記憶付きアンセンテッド粒子フィルタによる6自由度触覚位置推定
Memory Unscented Particle Filter for 6-DOF Tactile Localization
触覚の接触点情報のみから物体の6自由度姿勢をリアルタイム推定するMUPFを提案し、過去測定の記憶とUKFで多峰性分布に対応、iCubで実証した。
著者: Giulia Vezzani, Ugo Pattacini, Giorgio Battistelli, Luigi Chisci, Lorenzo Natale
分類: cs.RO
原文アブストラクト
This paper addresses 6-DOF (degree-of-freedom) tactile localization, i.e. the pose estimation of tridimensional objects given tactile measurements. This estimation problem is fundamental for the operation of autonomous robots that are often required to manipulate and grasp objects whose pose is a-priori unknown. The nature of tactile measurements, the strict time requirements for real-time operation and the multimodality of the involved probability distributions pose remarkable challenges and call for advanced nonlinear filtering techniques. Following a Bayesian approach, this paper proposes a novel and effective algorithm, named Memory Unscented Particle Filter (MUPF), which solves the 6-DOF localization problem recursively in real-time by only exploiting contact point measurements. MUPF combines a modified particle filter that incorporates a sliding memory of past measurements to better handle multimodal distributions, along with the unscented Kalman filter that moves the particles towards regions of the search space that are more likely with the measurements. The performance of the proposed MUPF algorithm has been assessed both in simulation and on a real robotic system equipped with tactile sensors (i.e., the iCub humanoid robot). The experiments show that the algorithm provides accurate and reliable localization even with a low number of particles and, hence, is compatible with real-time requirements.
