Estimation Methods of Flexible Tip-Steerable Needles: A Comparative Study

摘要

Needle insertion was used in percutaneous procedures such as biopsies and brachytherapy. Flexible Needles has attracted many attentions for its advantages in obstacle avoidance. In order to make the flexible needle reach the target position, the position and attitude of the flexible needle must be estimated. This paper uses three different non-linear estimation methods to estimate the st1ate of the flexible needle. The first one is the most commonly used extended kalman filter (EKF), the second is unscented kalman filter (UKF), the last one is the particle filter (PF). Principle, complexity and characteristic of the three estimation methods is different. In this paper, these different kind of filters is applied in the flexible needle with a comparative analysis. The uncertainty of the bevel-up needle model is not only reflected in the noises which these filters mainly deal with, but also the uncertainty of the model parameter. In this paper, the model parameter error is also considered. Kalman filter is based on the Gaussian noises. In this paper, we consider the situation of non-Gaussian noises.