Tracking Sensor Research Articles

OGC P14 A quantitative method to enhance endoscopic measurement

Abstract Background Upper gastrointestinal white light endoscopy is the most used procedure for assessing upper gastrointestinal tract condition and identification of various types of lesions. The characteristics of lesions may influence subsequent treatment steps. Among these characteristics, the size of a lesion plays a crucial role in determining the surveillance interval, appropriate resection method and predicting the risk of malignancy. Visual size estimation is highly subjective although this is the method currently used in clinical practice. Another important characteristic is the lesion location, which is currently estimated using the markers on the scope. However, in the stomach the endoscope may be curved to visualise the lesion, making location measurements inaccurate. Identifying the accurate location of the lesion is important for surgical planning and help in deciding the extent of surgical resection required. We aimed to improve endoscopic measurements by developing a quantitative method that can provide lesion size and location measurements in an accurate, objective, contactless and automatic way during endoscopy procedure. Methods A miniaturised electromagnetic tracking sensor was places inside the instrument channel of the endoscope to trace the pose of the camera scope. An algorithm based on feature points tracking was used to identify reference points on each image of interest; a triangulation technique was then used to determine the size of a lesion and its precise location with respect to desired anatomical landmark. Surgeons only require to take two different pictures of the lesion and one single picture of any anatomical landmark that they are interested in knowing the size and location information. At the same time corresponding tracking information were recorded for each captured image. Using these images and recorded tracking information (i.e. position and orientation), our proposed method computed the size of the lesion and its distance with respect to the selected anatomical landmark. The proposed quantitative endoscopy method was tested using a simulated plastic upper gastrointestinal endoscopy model with an artificial lesion. Results 10 different image sets of the same lesion were processed using the model of the upper gastrointestinal tract. A first picture was captured as soon as the endoscope had passed the gastroesophageal junction; two additional images were then captured from different angles when the scope was close to the lesion. The artificial lesion was a rounded shape lesion (diameter 12mm) placed 15.2 cm from gastroesophageal junction. Using the proposed method, the estimated distance from the gastrointestinal junction 15.17 ± 0.51 cm (mean ± SD); the assessed size was 12.76 ± 0.83 mm. The difference between the mean of 10 measurements and the actual values were 2.28 mm for the distance and 0.76 mm for the lesion size. The root mean square error was 5.14mm (distance) and 1.12 mm (size). Conclusions We have developed a quantitative endoscopy method and validated its accuracy in a simulated plastic upper gastrointestinal endoscopy model. This method may potentially be better than the deep learning approaches which require a huge amount of training data set. Our method can also compute the size directly whereas deep learning based algorithms can only classify the lesion into different size groups. There is also no limitation for the size of the lesion, as our method is tool independent. Using the same method and configuration the lesion can be accurately localised, which is also clinically important. The magnitude of error of the proposed quantitative endoscopy method is clinically acceptable and has the potential to highly improve the efficiency of endoscopic diagnosis and enhance patient outcome.

A Narrative Review on Wearable Inertial Sensors for Human Motion Tracking in Industrial Scenarios

Industry 4.0 has promoted the concept of automation, supporting workers with robots while maintaining their central role in the factory. To guarantee the safety of operators and improve the effectiveness of the human-robot interaction, it is important to detect the movements of the workers. Wearable inertial sensors represent a suitable technology to pursue this goal because of their portability, low cost, and minimal invasiveness. The aim of this narrative review was to analyze the state-of-the-art literature exploiting inertial sensors to track the human motion in different industrial scenarios. The Scopus database was queried, and 54 articles were selected. Some important aspects were identified: (i) number of publications per year; (ii) aim of the studies; (iii) body district involved in the motion tracking; (iv) number of adopted inertial sensors; (v) presence/absence of a technology combined to the inertial sensors; (vi) a real-time analysis; (vii) the inclusion/exclusion of the magnetometer in the sensor fusion process. Moreover, an analysis and a discussion of these aspects was also developed.

Comparative Analysis of Vehicle-Based and Driver-Based Features for Driver Drowsiness Monitoring by Support Vector Machines

Driver drowsiness is a serious threat to road safety. Most driver monitoring systems (DMSs) already embedded in vehicles to detect drowsiness use vehicle-based features (i.e., measures) computed by outward-facing cameras for lane tracking or steering wheel angle sensors to analyze lane keeping and steering control behavior. Such DMSs are referred to as indirect DMSs as they monitor drowsiness indirectly through driving performance. In this work, we extend this classical technique by using a driver monitoring camera for tracking driver-based features associated with eye blinking behavior and head movements. We refer to DMSs based only on a driver monitoring camera as direct DMSs as they monitor drowsiness directly through observable driver-based behavioral cues. In this work, we conduct a comparative analysis between an indirect and direct DMS. We also combine vehicle-based and driver-based features to examine the potential of a so-called hybrid DMS. To this end, we use a database collected from 70 participants in driving simulator experiments. The comparative analysis is performed by means of the correlation-based feature selection technique and support vector machines independently. With a balanced accuracy of 87.1%, the direct DMS significantly outperforms the indirect DMS, which reaches a balanced accuracy of only 77.9%. The hybrid DMS achieves a slightly better balanced accuracy of 87.7%. This work motivates the development and use of direct or hybrid DMSs to detect driver drowsiness and increase road safety.

Developing Acute Event Risk Profiles for Older Adults with Dementia in Long-Term Care Using Motor Behavior Clusters Derived from Deep Learning.

This paper uses deep (machine) learning techniques to develop and test how motor behaviors, derived from location and movement sensor tracking data, may be associated with falls, delirium, and urinary tract infections (UTIs) in long-term care (LTC) residents. Longitudinal observational study. A total of 23 LTC residents (81,323 observations) with cognitive impairment or dementia in 2 northeast Department of Veterans Affairs LTC facilities. More than 18months of continuous (24/7) monitoring of motor behavior and activity levels used objective radiofrequency identification sensor data to track and record movement data. Occurrence of acute events was recorded each week. Unsupervised deep learning models were used to classify motor behaviors into 5 clusters; supervised decision tree algorithms used these clusters to predict acute health events (falls, delirium, and UTIs) the week before the week of the event. Motor behaviors were classified into 5 categories (Silhouette score= 0.67), and these were significantly different from each other. Motor behavior classifications were sensitive and specific to falls, delirium, and UTI predictions 1week before the week of the event (sensitivity range= 0.88-0.91; specificity range= 0.71-0.88). Intraindividual changes in motor behaviors predict some of the most common and detrimental acute events in LTC populations. Study findings suggest real-time locating system sensor data and machine learning techniques may be used in clinical applications to effectively prevent falls and lead to the earlier recognition of risk for delirium and UTIs in this vulnerable population.

Humanoid robots coming to aid in dementia care provision: 16 humanoid robots working to augment care in 8 nursing homes across Minnesota.

Dementia is a growing problem that is the cause for an increase in care provision needs. Staff shortages are having an adverse effect by creating an atmosphere where the people affected with dementia feel neglected and isolated. This is resulting in reduced quality of life for the people residing in assisted living facilities and nursing homes. Family members and caregivers have been forced to reduce contact with the residents due to the fear of infection given the vulnerability of the residents. The residents are losing emotional stimulation and this is adversely affecting their disease progression. Freeing caregivers of mundane caregiving activities that require more patience and less caregiving in activities such as reminding/prompting, watching and monitoring can allow them to focus on activities that require more attention and will benefit the residents. Repetitive activities such as prompting, reminding of daily living activities such as brushing teeth and hair, washing hands, keeping the oxygen or face mask on etc. can be repetitive and stressful for caregivers. If these activities can be performed by a humanoid robot that can repeat the same activity or prompt without getting frustrated or changing the tone, the caregivers will have more time to tend to activities that require more attention such as medication management, tending to finances, medical followup appointments etc. University of Minnesota is collaborating with nursing homes in deploying robots to help augment care in the domains of emotion, cognition and physical wellbeing. These robots will have tasks assigned to them that they will perform throughout the day. They work in conjunction with sensors that will provide them with stimulation and feedback for tracking and monitoring activities of daily living. The sensors will also track the physiological well being to ensure the residents are not adversely affected by the robots. Humanoid robots are being deployed in nursing homes. A combination of humanoid robots and wearable and spatial sensors track and monitor care provision for residents at nursing homes. The humanoid robots are helping fill staff shortage gaps. Most importantly residents feel more engaged and have an improved sense of wellbeing.

Hand Gesture based Speech Recognition system for Hard of Hearing People

Abstract: For people who are intellectually challenged, this document provides a sign to speech translator. In contemporary society, the dumb have very little ability to converse with the normal person. Therefore, they are unable to interact with common people until common people, like ourselves, learn how to communicate through sign language. Dumb sign language is not something that everyone can learn because it is so challenging to perfect. Because of this, nobody can go and talk to these physically challenged persons. So, here is a system that would make it possible for those who are illiterate to speak with everyone. With today's technology, even the deaf may communicate by making hand gestures. Each hand gesture is set to play back a different audio message. Hand motions were captured by a flex sensor and an accelerometer, and the microcontroller then accepted these inputs. The flex sensor tracks hand movements and detects deflections, converting them into the right signals that trigger the system to produce recorded sound signals for transmission via the speakers. There are many different Arduino Boards available right now, and they can be used for a variety of things. The Arduino Mega 2560 is the ideal solution since it is the most cost-effective, allows faster computing, and can integrate and interface various sensors while consuming less power. The device under discussion is basically made up of a glove and a microcontroller-based system. Four strategically placed flex sensors are included in the data glove.

Autonomous Resonance-Tuning Mechanism for Environmental Adaptive Energy Harvesting.

An innovative autonomous resonance-tuning (ART) energy harvester is reported that utilizes adaptive clamping systems driven by intrinsic mechanical mechanisms without outsourcing additional energy. The adaptive clamping system modulates the natural frequency of the harvester's main beam (MB) by adjusting the clamping position of the MB. The pulling force induced by the resonance vibration of the tuning beam (TB) provides the driving force for operating the adaptive clamp. The ART mechanism is possible by matching the natural frequencies of the TB and clamped MB. Detailed evaluations are conducted on the optimization of the adaptive clamp tolerance and TB design to increase the pulling force. The energy harvester exhibits an ultrawide resonance bandwidth of over 30Hz in the commonly accessible low vibration frequency range (<100Hz) owing to the ART function. The practical feasibility is demonstrated by evaluating the ART performance under both frequency and acceleration-variant conditions and powering a location tracking sensor.

PID Controllers Performance On Dual Axis Tracking With Tetrahedron Based Sensor

This study compares control systems applied to a dual-axis tetrahedron-based sensor tracker. A tetrahedron-based sensor is a tracking sensor that can detect the coordinates of a light source. This study aims to determine a control system that can control sensors with high accuracy and precision and has a fast-tracking ability. Tests are carried out periodically by providing light at certain coordinates. After carrying out the testing and analysis process, it is concluded that the P controller is a better control system than the other controllers. This controller can control sensors with high accuracy and precision compared to PI, PD, and PID control systems. The P controller can also control the sensor to move towards the light coordinates with a travel time of 1.6 seconds on the X-axis and 3.1 seconds on the Y-axis, with a MAE value of 1.1 on the X-axis and 0.3 on the Y-axis. While the RSME value obtained is 1.33 on the X-axis and 0.55 on the Y-axis.

Design and Implementation of a Personalizable Alternative Mouse and Keyboard Interface for Individuals with Limited Upper Limb Mobility

People with neuromuscular diseases often experience limited upper limb mobility, which makes the handling of standard computer mice and keyboards difficult. Due to the importance of computers in private and professional life, this work aims at implementing an alternative mouse and keyboard interface that will allow for their efficient use by people with a neuromuscular disease. Due to the strongly differing symptoms of these diseases, personalization on the hardware and software levels is the focus of our work. The presented mouse alternative is based on a spectacle frame with an integrated motion sensor for head tracking, which enables the control of the mouse cursor position; the keyboard alternative consists of ten keys, which are used to generate word suggestions for the user input. The interface was tested in a user study involving three participants without disabilities, which showed the general functionality of the system and potential room for improvement. With an average throughput of 1.56 bits per second achieved by the alternative mouse and typing speeds of 8.44 words per minute obtained using the alternative keyboard, the proposed interface could be a promising input device for people with limited upper limb mobility.

ESS-IoT: The Smart Waste Management System for General Household Shen Yuong Wong, Huashuo Han, Kin Meng Cheng, Ah Choo Koo and Salman Yussof

With the urban population’s growth, unethical and unmanaged waste disposal may negatively impact the environment. In many cities, a massive flow of people in municipal buildings or offices has generated vast amounts of waste daily, which correlates to the enormous expenses of waste management. The critical issue for better waste management is waste collection and sorting. In this study, the Electronic Smart Sorting- Internet of Things (ESS-IoT) is proposed to assist people in better waste management. The ESS-IoT system uses Raspberry Pi 4b as the microcontroller with three modules, and it is designed with two main functions: waste collection and waste classification. The two main functions have been deployed separately in the literature, while this study has combined both functions to achieve a more comprehensive smart bin waste disposal solution. Waste collection is triggered by the overflow alarm mechanism that employs ultrasonic and tracker sensors. On the other hand, the waste classification is implemented using two classification algorithms: Random Forest (RF) prediction model and Convolutional Neural Network (CNN) prediction model. An experiment is conducted to evaluate the accuracy of the two classification algorithms in classifying various types of waste. The waste materials under investigation can be classified into four categories: kitchen waste, recyclables, hazardous waste, and other waste. The results show that CNN is the better classification algorithm between the two. Future work proposes the research extension by introducing an incentive mechanism to motivate the household communities using a cloud-based competition platform incorporated with the ESS-IoT system.

Sequential State and Unknown Parameter Estimation Strategy and its Application to a Sensor Fusion Problem

Soft-sensing and state-space-based methods play crucial roles in fault-tolerant sensor fusion schemes. Kalman filtering is one of the most employed sensor fusion methods. In practice, filtering (or state estimation) is sensitive to initial conditions, variations in model parameters, and faults in sensors and actuators. Cumulatively, these uncertainties that affect the state estimator can be modeled as unknown parameters. In this article, we present the development of a discrete-time minimum variance unbiased estimator (MVUE) that is robust to a class of unknown parameters. The estimation gain of MVUE is designed to facilitate unbiased state estimates in the presence of unknown parameters. We then use the innovations that are generated by the proposed state estimator (MVUE) to estimate the unknown parameters, thus proposing a sequential state and unknown parameter estimation strategy. The development of the decoupled state and unknown parameter estimation strategy is motivated by the fault-tolerant sensor fusion exercise for railway track geometry inspection. This exercise aims to formulate the state-space realization of the track geometry inspection problem to enable sequential state and unknown parameter estimation. The track geometry parameters, initial conditions, and sensor faults are considered unknown parameters for the application. The simulation analysis shows the importance of developing the trinity of the state-space realization for track geometry inspection, the state estimation framework, and the decoupled unknown parameter estimator.

The design, construction, operation and performance of the Belle II silicon vertex detector

The Silicon Vertex Detector of Belle II is a state-of-the-art tracking and vertexing system based on double-sided silicon strip sensors, designed and fabricated by a large international collaboration in the period 2012–2018. Since 2019 it has been in operation providing high quality data with a small number of defective channels (<1%), a large hit-finding efficiency (>99%), a good signal-to-noise ratio (well in excess of 10 for all sensor configurations and tracks). Together with the good control over the alignment, these are all essential factors to achieve good tracking reconstruction and physics performance. In this extended paper we try to document all the aspects of the SVD challenges and achievements, in the spirit of providing information to the broader community and help the development of high quality detector systems, which are essential tools to carry out physics research.

Distributed Network of Adaptive and Self-Reconfigurable Active Vision Systems

The performance of a computer vision system depends on the accuracy of visual information extracted by the sensors and the system’s visual-processing capabilities. To derive optimum information from the sensed data, the system must be capable of identifying objects of interest (OOIs) and activities in the scene. Active vision systems intend to capture OOIs with the highest possible resolution to extract the optimum visual information by calibrating the configuration spaces of the cameras. As the data processing and reconfiguration of cameras are interdependent, it becomes very challenging for advanced active vision systems to perform in real time. Due to limited computational resources, model-based asymmetric active vision systems only work in known conditions and fail miserably in unforeseen conditions. Symmetric/asymmetric systems employing artificial intelligence, while they manage to tackle unforeseen environments, require iterative training and thus are not reliable for real-time applications. Thus, the contemporary symmetric/asymmetric reconfiguration systems proposed to obtain optimum configuration spaces of sensors for accurate activity tracking and scene understanding may not be adequate to tackle unforeseen conditions in real time. To address this problem, this article presents an adaptive self-reconfiguration (ASR) framework for active vision systems operating co-operatively in a distributed blockchain network. The ASR framework enables active vision systems to share their derived learning about an activity or an unforeseen environment, which learning can be utilized by other active vision systems in the network, thus lowering the time needed for learning and adaptation to new conditions. Further, as the learning duration is reduced, the duration of the reconfiguration of the cameras is also reduced, yielding better performance in terms of understanding of a scene. The ASR framework enables resource and data sharing in a distributed network of active vision systems and outperforms state-of-the-art active vision systems in terms of accuracy and latency, making it ideal for real-time applications.

Characterization of irradiated passive CMOS sensors for tracking in HEP experiments

The upgrades to accelerator-based HEP detectors (e.g. ATLAS, CMS) as well as future projects (e.g. CEPC, FCC) feature large-area silicon-based tracking detectors. Using production lines of industrial CMOS foundries to fabricate silicon radiation detectors, both for pixels and for large-area strip sensors would be most beneficial in terms of availability, throughput and cost. First samples of pixel sensors coming from the LFoundry production line have already been tested and showed good performance up to irradiation levels of 1×1016neq cm−2 for their potential operations as sensors for the CMS inner tracker. This paper will focus on the characterization of pixel modules at high irradiation levels, up to 2×1016neq cm−2, studying the performance in terms of charge collection, position resolution and hit efficiency with measurements performed in the laboratory and with beam tests.

Analysis of Inertial Measurement Accuracy using Complementary Filter for MPU6050 Sensor

Inertial can be defined as disinclination to motion, action, or change. The inertia of an object is the propensity to remain at rest or if in motion, stays in motion at a steady speed. MPU6050 is one of the low-cost motion tracking sensors that contain a 3-axis gyroscope and 3-axis accelerometer orientation measurement. It is used to analyse the movement or location of a person in an indoor environment. This research is to analyse the accuracy of the inertial measurement of the MPU 6050 sensor. Next, is to improve the achievable accuracy rate up to 95% using the complementary filter and finally to visualize the results on an IoT platform. This MPU6050 sensor is beneficial to an emergency responder such as the firefighter’s department. The accurate inertial measurement and location will help to detect the movement and the motion of the firefighter during operation, especially in an indoor environment. The sensor will detect and collects the inertial measurement of an emergency responder and transmit the data wirelessly by using ESP8266 NodeMCU. Finally, the results can be viewed on an IoT platform. However, the results obtained from the MPU 6050 sensor is not perfectly accurate as there is noise during the measurement. Therefore, a complementary filter is used and analysed in this research. It is expected that the inertial location’s accuracy could be improved by 95% that will provide a precise movement and location of the firefighter during operation.

Fusion of Heterogenous Sensor Data in Border Surveillance.

Wide area surveillance has become of critical importance, particularly for border control between countries where vast forested land border areas are to be monitored. In this paper, we address the problem of the automatic detection of activity in forbidden areas, namely forested land border areas. In order to avoid false detections, often triggered in dense vegetation with single sensors such as radar, we present a multi sensor fusion and tracking system using passive infrared detectors in combination with automatic person detection from thermal and visual video camera images. The approach combines weighted maps with a rule engine that associates data from multiple weighted maps. The proposed approach is tested on real data collected by the EU FOLDOUT project in a location representative of a range of forested EU borders. The results show that the proposed approach can eliminate single sensor false detections and enhance accuracy by up to 50%.

Uncertainty Analysis Study on Seakeeping Tests of Benchmark Model

Experimental uncertainty and measurement, particularly in engineering, have been increasingly emphasized over the last 20 years. Uncertainty analysis of ship model testing in seakeeping experiments based on International Organization for Standardization, Guide for Uncertainty of Measurements (ISO-GUM), report with recommendations published by ITTC (2008). In this paper, the uncertainties for free-running tests were performed for a model scale 1:62 as a benchmark model test at the Indonesian Hydrodynamic Laboratory (IHL). The benchmark model is representative of a full scale of 186 meters. The model has full appendages as well as a rudder and propellers to allow for free-running tests, natural roll period, free roll decay, and swing tests. Factors of basic uncertainty (type A and type B) are estimated using model tests. The standard deviation of the mean value of a repeated measurement is uncertainty type A. However, Type B uncertainty is estimated from manufacturing specifications. The uncertainty components resulting from both types are quantified by standard deviations. The uncertainty value of Type B is approximated by a corresponding variance. This study helps to understand the underlying uncertainty associated with this research, which resulted in a total geometry uncertainty value (Lpp, B, T, H, , and kyy) of 0.17598% and the total uncertainty value of the instrument calibration results (wave height sensor and qualysis motion tracking) is 0.01161%. The results of this study are expected to minimize the impact of sources of uncertainty and allow the potential for underlying uncertainty to be corrected for the next seaworthiness test.

Development and integration of laser sensor tracking system in robotic arm for path correction during welding operation

&lt;span lang="EN-US"&gt;An industrial robot is mainly used for manufacturing. Industrial robots are 6 or more axes, which can be automatically controlled by programming. Typical applications of robots include welding, painting, picking, and placing printed circuit boards, packaging and labelling, palletizing, product inspection, and testing with high accuracy, precision, and fast speed. Robotic welding is a complex, nonlinear and time varying process which can be affected by various natural or any random disturbances. Due to the effect of various factors, the actual welding path may differ. So, welding robots should be able to detect the actual welding path, then adjust the difference in welding path and complete the welding process accurately. Laser welding is one of the most important technologies in the manufacturing field. It is the most frequently used technology which has made new demands. So, the manufacturer ensures to meet the quality of laser welding and improve the production efficiency. Due to the increase in demand of quality, accuracy, precise, productivity, flexibility and adaptive control of welding robot, an automatic laser seam tracking system is developed with welding robot to precisely follow the welding path and make the necessary corrections during welding operations.&lt;/span&gt;

Sensor Action Recognition, Tracking, and Optimization Analysis in Training Process Based on Virtual Reality Technology

In order to solve the problem of improving the training efficiency and competition performance of athletes, a system of sensor action recognition, tracking, and optimization analysis based on virtual reality technology is proposed. After using the perceptual interaction subsystem to build a three-dimensional training scene through the virtual environment producer, the picture in the three-dimensional scene is rendered through the view module in the model operator system. At the same time, the logic module calculates the physical attenuation of athletes and sends the calculation results back to the perceptual interaction subsystem effect generator through the interface to present the interactive effect of athletes’ training. The experimental results show that the maximum improvement of the performance of five randomly selected athletes in the 100-meter, 200-meter, and 400-meter dash competitions after using the system can reach 5 s, 4.5 s, and 6 s, respectively. Conclusion. The system can effectively complete the calculation of athletes’ physical fitness attenuation and has good carrying capacity. It can improve athletes’ physical fitness training effect and competition simulation efficiency and can improve athletes’ competition performance after application.

Enhancing motion tracking accuracy of a low-cost 3D video sensor using a biomechanical model, sensor fusion, and deeplearning.

Low-cost 3D video sensors equipped with routines for extracting skeleton data facilitate the widespread use of virtual reality (VR) for rehabilitation. However, the accuracy of the extracted skeleton data is often limited. Accuracy can be improved using a motion tracker, e.g., using a recurrent neural network (RNN). Yet, training an RNN requires a considerable amount of relevant and accurate training data. Training databases can be obtained using gold-standard motion tracking sensors. This limits the use of the RNN trackers in environments and tasks that lack accessibility to gold-standard sensors. Digital goniometers are typically cheaper, more portable, and simpler to use than gold-standard motion tracking sensors. The current work suggests a method for generating accurate skeleton data suitable for training an RNN motion tracker based on the offline fusion of a Kinect 3D video sensor and an electronic goniometer. The fusion applies nonlinear constraint optimization, where the constraints are based on an advanced shoulder-centered kinematic model of the arm. The model builds on the representation of the arm as a triangle (the arm triangle). The shoulder-centered representation of the arm triangle motion simplifies constraint representation and consequently the optimization problem. To test the performance of the offline fusion and the RNN trained using the optimized data, arm motion of eight participants was recorded using a Kinect sensor, an electronic goniometer, and, for comparison, a passive-marker-based motion tracker. The data generated by fusing the Kinect and goniometer recordings were used for training two long short-term memory (LSTM) RNNs. The input to one RNN included both the Kinect and the goniometer data, and the input to the second RNN included only Kinect data. The performance of the networks was compared to the performance of a tracker based on a Kalman filter and to the raw Kinect measurements. The accuracy of the fused data was high, and it considerably improved data accuracy. The accuracy for both trackers was high, and both were more accurate than the Kalman filter tracker and the raw Kinect measurements. The developed methods are suitable for integration with immersive VR rehabilitation systems in the clinic and the home environments.