论文部分内容阅读
针对常规UKF在组合测姿中自适应性不足的问题,提出一种基于遗传模糊推理的自适应UKF组合测姿滤波算法。首先建立了基于模糊推理的自适应UKF,利用模糊推理系统对组合测姿系统的量测噪声统计特性进行调整,以实现状态的准确估计。然后利用遗传算法对模糊推理系统的隶属度函数参数进行了离线优化,以提高系统精度。最后以陀螺仪、加速度计和磁强计组成的组合测姿系统进行了实验。实验结果表明,基于遗传模糊推理的自适应UKF在量测噪声变化时仍能保持较好的测量精度,具有较强的自适应能力。
Aiming at the problem of the conventional UKF’s lack of adaptability in the combined attitude measurement, an adaptive UKF combined attitude detection and filtering algorithm based on genetic fuzzy reasoning is proposed. First of all, an adaptive UKF based on fuzzy inference is established. The fuzzy inference system is used to adjust the statistical characteristics of the measured noise of the combined attitude-measuring system to achieve an accurate estimation of the state. Then the genetic algorithm is used to optimize the membership function of fuzzy inference system to improve the system accuracy. Finally, a combination of gyroscope, accelerometer and magnetometer composed of attitude measurement system experiments. Experimental results show that the adaptive UKF based on genetic fuzzy reasoning can maintain good measurement accuracy and has strong self-adaptability when the measurement noise changes.