Distributed cooperative state estimation algorithms of nonlinear systems and their application to SLAM problems

2022 Fiscal Year Final Research Report

• Project/Area Number: 19K04447
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 21040:Control and system engineering-related
• Research Institution: Ritsumeikan University
• Principal Investigator: Takaba Kiyotsugu 立命館大学, 理工学部, 教授 (30236343)
• Co-Investigator(Kenkyū-buntansha): 大橋 あすか 香川高等専門学校, 一般教育科, 助教 (50782166)
• Project Period (FY): 2019-04-01 – 2023-03-31
• Keywords: 分散協調状態推定 / ベイズ推定 / カルマンフィルタ / SLAM

Outline of Final Research Achievements

This research is concerned with the distributed state estimation problem in sensor networks. The conventional distributed Kalman filter does not guarantee the optimality of the state estimates, and the estimation accuracy may be significantly degraded in some cases. Therefore, in the case of ring networks, we employed the Bayesian inference to derive a new distributed Kalman filter that provides the optimal estimates. We also extended this to nonlinear systems and proposed a distributed version of the unscented Kalman filter (UKF). Furthermore, we applied this distributed UKF to the multi-robot SLAM problem and verified its effectiveness and practical applicability through numerical experiments with real data.

Free Research Field

制御工学

Academic Significance and Societal Importance of the Research Achievements

センサネットワークやSLAM(移動ロボットの位置推定と環境把握の同時実行)の技術は現代社会に深く浸透しようとしている．本研究は，これらの技術の基礎となる分散協調型状態推定問題を研究したものであり，非線形システムに対する分散型最適推定アルゴリズムを提案している．このアルゴリズムの完成度を上げることにより，より多様なセンサネットワークやSLAMのシステムに対して高精度な計測・情報抽出を実現することが期待できる．