Verification and application of a path accumulation theory of social insects from an engineering perspective

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K12007
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61050:Intelligent robotics-related
• Research Institution: Chiba University
• Principal Investigator: Okawa Kazuya 千葉大学, 大学院工学研究院, 准教授 (50344966)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: LSTM / 位置推定 / IMU

Outline of Final Research Achievements

Ants called Cataglyphis, which live in the Sahara Desert, are estimated their position by integrating their trajectories of movement. However, there is no convincing theory about how to integrate the displacement and orientation. In this research, LSTM, which is one of deep learning and can handle time series information, is applied for the integration.
In order to verifiy the effectiveness of the LSTM, a mobile robot was built. The robot is equipped with an acceleration sensor to measure the amount of movement and an angular velocity sensor to measure the orientation. In addition, the position and velocity obtained from highly accurate RTK-GNSS were taken as true values, and LSTM was trained based on them. As a result, it was confirmed that the position can be estimated with higher accuracy than the conventional method calculated by mathematical numerical integration.

Free Research Field

ロボット工学

Academic Significance and Societal Importance of the Research Achievements

サハラ砂漠で生息するCataglyphisという蟻は，連続した細かい移動の軌跡を足していく経路積算によって位置推定していると考えられているが，蟻が餌を探してさまよいながらこれらの計算をしているとは考えにくい．その一方で，近年の深層学習を用いることで動物のような知的さを人工的に機械に持たせられるかもしれない．そこで，生物学者による「経路積算説」を採用するものの，謎とされている「距離」と「向き」の統合部分を深層学習で学習させる手法を提案することとした．結果としては，計測誤差の影響を受けにくい計算ができるようになり，従来手法よりも良い結果が得られた．