Geometrical kinematic modeling on human motion using method of multi-sensor fusion

Xu, Chunjing; He, Jie; Zhang, Xiaolin; Yao, Chi; Tseng, Po-Hsuan

doi:10.1016/j.inffus.2017.09.014

Cited by 42 publications

(28 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in Figure 1, the human body consists of 15 joints and 14 limbs. To better describe the complicated human body movements, Denavit-Hartenbe (D-H) equation [22,25] is generally used in the description of HMT features and parameters. Its measurement parameters mainly include the distances between joints (Figure 1b,c) joint and limbs ( Figure 1d) and the relative angles between limbs (Figure 1e,f).…”

Section: Model Descriptionmentioning

confidence: 99%

“…According to the Bayesian theorem [22,25], at time t, the state estimation x t−1 could be represented as:…”

Section: Model Presentationmentioning

confidence: 99%

“…State-of-the-art studies seldom concentrate on the fundamental limits of IMU/TOA fusion methods, and the error correction effects are not satisfying. For example, Nilsson et al [27] proposed a cooperative localization method by fusion of dual foot-mounted inertial sensors and inter-agent UWB ranging, but the experiment results show that, compared with the performance lower bounds [22], there is still a lot of room for improvement.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Towards Human Motion Tracking: Multi-Sensory IMU/TOA Fusion Method and Fundamental Limits

Zhang

et al. 2019

Electronics

Self Cite

View full text Add to dashboard Cite

Human motion tracking could be viewed as a multi-target tracking problem towards numerous body joints. Inertial-measurement-unit-based human motion tracking technique stands out and has been widely used in body are network applications. However, it has been facing the tough problem of accumulative errors and drift. In this paper, we propose a multi-sensor hybrid method to solve this problem. Firstly, an inertial-measurement-unit and time-of-arrival fusion-based method is proposed to compensate the drift and accumulative errors caused by inertial sensors. Secondly, Cramér-Rao lower bound is derived in detail with consideration of both spatial and temporal related factors. Simulation results show that the proposed method in this paper has both spatial and temporal advantages, compared with traditional sole inertial or time-of-arrival-based tracking methods. Furthermore, proposed method is verified in 3D practical application scenarios. Compared with state-of-the-art algorithms, proposed fusion method shows better consistency and higher tracking accuracy, especially when moving direction changes. The proposed fusion method and comprehensive fundamental limits analysis conducted in this paper can provide a theoretical basis for further system design and algorithm analysis. Without the requirements of external anchors, the proposed method has good stability and high tracking accuracy, thus it is more suitable for wearable motion tracking applications.

show abstract

Section: Model Descriptionmentioning

confidence: 99%

“…According to the Bayesian theorem [22,25], at time t, the state estimation x t−1 could be represented as:…”

Section: Model Presentationmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Towards Human Motion Tracking: Multi-Sensory IMU/TOA Fusion Method and Fundamental Limits

Zhang

et al. 2019

Electronics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Information security risk refers to the risks caused by the hardware and software defects, system integration defects and information security management potential of the weak links, existing in various types of application systems and their basic networks and the data during the information construction process [4] [5].…”

Section: The Risk Of Iotmentioning

confidence: 99%

Analytic Hierarchy Process Based Risk Assessment Method for Internet of Things

Wang¹

2018

dteees

View full text Add to dashboard Cite

Abstract. Internet of Things (IoT for short) is an emerging Internet-based global information architecture that promotes the exchange of goods and services in the global supply chain network. IoT not only brings a lot of convenience to society but also faces a growing number of complex security risks. The basis of solving risk is accurate assessment of the possibility and the loss of the risk occurrence. However, in fact, when assess the risk, most companies still rely on personal work experience and knowledge to carry out emotional qualitative judgments, making the assessment results are not accurate. Through the combination of Analytic Hierarchy Process (AHP) and the practice of risk assessment, forming the risk assessment method using AHP, and the simulation experiment is carried out on the typical application of positioning system in IoT. The results show that the risk assessment method can be used to solve the risk assessment method which is combined with qualitative and quantitative analysis, which makes the result of risk assessment more scientific and accurate.

show abstract

“…To solve the first problem, most related works use data-driven methods which tend to take the advantage of multiple sensors [4], such as accelerometer, gyroscope, compass sensor, and humidity sensor, to name but a few, to enlarge the input data set to achieve more information. More than one sensor node is mounted onto different body-parts to monitor human motions with multiple degrees of freedom.…”

Section: Introductionmentioning

confidence: 99%

Recurrent Transformation of Prior Knowledge Based Model for Human Motion Recognition

Zhang

et al. 2018

Computational Intelligence and Neuroscience

Self Cite

View full text Add to dashboard Cite

Motion related human activity recognition using wearable sensors can potentially enable various useful daily applications. So far, most studies view it as a stand-alone mathematical classification problem without considering the physical nature and temporal information of human motions. Consequently, they suffer from data dependencies and encounter the curse of dimension and the overfitting issue. Their models are hard to be intuitively understood. Given a specific motion set, if structured domain knowledge could be manually obtained, it could be used for better recognizing certain motions. In this study, we start from a deep analysis on natural physical properties and temporal recurrent transformation possibilities of human motions and then propose a useful Recurrent Transformation Prior Knowledge-based Decision Tree (RT-PKDT) model for recognition of specific human motions. RT-PKDT utilizes temporal information and hierarchical classification method, making the most of sensor streaming data and human knowledge to compensate the possible data inadequacy. The experiment results indicate that the proposed method performs superior to those adopted in related works, such as SVM, BP neural networks, and Bayesian Network, obtaining an accuracy of 96.68%.

show abstract

Geometrical kinematic modeling on human motion using method of multi-sensor fusion

Cited by 42 publications

References 23 publications

Towards Human Motion Tracking: Multi-Sensory IMU/TOA Fusion Method and Fundamental Limits

Towards Human Motion Tracking: Multi-Sensory IMU/TOA Fusion Method and Fundamental Limits

Analytic Hierarchy Process Based Risk Assessment Method for Internet of Things

Recurrent Transformation of Prior Knowledge Based Model for Human Motion Recognition

Contact Info

Product

Resources

About