Ting Zhang

Robust Cyber-Physical Systems

In this paper we comprehensively survey the concept and strategies for building a resilient and integrated cyber-physical system (CPS). Here resilience refers to a 3S-oriented design, that is, stability, security, and systematicness: Stability means the CPS can achieve a stable sensing-actuation close-loop control even though the inputs (sensing data) have noise or attacks; Security means that the system can overcome the cyber-physical interaction attacks; and Systematicness means that the system has a seamless integration of sensors and actuators. We will also explain the CPS modeling issues since they serve as the basics of 3S design. We will use two detailed examples from our achieved projects to explain how to achieve arobust, systematic CPS design: Case study 1 is on the design of a rehabilitation system with cyber (sensors) and physical (robots) integration. Case Study 2 is on the implantable medical device design. It illustrates the nature of CPS security principle. The dominant feature of this survey is that it has both principle discussions and practical cyber-physical coupling design. Comprehensive survey on entire CPS design process.Qualitative and quantitative descriptions on CPS resilience.From basic concepts to case studies.Point out the future research trends.

Using decision trees to measure activities in people with stroke

Improving community mobility is a common goal for persons with stroke. Measuring daily physical activity is helpful to determine the effectiveness of rehabilitation interventions. In our previous studies, a novel wearable shoe-based sensor system (SmartShoe) was shown to be capable of accurately classify three major postures and activities (sitting, standing, and walking) from individuals with stroke by using Artificial Neural Network (ANN). In this study, we utilized decision tree algorithms to develop individual and group activity classification models for stroke patients. The data was acquired from 12 participants with stroke. For 3-class classification, the average accuracy was 99.1% with individual models and 91.5% with group models. Further, we extended the activities into 8 classes: sitting, standing, walking, cycling, stairs-up, stairs-down, wheel-chair-push, and wheel-chair-propel. The classification accuracy for individual models was 97.9%, and for group model was 80.2%, demonstrating feasibility of multi-class activity recognition by SmartShoe in stroke patients.

Classification of posture and activities by using decision trees

Obesity prevention and treatment as well as healthy life style recommendation requires the estimation of everyday physical activity. Monitoring posture allocations and activities with sensor systems is an effective method to achieve the goal. However, at present, most devices available rely on multiple sensors distributed on the body, which might be too obtrusive for everyday use. In this study, data was collected from a wearable shoe sensor system (SmartShoe) and a decision tree algorithm was applied for classification with high computational accuracy. The dataset was collected from 9 individual subjects performing 6 different activities-sitting, standing, walking, cycling, and stairs ascent/descent. Statistical features were calculated and the classification with decision tree classifier was performed, after which, advanced boosting algorithm was applied. The computational accuracy is as high as 98.85% without boosting, and 98.90% after boosting. Additionally, the simple tree structure provides a direct approach to simplify the feature set.

Preprocessing Design in Pyroelectric Infrared Sensor-Based Human-Tracking System: On Sensor Selection and Calibration

This paper presents an information-gain-based sensor selection approach as well as a sensor sensing probability model-based calibration process for multihuman tracking in distributed binary pyroelectric infrared sensor networks. This research includes three contributions: 1) choose the subset of sensors that can maximize the mutual information between sensors and targets; 2) find the sensor sensing probability model to represent the sensing space for sensor calibration; and 3) provide a factor graph-based message passing scheme for distributed tracking. Our approach can find the solution for sensor selection to optimize the performance of tracking. The sensing probability model is efficiently optimized through the calibration process in order to update the parameters of sensor positions and rotations. An application for mobile calibration and tracking is developed. Simulation and experimental results are provided to validate the proposed framework.

Virtual Reality Enhanced Robotic Systems for Disability Rehabilitation

This chapter mainly introduced the virtual reality as many benefits of robots involved in disability rehabilitation. According to the vision feedback and force feedback, the therapist can adjust his operation. Virtual reality technology can provide repeated practice, performance feedback and motivation techniques for rehabilitation training. Patients can learn motor skills in a virtual environment, and then transfer the skills to the real world. It is hopeful to achieve satisfactory outcome in the field of rehabilitation in the future. VR is mainly used for the upper-limb rehabilitation robot system in this article. The objective of robotic systems for disability rehabilitation are explored to divide the whole rehabilitation training process into three parts, earliest rehabilitation training, medium-term rehabilitation training and late rehabilitation training, respectively. Accordingly, brain-computer training modes, the masterslave training modes and the electromyogram (EMG) signals training modes are developed to be used in rehabilitation training to help stroke patients with hemiplegia to restore the motor function of upper limb. Aimed at the rehabilitation goal, three generations of VR rehabilitation system has designed. The first generation of VR rehabilitation system includes haptic device (PHANTOM Omni), an advanced inertial sensor (MTx) and a computer. The impaired hand grip the stylus of haptic device, the intact hand can control the impaired hand’s motion based on the virtual reality scene. The second generation of the VR rehabilitation system is the exoskeleton robots structure. Two virtual upper limbs are portrayed in the virtual environment, simulated the impaired hand and the intact hand, respectively. The third generation is a novel VR-based upper limb rehabilitation robot system. In the system, the realization of virtual reality environment is implemented, which can potentially motivate patients to exercise for longer periods of time. Not only virtual images but also position and force information are sent to the doctors. The development of this system can be a promising approach for further research in the field of tele-rehabilitation science. Virtual Reality Enhanced Robotic Systems for Disability Rehabilitation

A sensor-based virtual piano biofeedback system for stroke rehabilitation.

Approximately 15 million people suffer a stroke every year globally. People with stroke usually have less mobility and this may further reduce the fitness level and emotional wellbeing. Traditional stroke rehabilitation therapy is usually performed in a clinic or hospital, which evolves the care from the therapist. Robot-assisted stroke therapy can be done at the patients home, but the cost of the systems might be too high for some patients. In this study, we propose a novel sensor-based system for stroke rehabilitation. The system consists of a sensor-based digital glove and software running on a computer with a user interface to a piano. During the rehabilitation treatment, the patient plays the keys of a piano guided by the user interface on the screen. The system may enhance the mobility and flexibility of the affected upper limb and the fingers of stroke patients in an entertaining way. Our experimental results show that the system has a high accuracy of biofeedback.

Measuring gait symmetry in children with cerebral palsy using the SmartShoe

Cerebral palsy (CP) is a group of non-progressive neuro-developmental conditions occurring in early childhood that causes movement disorders and physical disability. Many affected children have impaired function in movement and limitations in mobility. Measuring gait symmetry is essential in assessing clinical outcomes of rehabilitation. Modern sensor technology has made it possible to measure gait unobtrusively in the community. However, no wearable systems that allow for gait symmetry measurement in free living have been investigated for children with CP. In this study, data was collected from three children with CP by a wearable shoe sensor system (SmartShoe) in a community environment and the gait symmetry ratio was estimated from the sensor data prior and post rehabilitation therapy. The sensor data were processed by algorithms including data preprocessing, posture and activity classification, and calculation of symmetry ratio of stance. The gait symmetry metrics extracted by the automatic algorithms closely match the metrics manually estimated on the sensor data with an average mean absolute error of 1.235%), suggesting that the proposed method may be an effective way to evaluate rehabilitation progress in the community setting.

Measuring activities and counting steps with the SmartSocks - An unobtrusive and accurate method

Physical inactivity is an important contributor to non-communicable diseases in countries of high income, and increasingly so in those of low and middle income. Physical inactivity is the leading cause of many diseases. It has been estimated that as many as 250,000 deaths per year in the United States, approximately 12% of the total, are attributable to a lack of regular physical activity. Measuring physical activities and counting steps is an effective method to diagnose some diseases. It can also serve as an effective method to encourage people to increase their physical activity. Pedometers have been invented as a convenient way of counting steps. However most of them lack the functionality of differentiating activities. Pressure sensor pads can measure steps and gait, but as the pad has a limited size, it can not meet the need of anytime and anywhere usage. In this study, we made the Sensor Socks for measuring physical activities and counting steps. It is unobtrusive and convenient for everyday usage. Our experimental results show that the system has a high accuracy of the classification of physical activities and counting steps in a home or community environment.

Wavelet Enhanced Image Preprocessing and Neural Networks for Hand Gesture Recognition

This paper presents a novel approach for hand gesture recognition based on wavelet enhanced image preprocessing and supervised Artificial Neural Networks (ANNs). Six different hand gestures are tested. The image preprocessing handles the hand gesture contour segmentation. This research includes three contributions: (1) it provides two dimensional hand gesture contour images to one dimensional signal conversion using reference points, (2) it implements wavelet decomposition for the 1D signals converted from 2D hand gesture contour images, and (3) it extracts 4 statistical features of the wavelet coefficients. The experimental results are provided to validate our proposed framework. The ANN is built to classify different hand gestures. There are totally 1240 images used for training and 240 images are used for testing. By using the proposed framework, our approach can provide classification accuracy of 97% and is fast in feature extraction and computation.