Investigation of Attitude Tracking Using an Integrated Inertial and Magnetic Navigation System for Hand-Held Surgical Instruments
Investigation of Attitude Tracking Using an Integrated Inertial and Magnetic Navigation System for Hand-Held Surgical Instruments
复制标题
DOI：
10.1109/tmech.2010.2095504
复制
发表时间：
2012-04-01
影响因子：
6.4
通讯作者：
Kazanzides, Peter
中科院分区：
文献类型：
--
作者：
Ren, Hongliang;Kazanzides, Peter
由于在微创手术中需要精确导航，已经将许多方法引入手术室用于跟踪器械的位置和取向。本文研究了利用集成惯性和磁传感器跟踪手术器械姿态（方向）的子问题。在这种情况下，通常假设传感器是准静态的，周围的磁场是稳定的。对于实际的手持式手术器械，由于预期和非预期（例如，例如，在一个实施例中，震颤）运动以及由于周围磁场的畸变。本文研究了小扰动惯性传感器坐标系下的重磁场估计问题。重力和磁场的动力学的扰动下进行了研究，他们的陀螺仪测量的关系进行了分析，卡尔曼滤波器（KFs）制定，以减少这些扰动。估计的重力和磁性值（输出的KF），随后使用在扩展的KF的姿态估计。在该滤波器中，预测模型由系统动力学给出，使用四元数制定，观测模型由估计的重力和磁场的矢量分析给出。在临床上真实的运动进行实验，以验证算法。完整的系统证明了在存在小扰动的情况下姿态估计精度的提高，并满足1度的指定精度要求。
Due to the need for accurate navigation in minimally invasive surgery, many methods have been introduced to the operating room for tracking the position and orientation of instruments. This paper considers the subproblem of using integrated inertial and magnetic sensing to track the attitude (orientation) of surgical instruments. In this scenario, it is usually assumed that the sensor is quasi-static and the surrounding magnetic field is steady. For practical hand-held surgical instruments, perturbations exist due to intended and unintended (e. g., tremor) motion and due to distortion of the surrounding magnetic field. We consider the problem of estimating the gravity and magnetic field in the inertial sensor frame with small perturbations. The dynamics of the gravity and magnetic field is studied under perturbations, their relationships to gyroscope measurements are analyzed, and Kalman filters (KFs) are formulated to reduce these perturbations. The estimated gravity and magnetic values (outputs of the KFs) are subsequently used in an extended KF for attitude estimation. In this filter, the prediction model is given by the system dynamics, formulated using quaternions, and the observation model is given by vector analysis of the estimated gravity and magnetic field. Experiments are performed to validate the algorithms under clinically realistic motions. The complete system demonstrates an improvement in the accuracy of the attitude estimate in the presence of small perturbations, and satisfies the specified accuracy requirement of 1 degrees.