CyclePro: A Robust Framework for Domain-Agnostic Gait Cycle Detection

Ma, Youchun; Ashari, Zhila Esna; Pedram, Mahdi; Amini, Navid; Tarquinio, Daniel; Nouri‐Mahdavi, Kouros; Pourhomayoun, Mohammad; Catena, Robert D.; Ghasemzadeh, Hassan

doi:10.1109/jsen.2019.2893225

Cited by 19 publications

(3 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the past decades, with the development of micro-electro-mechanical (MEMS) technologies, MEMS inertial sensors have been extensively used for pedestrian 2 navigation [1,7,14,15,16].These sensors consisting of a tri-axial gyroscope and a tri-axial accelerometer are usually attached to a certain part of one's body for motion monitoring or biomedical rehabilitation purposes [17,19,18,20,21,22].…”

Section: Related Workmentioning

confidence: 99%

Online calibrated, energy-aware and heading corrected pedestrian navigation with foot-mounted MARG sensors

Zhou

Zhang

et al. 2023

Measurement

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Online calibrated, energy-aware and heading corrected pedestrian navigation with foot-mounted MARG sensors

Zhou

Zhang

et al. 2023

Measurement

View full text Add to dashboard Cite

“…The recognition of cycles or patterns with different lengths are still a problem [22], thus most existing works finding a cycle require that the approximate length of the time series is known in advance [23]. For example, gait recognition applications use a pre-defined reference signal to recognize a step, [24], [25]. However, the increasing data space capacity makes it possible to analyze patterns with different length [18]- [20].…”

Section: Cycle Recognitionmentioning

confidence: 99%

Load Profile Cycle Recognition for Industrial DC Microgrids with Energy Storage Systems

Mannel

Müller

Knochelmann

et al. 2020

2020 IEEE 29th International Symposium on Industrial Electronics (ISIE)

View full text Add to dashboard Cite

Many consumers in production plants like industrial robots or tool machines perform repetitive movements, which lead to a cyclic load demand. However, these load profiles can usually only be roughly estimated at the planning stage. Hence, a subsequent online adaptation of the energy distribution is useful for cases, such as balancing between the charging and discharging amount of energy storage systems to improve those lifetime and usage. This paper presents a novel method of online adaptation for the load distribution of production processes within industrial direct current (DC) microgrids. The online load profile cycle recognition was used to adapt the energy distribution among the sources and loads in the DC microgrid. These sources can be inverters, rectifiers, energy storage systems or decentralized power supply units, such as photo voltaic systems. The approach consists of three major points, the load profile cycle recognition, the load profile analysis and the online adaptation of the energy distribution. This solution was tested in simulation and in experiment with a test rig, that contains an inverter and an energy storage system. The results show, that the load profile will be recognized latest from the third cycle and that the imbalance between charging and discharging amounts of the energy storage is less than 0.6% for each cycle after adaptation.

show abstract

“…Smart devices such as wearable and mobile sensors are increasingly utilized for health monitoring and personalized behavioral medicine. These technologies use machine-learning/deep-learning algorithms to detect lifestyle and physiological biomarkers and to provide real-time clinical interventions [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ]. However, the machine learning models are designed based on labeled training data collected in a particular domain, such as with a specific sensor modality, wearing site, or user.…”

Section: Introductionmentioning

confidence: 99%

Model-Agnostic Structural Transfer Learning for Cross-Domain Autonomous Activity Recognition

Alinia

Arefeen

Ashari

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

Activity recognition using data collected with smart devices such as mobile and wearable sensors has become a critical component of many emerging applications ranging from behavioral medicine to gaming. However, an unprecedented increase in the diversity of smart devices in the internet-of-things era has limited the adoption of activity recognition models for use across different devices. This lack of cross-domain adaptation is particularly notable across sensors of different modalities where the mapping of the sensor data in the traditional feature level is highly challenging. To address this challenge, we propose ActiLabel, a combinatorial framework that learns structural similarities among the events that occur in a target domain and those of a source domain and identifies an optimal mapping between the two domains at their structural level. The structural similarities are captured through a graph model, referred to as the dependency graph, which abstracts details of activity patterns in low-level signal and feature space. The activity labels are then autonomously learned in the target domain by finding an optimal tiered mapping between the dependency graphs. We carry out an extensive set of experiments on three large datasets collected with wearable sensors involving human subjects. The results demonstrate the superiority of ActiLabel over state-of-the-art transfer learning and deep learning methods. In particular, ActiLabel outperforms such algorithms by average F1-scores of 36.3%, 32.7%, and 9.1% for cross-modality, cross-location, and cross-subject activity recognition, respectively.

show abstract

CyclePro: A Robust Framework for Domain-Agnostic Gait Cycle Detection

Cited by 19 publications

References 37 publications

Online calibrated, energy-aware and heading corrected pedestrian navigation with foot-mounted MARG sensors

Online calibrated, energy-aware and heading corrected pedestrian navigation with foot-mounted MARG sensors

Load Profile Cycle Recognition for Industrial DC Microgrids with Energy Storage Systems

Model-Agnostic Structural Transfer Learning for Cross-Domain Autonomous Activity Recognition

Contact Info

Product

Resources

About