Zero Velocity Detection Without Motion Pre-Classification: Uniform AI Model for All Pedestrian Motions (UMAM)

Kone, Yacouba; Zhu, Ni; Renaudin, Valérie

doi:10.1109/jsen.2021.3099860

Cited by 10 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to factors such as noise and external disturbances, the actual velocity has some error, and it is not exactly zero during the standstill period. The velocity output during this period becomes the observation for the filter, as shown in Equation (10). This allows estimation of horizontal attitude errors, position errors, etc., which are then fed back into the navigation calculation system to obtain corrected navigation parameters.…”

Section: Basic Zero-velocity Correction Pedestrian Dead Reckoning Alg...mentioning

confidence: 99%

“…To enhance the accuracy of zero-velocity detection in mixed-motion modes, Yujie Sun [7], Mingkun Yang [8], and Seong Yun Cho [9] have proposed innovative zero-velocity interval (ZVI) detectors. To accommodate various motion patterns, Ni Zhu [10] proposed a machine learning model for detecting zero-velocity moments without any pre-classification step, named the Uniform AI Model for All pedestrian Motions (UMAM). Pedestrian gait detection helps improve the accuracy of foot-mounted pedestrian autonomous navigation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Improved Method for Indoor 3D Pedestrian Positioning Based on Dual Foot-Mounted IMU System

Jia,

Yu,

et al. 2023

Micromachines

View full text Add to dashboard Cite

Micro-Electro-Mechanical System (MEMS) inertial sensors, characterized by their small size, low cost, and low power consumption, are commonly used in foot-mounted wearable pedestrian autonomous positioning systems. However, they also have drawbacks such as heading drift and poor repeatability. To address these issues, this paper proposes an improved pedestrian autonomous 3D positioning algorithm based on dual-foot motion characteristic constraints. Two sets of small-sized Inertial Measurement Units (IMU) are worn on the left and right feet of pedestrians to form an autonomous positioning system, each integrated with low-cost, low-power micro-inertial sensor chips. On the one hand, an improved adaptive zero-velocity detection algorithm is employed to enhance discrimination accuracy under different step-speed conditions. On the other hand, considering the dual-foot gait characteristics and the height difference feature during stair ascent and descent, horizontal position update algorithms based on dual-foot motion trajectory constraints and height update algorithms based on dual-foot height differences are, respectively, designed. These algorithms aim to re-correct the pedestrian position information updated at zero velocity in both horizontal and vertical directions. The experimental results indicate that in a laboratory environment, the 3D positioning error is reduced by 93.9% compared to unconstrained conditions. Simultaneously, the proposed approach enhances the accuracy, continuity, and repeatability of the foot-mounted IMU positioning system without the need for additional power consumption.

show abstract

Section: Basic Zero-velocity Correction Pedestrian Dead Reckoning Alg...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The Improved Method for Indoor 3D Pedestrian Positioning Based on Dual Foot-Mounted IMU System

Jia,

Yu,

et al. 2023

Micromachines

View full text Add to dashboard Cite

show abstract

“…Since SELDA only estimates stride length, for illustration's purpose, we pile up three modules namely stride detection, SELDA, and heading estimation to build a positioning system. Stride instants and walking directions are provided by our footmounted reference tracker ( [34] and [35]).…”

Section: A Selda: Pedestrian Stride-length Estimation Based On Lstm A...mentioning

confidence: 99%

Analysis of the Recent AI for Pedestrian Navigation With Wearable Inertial Sensors

Renaudin

Kone

et al. 2023

J. Ind. Sea. Pos. Nav.

View full text Add to dashboard Cite

Wearable devices embedding inertial sensors enable autonomous, seamless, and low-cost pedestrian navigation. Appealing it is, the approach faces several challenges: measurement noises, different device-carrying modes, different user dynamics, and individual walking characteristics. Recent research applies Artificial Intelligence (AI) to improve inertial navigation's robustness and accuracy. Our analysis identifies 2 main categories of AI approaches depending on the inertial signals segmentation: either using human gait events (steps or strides) or fixed-length inertial data segments. A theoretical analysis of the fundamental assumptions is carried out for each category. Two state-of-the-art AI algorithms (SELDA, RoNIN), representative of each category, and a gait-driven non-AI method (SmartWalk) are evaluated in a 2,17 km long open access dataset, representative of the diversity of pedestrians' mobility surroundings (open-sky, indoors, forest, urban, parking lot). SELDA is an AI-based stride length estimation algorithm, RoNIN is an AI-based positioning method, and SmartWalk is a gait-driven non-AI positioning method. The experimental assessment shows the distinct features in each category and their limits with respect to the underlying hypotheses. On average, SELDA, RoNIN, and SmartWalk achieve 8 m, 22 m, and 17 m average positioning errors (RMSE) respectively, on six testing tracks recorded with two volunteers in various environments.

show abstract

“…Stance phase detection: To ensure a good aiding by ZUPT, a clear detection of stance phases of natural walk is an obvious topic for research. Besides the classical techniques like those of Skog et al [13], the available methods for this task occupy meanwhile a wide space in the literature of pedestrian navigation and range from rather general approaches like artificial intelligence [15], machine learning [16], and neural networks [17] to more targeted techniques like motion classification [18], detection of special environments like escalators [19], decomposition of the single steps [20], or usage of insole sensors [21].…”

Section: Research Topics On Zupt-based Pedestrian Navigationmentioning

confidence: 99%

Reevaluation of Algorithmic Basics for ZUPT-Based Pedestrian Navigation

2022

View full text Add to dashboard Cite

During the last 10 to 15 years, pedestrian navigation based on zero velocity updates (ZUPT) has become very popular. One of the main reasons for this is the increasing availability of small, low-cost inertial measurement units. However, the processing of the data from these units for pedestrian navigation is almost exclusively based on algorithmic features that originate from classical inertial navigation with highgrade sensors. In addition, the historical background of the ZUPT approach presupposes also sensors with high accuracy. This leads to the problem that neither the ZUPT approach nor the algorithmic features mentioned are consistent with the accuracy level of the inertial measurement units used normally for pedestrian navigation. Therefore, the question arises of whether the usual algorithmic basics and numerical procedures employed by ZUPT-based pedestrian navigation are adequate. Supported by a literature review showing the state of the art, this study investigates the effect of basics such as the system states and the usage of Runge-Kutta method on the navigation accuracy from pedestrian inertial data aided by ZUPT. To this end, a comparative processing of a well-known, published dataset of a short walk and of own data from the authors using different data fusion algorithms was employed. The important results of this study are that the oftenused omission of inertial sensor biases cannot be recommended, that a continuous-discrete Kalman filter in combination with a Runge-Kutta algorithm performs better than traditional filter configurations, and that total navigation states are an interesting alternative to classical navigation error-states.

show abstract

Zero Velocity Detection Without Motion Pre-Classification: Uniform AI Model for All Pedestrian Motions (UMAM)

Cited by 10 publications

References 20 publications

The Improved Method for Indoor 3D Pedestrian Positioning Based on Dual Foot-Mounted IMU System

The Improved Method for Indoor 3D Pedestrian Positioning Based on Dual Foot-Mounted IMU System

Analysis of the Recent AI for Pedestrian Navigation With Wearable Inertial Sensors

Reevaluation of Algorithmic Basics for ZUPT-Based Pedestrian Navigation

Contact Info

Product

Resources

About