Scholarlink

<p>This study proposes a method for estimating the position and attitude of targets based on multi-sensor nonlinear measurement systems (MSNMS). We first establish an integrated model coupling translational and rotational dynamics, alongside a measurement model incorporating distance and radial velocity. The necessary and sufficient conditions for system observability are rigorously derived. Analysis reveals that conventional filtering methods suffer from linearization errors and state-representation mismatch, and we further prove that such external measurement systems are not group affine, rendering invariant observer frameworks inapplicable. To overcome these limitations, the unscented Kalman filter on Lie groups (UKF-LG) within the framework of MSNMS is developed. An adaptive update method for the state and measurement noise covariance matrices is designed, leading to the proposed adaptive UKF-LG. This method effectively enhances estimation accuracy and robustness against complex environments and inaccurate prior noise statistics. Simulation experiments demonstrate the superior performance of the proposed method.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Motion estimation of rigid targets based on an adaptive unscented kalman filter on lie groups

  • 期刊论文

  • 发表日期：

    2026年1月20日

查看全文
  • Yu Lu,
  • Bowen Hou,
  • Zhangming He,
  • Haiyin Zhou,
  • Dayi Wang,
  • Jiongqi Wang

摘要

This study proposes a method for estimating the position and attitude of targets based on multi-sensor nonlinear measurement systems (MSNMS). We first establish an integrated model coupling translational and rotational dynamics, alongside a measurement model incorporating distance and radial velocity. The necessary and sufficient conditions for system observability are rigorously derived. Analysis reveals that conventional filtering methods suffer from linearization errors and state-representation mismatch, and we further prove that such external measurement systems are not group affine, rendering invariant observer frameworks inapplicable. To overcome these limitations, the unscented Kalman filter on Lie groups (UKF-LG) within the framework of MSNMS is developed. An adaptive update method for the state and measurement noise covariance matrices is designed, leading to the proposed adaptive UKF-LG. This method effectively enhances estimation accuracy and robustness against complex environments and inaccurate prior noise statistics. Simulation experiments demonstrate the superior performance of the proposed method.