Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: MEMS-IMU

Sortuj według:

Ogranicz wyniki do:

Neural inertial navigation method for wheeled robots based on self-supervised learning

Huang Fengrong, Gao Mengqi, Liu Qinglin, Gao Min

Metrology and Measurement Systems

2025

Vol. 32, nr 2

1--19

Low-cost Micro-Electro-Mechanical System Inertial Measurement Units (MEMS-IMUs) are plagued by large, complex, and variable errors. Traditional strap-down inertial navigation systems that utilize MEMS-IMUs are unable to meet the positioning requirements of wheeled robots. Although inertial navigation based on deep learning has been explored, it necessitates a substantial amount of carefully selected and labelled data, resulting in high costs. Consequently, this paper proposes a self-supervised neural inertial navigation method for wheeled robots that solely depends on MEMS-IMU data. Firstly, a representation learning model is established to extract general IMU features for self-supervised denoising. Subsequently, an intelligent framework employing contrastive learning is adopted to explore the latent information of the IMU and acquire the motion state of the robot. Specific motion state information is regarded as observations, and an invariant extended Kalman filter (IEKF) is applied for information fusion to enhance positioning accuracy. Experiments conducted on public datasets demonstrate that, in the absence of additional ground truth values, the Absolute Trajectory Error (ATE) and Temporal Relative Trajectory Error (T-RTE) of the proposed method are 20.23% and 30.71% lower than those of supervised learning-based methods, respectively. The proposed method offers a more cost-effective and practical solution for the development of inertial navigation technology for wheeled robots.

Monitoring and evaluation of rowing performance using mobile mapping data

Mpimis A., Gikas V.

Archiwum Fotogrametrii, Kartografii i Teledetekcji

2011

Vol. 22

337--349

Traditionally, the term mobile mapping refers to a means of collecting geospatial data using mapping sensors that are mounted on a mobile platform. Historically, this process was mainly driven by the need for highway infrastructure mapping and transportation corridor inventories. However, the recent advances in mapping sensor and telecommunication technologies create the opportunity that, completely new, emergent application areas of mobile mapping to evolve rapidly. This article examines the potential of mobile mapping technology (MMT) in sports science and in particular in competitive rowing. Notably, in this study the concept definition of mobile mapping somehow differs from the traditional one in a way that, the end result is not relevant to the geospatial information acquired as the moving platform travels in space. In contrast, the interest is placed on the moving platform (rowing boat) itself and on the various subsystems which are also in continuous motion. As an initial step of an on-going research this article discusses the biomechanics of rowing in relation to applied technique and equipment. Also, it reviews the current practices and sensor systems used for monitoring and evaluating performance in rowing. Finally, it presents an integrated data acquisition and processing scheme for rowing based on modern MMT. To this effect, a critical assessment of the various types of sensors as well as their installation and integration into a recording system is detailed. This analysis benefits by a number of preliminary tests using real data recordings. The boat kinematics (velocity, acceleration and attitude), techniques for their noise elimination and processes for computing average stroke characteristics are studied thoroughly. Also, the pattern of motion in rowing (stroke cycle) is examined in relation to athlete technique and capacity.