Search results

This paper presents the concept and design of a strain gauge sensor, which is a part of a pipe crawling robot and can be used for the purpose of detecting the shape of a collapsed pipe or tunnels like voids within rubble. The paper illustrates the sensor's working principle. The sensor uses a spline interpolation algorithm combined with a looking‐up table method to estimate the void shape. The paper also presents the software interface and the laboratory experiment of the sensor. Discussions about the experimental results are given in the later part of the paper.

Dyadic interactions are significant for human life. Most body sensor networks-based research studies focus on daily actions, but few works have been done to recognize affective actions during interactions. The purpose of this paper is to analyze and recognize affective actions collected from dyadic interactions.

A framework that combines hidden Markov models (HMMs) and k-nearest neighbor (kNN) using Fisher kernel learning is presented in this paper. Furthermore, different features are considered according to the interaction situations (positive situation and negative situation).

Three experiments are conducted in this paper. Experimental results demonstrate that the proposed Fisher kernel learning-based framework outperforms methods using Fisher kernel-based approach, using only HMMs and kNN.

The research may help to facilitate nonverbal communication. Moreover, it is important to equip social robots and animated agents with affective communication abilities.

The presented framework may gain strengths from both generative and discriminative models. Further, different features are considered based on the interaction situations.

Existing studies on human activity recognition using inertial sensors mainly discuss single activities. However, human activities are rather concurrent. A person could be walking while brushing their teeth or lying while making a call. The purpose of this paper is to explore an effective way to recognize concurrent activities.

Concurrent activities usually involve behaviors from different parts of the body, which are mainly dominated by the lower limbs and upper body. For this reason, a hierarchical method based on artificial neural networks (ANNs) is proposed to classify them. At the lower level, the state of the lower limbs to which a concurrent activity belongs is firstly recognized by means of one ANN using simple features. Then, the upper-level systems further distinguish between the upper limb movements and infer specific concurrent activity using features processed by the principle component analysis.

An experiment is conducted to collect realistic data from five sensor nodes placed on subjects’ wrist, arm, thigh, ankle and chest. Experimental results indicate that the proposed hierarchical method can distinguish between 14 concurrent activities with a high classification rate of 92.6 per cent, which significantly outperforms the single-level recognition method.

In the future, the research may play an important role in many ways such as daily behavior monitoring, smart assisted living, postoperative rehabilitation and eldercare support.

To provide more accurate information on people’s behaviors, human concurrent activities are discussed and effectively recognized by using a hierarchical method.

In sensor-based activity recognition, most of the previous studies focused on single activities such as body posture, ambulation and simple daily activities. Few works have been done to analyze complex concurrent activities. The purpose of this paper is to use a statistical modeling approach to classify them.

In this study, the recognition problem of concurrent activities is explored with the framework of parallel hidden Markov model (PHMM), where two basic HMMs are used to model the upper limb movements and lower limb states, respectively. Statistical time-domain and frequency-domain features are extracted, and then processed by the principal component analysis method for classification. To recognize specific concurrent activities, PHMM merges the information (by combining probabilities) from both channels to make the final decision.

Four studies are investigated to validate the effectiveness of the proposed method. The results show that PHMM can classify 12 daily concurrent activities with an average recognition rate of 93.2 per cent, which is superior to regular HMM and several single-frame classification approaches.

A statistical modeling approach based on PHMM is investigated, and it proved to be effective in concurrent activity recognition. This might provide more accurate feedback on people’s behaviors.

The research may be significant in the field of pervasive healthcare, supporting a variety of practical applications such as elderly care, ambient assisted living and remote monitoring.

This work introduces the concept of a shape reconfigurable brush robot used for work in collapsed buildings or tunnels. This paper presents the bristle mechanism and traction experiments relating to a robot, which is designed to be able to negotiate pipes with variable cross section or ill constrained tunnel‐like voids within rubble. Traction experiments in the laboratory were used to investigate the characteristics of bristles and the performance of the brush units of different shapes. The experimental results are used to analyse the interaction between brush units and different shaped boxes and related to bristle characteristics with a view to give guidance for the design of a future brush‐based shape reconfigurable robot.

The purpose of this paper is to reduce the calculation burden and speed up the estimation process of Allan variance method while ensuring the exactness of the analysis results.

A series of six‐hour static tests have been implemented at room temperature, and the static measurements have been collected from MEMS IMU. In order to characterize the various types of random noise terms for the IMU, the basic definition and main procedure of the Allan variance method are investigated. Unlike the normal Allan variance method, which has the shortcomings of processing large data sets and requiring long computation time, a modified Allan variance method is proposed based on the features of data distribution in the log‐log plot of the Allan standard deviation versus the averaging time.

Experiment results demonstrate that the modified Allan variance method can effectively estimate the noise coefficients for MEMS IMU, with controllable computation time and acceptable estimation accuracy.

This paper proposes a time‐controllable Allan variance method which can quickly and accurately identify different noise terms imposed by the stochastic fluctuations.

This study aims to investigate locomotion mechanisms of different urban search and rescue (USAR) robots currently being researched or commercially available on the market.

USAR robots are categorized by the type of their mobility. Detailed illustration and analysis have been given for each USAR robot in the paper.

The paper finds that none of current USAR robots can practically and autonomously carry out rescue work in a complex and unstructured environment. Hence, responding to the practical requirements of highly challenging USAR tasks, a team of USAR robots based on different locomotion mechanisms may be a good solution to undertake rescue activities.

The paper provides guidance in the design of future USAR robots.

The paper investigates locomotion mechanisms of different USAR robots in detail.

The purpose of this paper is to present an adaptive numerical algorithm for forward kinematics analysis of general Stewart platform.

Unlike the convention of developing a set of kinematic equations and then solving them, an alternative numerical algorithm is proposed in which the principal components of link lengths are used as a bridge to analyze the forward kinematics of a Stewart platform. The values of link lengths are firstly transformed to the values of principal components through principal component analysis. Then, the computation of the values of positional variables is transformed to a two‐dimensional nonlinear minimization problem by using the relationships between principal components and positional variables. A hybrid Nelder Mead‐particle swarm optimizer (NM‐PSO) algorithm and a modified NM algorithm are used to solve the two‐dimensional nonlinear minimization problem.

Simulation experiments have been conducted to validate the numerical algorithm and experimental results show that the numerical algorithm is valid and can achieve good accuracy and high efficiency.

This paper proposes an adaptive numerical algorithm for forward kinematics analysis of general Stewart platform.

Due to the complex mechanism during walking, human gait takes plenty of information reflecting human motion. The method of quantitative measurement of gait makes a profound influence in many fields, such as clinical medicine, biped robot control strategy and so on. The purpose of this paper is to present a gait analysis system based on inertial measurement unit (IMU) and combined with body sensor network (BSN).

The authors placed two wireless inertial nodes on the left and right ankles, so that the acceleration and angular velocity could be obtained from both sides at the same time. By using the kinematical model of the human gait, many methods such as time series analysis, pattern recognition and numerical analysis, are introduced to fuse the inertial data and estimate the sagittal gait parameters.

The gait parameters evaluation gains a practical precision, especially in the gait phase detection and the process of how the two feet cooperate with each other has been analyzed to learn about the mechanism of biped walking.

The gait analysis procedure is off line, so that the system ensures sampling at a high rate.

This gait analysis system can be utilized to measure quantitative gait parameters. Further, the coordination of dual gait pattern is presented. Last but not least, the system can also be put into capturing and analyzing the motion of other parts of the body.

The purpose of this paper is to develop a health monitoring system that can measure human vital signs and recognize human activity based on body sensor network (BSN).

The system is mainly composed of electrocardiogram (ECG) signal collection node, blood oxygen signal collection node, inertial sensor node, receiving node and upper computer software. The three collection nodes collect ECG signals, blood oxygen signals and motion signals. And then collected signals are transmitted wirelessly to receiving node and analyzed by software in upper computer in real-time.

Experiment results show that the system can simultaneously monitor human ECG, heart rate, pulse rate, SpO₂ and recognize human activity. A classifier based on coupled hidden Markov model (CHMM) is adopted to recognize human activity. The average recognition accuracy of CHMM classifier is 94.8 percent, which is higher than some existent methods, such as supported vector machine (SVM), C4.5 decision tree and naive Bayes classifier (NBC).

The monitoring system may be used for falling detection, elderly care, postoperative care, rehabilitation training, sports training and other fields in the future.

First, the system can measure human vital signs (ECG, blood pressure, pulse rate, SpO₂, temperature, heart rate) and recognizes some specific simple or complex activities (sitting, lying, go boating, bicycle riding). Second, the researches of using CHMM for activity recognition based on BSN are extremely few. Consequently, the classifier based on CHMM is adopted to recognize activity with ideal recognition accuracies in this paper.