a local optimum of accelerometer calibration optimization and defects in application by particle swarm optimization（ PSO） and genetic algorithm（ GA） algorithm,the paper proposes the research applying Covariance Matrix Adaptation Evolution Strategy（ CMAES） into accelerometer calibration. Firstly,Sphere,Rastrigin and Rosen test functions are used to testing the overall performance of CMAES algorithm,which has different functional characteristics.Then,accelerometer model is established based on observation model calibration method. Lastly,24 positions of accelerometers are selected for simulation observation. Final calibration simulation results show that the proposed algorithm has an advantage on convergence speed and accuracy. Besides,the calibration accuracy of the proposed method is increased up to 12 order of magnitudes.This new method lays a good foundation for other algorithms to calibrate accelerometers and also is of great significance to displacement system measurement research.
作者
仲志丹
张玮琪
杜慧颖
ZHONG Zhidan;ZHANG Weiqi;DU Huiying(Mechatronics Engineering College, Henan University of Science ＆ Technology, Luoyang Henan 471000, China;Luoyang Campus, Information Engineering University, Luoyang Henan 471003, China)
出处
《智能计算机与应用》
2018年第2期30-34,共5页
Intelligent Computer and Applications
基金
国家自然科学基金(50906022)
河南省高等学校重点科研项目(15A460023)
关键词
自适应协方差矩阵进化策略
加速度计标定优化
标定
多位置
遗传算法
Covariance Matrix Adaptation Evolution Strategy
calibration optimization of accelerometer
calibration
multi-position
genetic algorithm
