| 付雷,章政,余义.基于动态权值共轭梯度的自适应互补滤波姿态估计算法[J].高技术通讯(中文),2019,29(10):977~984 |
| 基于动态权值共轭梯度的自适应互补滤波姿态估计算法 |
| Attitude estimation based on the fusion algorithm of dynamic weight conjugate gradient and adaptive complementary filter |
| |
| DOI：10.3772/j.issn.1002-0470.2019.10.005 |
| 中文关键词: 姿态估计； 共轭梯度法； 互补滤波； 四旋翼飞行器； 微机电系统（MEMS） |
| 英文关键词: attitude estimation, conjugate gradient with a dynamic weight, complementary filter, quadrotor, micro-electro-mechanical system (MEMS) |
| 基金项目: |
|
| 摘要点击次数: 2748 |
| 全文下载次数: 1755 |
| 中文摘要: |
| 针对低成本微机电系统（MEMS）惯性测量器件易发散，单一的姿态估计算法存在精度低、抗干扰性弱等问题，提出了一种动态权值共轭梯度法与自适应互补滤波融合的姿态估计算法。该算法在共轭梯度算法中加入动态权值，根据载体运动加速度大小对权值进行动态调整，消除运动加速度对姿态估计的不利影响；将加速度计的输出采用动态权值共轭梯度法估计出姿态四元数，并将其与陀螺仪的输出通过自适应互补滤波算法融合以减小惯性测量单元(IMU)的漂移和噪声干扰，提高微型四旋翼飞行器姿态估计的跟踪精度。为了验证所设计算法的可行性和有效性，搭建了基于STM32单片机的四旋翼飞行器实验平台，实验结果表明该算法提高了姿态估计的跟踪精度以及非重力运动加速度干扰下的抗干扰能力。 |
| 英文摘要: |
| Based on the conjugate gradient method with a dynamic weight and the adaptive complementary filter for attitude estimation, a fusion algorithm is proposed to solve such problems as divergence of low-cost MEMS inertial sensors, low precision and low anti-jamming ability of the single attitude estimation algorithm. The weight of the conjugate gradient method is adjusted by the carrier acceleration to remove the adverse effect of the motion acceleration. In order to reduce the interference of the drift and noise from the inertial measurement unit (IMU), the attitude quaternion which is calculated by the output of the accelerometer with dynamic weight conjugate gradient algorithm is fused with the output of gyroscope by the adaptive complementary filter. Finally, experimental results on the quadrotor with STM32 demonstrate that the proposed algorithm improves the tracking accuracy of attitude estimation and the anti-jamming ability of quadrotor with the disturbance of non-gravity motion acceleration. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
