測距型ローカライゼーション網のリアルタイム再構成アルゴリズム
An algorithm for real-time restructuring of a ranging-based localization network
UWB測距と慣性センサを用いた6自由度位置推定において、移動アンカーを動かして測位網をリアルタイムに再構成し、最小二乗誤差の共分散行列式を最小化することで位置推定精度を向上させる手法を提案・実験検証した。
著者: Saman Fahandezh-Saadi, Mark W. Mueller
分類: cs.RO
原文アブストラクト
This paper presents a method to improve the localization accuracy of robots operating in a range-based localization network. The method is favorable especially when the robots operate in harsh environments where the access to a robust and reliable localization system is limited. A state estimator is used for a six degree of freedom object using inertial sensors as well as an Ultra-wideband (UWB) range measurement sensor. The estimator is incorporated into an adaptive algorithm, improving the localization quality of an agent by using a mobile UWB ranging sensor, where the mobile anchor moves to improve localization quality. The algorithm reconstructs localization network in real-time to minimize the determinant of the covariance matrix in the sense of least square error. Finally, the proposed algorithm is experimentally validated in a network consisting of one mobile and four fixed anchors.
