Inertial Measurement Unit Data Research Articles

Ergonomic Robotic Console Configuration in Gynecologic Surgery: An Interventional Study

The objectives of this study were to (1) pilot a robotic console configuration methodology to optimize ergonomic posture, and (2) determine the effect of this intervention on surgeon posture and musculoskeletal discomfort. This was an institutional review board-approved prospective cohort study conducted from February 2017 to October 2017. A single tertiary care midwestern academic medical center. Six fellowship-trained gynecologic surgeons, proficient in robotic hysterectomy, were recruited: 3 men and 3 women. Each surgeon performed 3 robotic hysterectomies using their self-selected robotic console settings (preintervention). Then, a robotic console ergonomic intervention protocol was implemented by trained ergonomists to improve posture and decrease time in poor ergonomic positions. Each surgeon then performed 3 robotic hysterectomies using the ergonomic intervention settings (postintervention). All surgeries used the da Vinci Xi surgical system (Intuitive Surgical, Inc., Sunnyvale, CA) and were the first case of the day. The surgeons wore inertial measurement unit (IMU) sensors on their head, chest, and bilateral upper arms during surgery. The IMU sensors are equipped with accelerometers, gyroscopes, and magnetometers to give objective measurements of body posture. IMU data were then analyzed to determine the percentage of time spent in ergonomically risky postures as categorized using a modified rapid upper limb assessment. Before and after each hysterectomy, the surgeons completed identical questionnaires for an assessment of musculoskeletal pain/discomfort. The outcome measurements were compared pre- versus postintervention on the basis of fitting generalized linear mixed models that handled the individual surgeon as a random effect and "setting" as a fixed effect. With regard to the IMU posture results, there was a significant decrease in time spent in the moderate- to high-risk neck position and a decrease in average neck angle after the ergonomic intervention. The average percentage of time spent in moderate- to high-risk categories was significantly lower for the neck (mean, 54.3% vs 21.0%; p = .008) and right upper arm (mean, 15.5% vs 0.9%; p = .02) when using the intervention settings compared with the surgeons' settings. Pain score results: There were fewer reported increases in neck (4 [22%] vs 1 [6%]) and right shoulder (4 [22%] vs 2 [11%]) pain or discomfort after completion of robotic hysterectomy postintervention versus preintervention; however, these differences did not attain statistical significance (p = .12 and p = .37, respectively). An ergonomic robotic console intervention demonstrated effectiveness and improved objective surgeon posture at the console when compared with the surgeons' self-selected settings.

Convolutional Neural Network for Driving Maneuver Identification Based on Inertial Measurement Unit (IMU) and Global Positioning System (GPS)

Identification and translation of different driving manoeuvre are some of the key elements to analysis driving risky behavior. However, the major obstacles to manoeuvre identification are the wide variety of styles of driving manoeuvre which are performed during driving. The objective in this contribution through the paper is to automatic identification of driver manoeuvre e.g. driving in roundabouts, left and right turns, breaks, etc. based on Inertia Measurement Unit (IMU) and Global Positioning System (GPS). Here, several Machine Learning (ML) algorithms i.e. Artificial Neural Network (ANN), Convolutional Neural Network (CNN), K-nearest neighbor (k-NN), Hidden Markov Model (HMM), Random Forest (RF), and Support Vector Machine (SVM) have been applied for automatic feature extraction and classification on the IMU and GPS data sets collected through a Naturalistic Driving Studies (NDS) under an H2020 project called SimuSafe . The CNN is further compared with HMM, RF, ANN, k-NN and SVM to observe the ability to identify a car manoeuvre through roundabouts. According to the results, CNN outperforms (i.e. average F1-score of 0.88 both roundabout and not roundabout) among the other ML classifiers and RF presents better correlation than CNN, i.e. MCC = -.022.

Recurrent Neural Network-Based Hybrid Localization for Worker Tracking in an Offshore Environment

Accidents involving marine crew members and passengers are still an issue that must be studied and obviated. Preventing such accidents at sea can improve the quality of life on board by ensuring a safe ship environment. This paper proposes a hybrid indoor positioning method, an approach which is becoming common on land, to enhance maritime safety. Specifically, a recurrent neural network (RNN)-based hybrid localization system (RHLS) that provides accurate and efficient user-tracking results is proposed. RHLS performs hybrid positioning by receiving wireless signals, such as Wi-Fi and Bluetooth, as well as inertial measurement unit data from smartphones. It utilizes the RNN to solve the problem of tracking accuracy reduction that may occur when using data collected from various sensors at various times. The results of experiments conducted in an offshore environment confirm that RHLS provides accurate and efficient tracking results. The scalability of RHLS provides managers with more intuitive monitoring of assets and crews, and, by providing information such as the location of safety equipment to the crew, it promotes welfare and safety.

Comparing different methods of gait speed estimation using wearable sensors in individuals with varying levels of mobility impairments.

Wearable sensors, such as inertial measurement units (IMU), provide the ability to quantify gait parameters outside of traditional gait laboratory settings. Walking speed has been shown to be associated with morbidity and mortality. Therefore, the ability of a clinician to easily and inexpensively measure gait speed within their clinic or patients' home setting can improve patient management and care. This study highlights multiple methods used to estimate patient walking speeds based only on IMU data and minimal anthropometric data, and identifies the algorithm appearing to be the most robust; one relying on identifying swing phases of gait first.Clinical relevance- Providing a clinician with a simple, inexpensive and reliable protocol for measuring patients' gait speed and other parameters could offer prevention and individualized care.

A Comparative Severity Assessment of Impaired Balance due to Cerebellar Ataxia using Regression Models.

Cerebellar ataxia (CA) refers to the impaired balance and coordination resulting from injury or degeneration of the cerebellum. Testing balance is one of the simplest means of assessing CA. This study compares instrumented assessment and clinical assessment scales of the balance test called Romberg's test. Inertial Measurement Unit (IMU) data were collected from a sensor attached to their chest of 53 subjects while they performed the test. The corresponding clinical scores were also tabulated. Using this data, 99 features were extracted to quantify acceleration, tremor and displacement of body sway. These features were filtered to identify the subset that better characterize the distinctive behavior of CA subjects. Elastic Net Regression model resulted a greater agreement (0.70 Pearson coefficient) with the clinical SARA scores. The overall results indicated that data from a single IMU sensor is sufficient to accurately assess balance in CA. The significance of this study is that evaluation of balance using Recurrence Quantification Analysis produces a comprehensive framework for the assessment of CA.

Road Profile Estimation Using a 3D Sensor and Intelligent Vehicle.

Autonomous vehicles can achieve accurate localization and real-time road information perception using sensors such as global navigation satellite systems (GNSSs), light detection and ranging (LiDAR), and inertial measurement units (IMUs). With road information, vehicles can navigate autonomously to a given position without traffic accidents. However, most of the research on autonomous vehicles has paid little attention to road profile information, which is a significant reference for vehicles driving on uneven terrain. Most vehicles experience violent vibrations when driving on uneven terrain, which reduce the accuracy and stability of data obtained by LiDAR and IMUs. Vehicles with an active suspension system, on the other hand, can maintain stability on uneven roads, which further guarantees sensor accuracy. In this paper, we propose a novel method for road profile estimation using LiDAR and vehicles with an active suspension system. In the former, 3D laser scanners, IMU, and GPS were used to obtain accurate pose information and real-time cloud data points, which were added to an elevation map. In the latter, the elevation map was further processed by a Kalman filter algorithm to fuse multiple cloud data points at the same cell of the map. The model predictive control (MPC) method is proposed to control the active suspension system to maintain vehicle stability, thus further reducing drifts of LiDAR and IMU data. The proposed method was carried out in outdoor environments, and the experiment results demonstrated its accuracy and effectiveness.

GraphIPS: Calibration-Free and Map-Free Indoor Positioning Using Smartphone Crowdsourced Data

Indoor positioning plays an important role in a variety of applications under Internet of Things (IoT). Conventional WiFi fingerprinting-based indoor positioning systems (IPSs) usually require extensive manual calibrations to construct radio maps. This process severely limits the system scalability and adaptiveness. Pedestrian dead reckoning (PDR) is a popular method that can avoid the calibration process. However, PDR-based IPSs typically suffer from accumulated errors. To tackle this problem, many refinement methods require map information or floorplans which may not be available or up-to-date in practice. With the development of IoT, various types of crowdsourced data become available. In this work, we propose GraphIPS, a calibration-free and map-free IPS which dynamically generates accurate radio maps by utilizing smartphone crowdsourced WiFi and inertial measurement unit (IMU) data. GraphIPS fuses the crowdsourced data into a graph-based formulation and applies the multidimensional scaling (MDS) algorithm to compute the positions of the user's steps. The experimental results show that GraphIPS achieves comparable accuracy to the calibration-based method in a significantly shorter run time than optimization-based methods. In addition to smartphones, GraphIPS is also potentially applicable for the smart wearables with embedded WiFi modules and IMUs.

Fusing and refining convolutional neural network models for assembly action recognition in smart manufacturing

Assembly carries paramount importance in manufacturing. Being able to support workers in real time to maximize their positive contributions to assembly is a tremendous interest of manufacturers. Human action recognition has been a way to automatically analyze and understand worker actions to support real-time assistance for workers and facilitate worker–machine collaboration. Assembly actions are distinct from activities that have been well studied in the action recognition literature. Actions taken by assembly workers are intricate, variable, and may involve very fine motions. Therefore, recognizing assembly actions remains a challenging task. This paper proposes to simply use only two wearable devices that respectively capture the inertial measurement unit data of each hand of workers. Then, two convolutional neural network models with an identical architecture are independently trained using the two sources of inertial measurement unit data to respectively recognize the right-hand and the left-hand actions of an assembly worker. Classification results of the two convolutional neural network models are fused to yield a final action recognition result because the two hands often collaborate in assembling operations. Transfer learning is implemented to adapt the action recognition models to subjects whose data have not been included in dataset for training the models. One operation in assembling a Bukito three-dimensional printer, which is composed of seven actions, is used to demonstrate the implementation and assessment of the proposed method. Results from the study have demonstrated that the proposed approach effectively improves the prediction accuracy at both the action level and the subject level. Work of the paper builds a foundation for building advanced action recognition systems such as multimodal sensor-based action recognition.

Mobile Robot Localization Based on Vision and Multisensor

To deal with the low accuracy of positioning for mobile robots when only using visual sensors and an IMU, a method based on tight coupling and nonlinear optimization is proposed to obtain a high-precision visual positioning scheme by combining measured value of the preintegrated inertial measurement unit (IMU) and values of the odometer and characteristic observations. First, the preprocessing part of the observation data includes tracking of the image data and the odometer data, and preintegration of IMU data. Second, the initialization part of the above three sensors includes IMU preintegration, odometer preintegration, and gyroscope bias calculation. It also includes the alignment of speed, gravity, and scale. Finally, a local BA (bundle adjustment) joint optimization and global graph optimization are established, so as to obtain more accurate positioning results.

Stereo Visual Odometry for Indoor Localization of Ship Model

Typically, ships are designed for open sea navigation and thus research of autonomous ships is mostly done for that particular area. This paper explores the possibility of using low-cost sensors for localization inside the small navigation area. The localization system is based on the technology used for developing autonomous cars. The main part of the system is visual odometry using stereo cameras fused with Inertial Measurement Unit (IMU) data coupled with Kalman and particle filters to get decimetre level accuracy inside a basin for different surface conditions. The visual odometry uses cropped frames for stereo cameras and Good features to track algorithm for extracting features to get depths for each feature that is used for estimation of ship model movement. Experimental results showed that the proposed system could localize itself within a decimetre accuracy implying that there is a real possibility for ships in using visual odometry for autonomous navigation on narrow waterways, which can have a significant impact on future transportation.

The First Attempt of SAR Visual-Inertial Odometry

This article proposes a novel synthetic aperture radar visual-inertial odometry (SAR-VIO) consisting of an SAR and an inertial measurement unit (IMU), which aims to enable the observation platform to complete successfully a continuous observation mission in the context of low-cost demand and lack of enough navigation information. First, we establish the observation models of the SAR in a continuous observation process based on the SAR frequency-domain imaging algorithm and the SAR time-domain imaging algorithm, respectively. With the preintegrated IMU data, we then propose a method for estimating the geographic locations of the matched targets in the SAR images and verify the condition and correctness of the method. The optimization of the track and the locations of the targets is achieved by bundle adjustment according to the minimum reprojection error criterion, and a sparse point-cloud map can be obtained. Finally, these methods and models are organized into a complete SAR-VIO framework, and the feasibility of the framework is verified through experiments.

PassengerFlows: A Correlation-Based Passenger Estimator in Automated Public Transport

Human mobility information is widely needed by many sectors in smart cities, especially for public transport. This work designs lightweight algorithms to perform stop detection , passenger flow tracking , and passenger estimation in an automated train. The on-board learning and detection algorithms are running on an edge node that is integrated with Wi-Fi sniffing technology, a GPS sensor, and an inertial measurement unit (IMU) sensor. The stop detection algorithm determines if the automated train is static at which train station based on GPS and IMU data. When the train is moving, the correlations between statistical properties extracted from Wi-Fi probes and the actual number of passengers change. Therefore, two algorithms, passenger flow tracking and passenger estimation, are designed to analyze passenger mobility. The passenger flow tracking algorithm analyzes the number of incoming and outgoing passengers in the train. The passenger estimation algorithm approximates the number of passengers inside the train based on a multi-dimensional regression model created by statistical properties extracted from different device brands. The designed prototype is deployed in an automated hanging train to conduct real-world experiments. The experimental results indicate that the proposed passenger flow tracking algorithm reduces the average errors of $62.5\%$ and $70\%$ compared against two existing clustering algorithms respectively. When the devices’ brands are split for creating a regression model, compared with two counting-based approaches, the proposed passenger estimation algorithm results in lower errors with a mean of 3.15 and a variance of 9.29.

Intraprocedural ergonomics of vascular surgeons

ObjectiveThe objective of this study was to estimate the ergonomic postural risk (EPR) for musculoskeletal posture of vascular surgeons performing open and endovascular procedure types and with various adjunctive equipment using wearable inertial measurement unit (IMU) sensors. The hypothesis was that EPR will increase with increased physical and mental demand as well as with procedural complexity. MethodsA prospective, observational study was conducted at a large, quaternary academic hospital located at two sites. Sixteen vascular surgeons (13 male) participated in the study. Participants completed a presurgery and postsurgery survey consisting of a body part discomfort scale and a modified NASA-Task Load Index. Participants wore IMU sensors on the head and upper body to measure EPR during open and endovascular procedures. ResultsVascular surgeons have increased EPR scores of the neck as measured by the IMUs and increased lower back pain when performing open surgery compared with non-open surgery (P < .05). Open procedures were rated as more physically demanding. The use of loupes resulted in increased EPR scores for the neck and torso (P < .05), and they were significantly associated with higher levels of lower back pain during procedures (P < .05) as well as with higher levels of physical demand (P < .05). The use of headlights also resulted in increased subjectively measured levels of physical demand and lower back pain. In comparing survey responses with IMU data, surveyed physical demand was strongly and significantly correlated with the neck (r = 0.61; P < .0001) and torso (r = 0.59; P < .0001) EPR scores. The use of lead aprons did not affect EPR or most surveyed measures of workload but resulted in significantly higher levels of distraction (P < .01). The data presented highlight the potential of using wearable sensors to measure the EPR of surgeons during vascular surgical procedures. ConclusionsVascular surgeons should be aware of EPR during the performance of their duties. Procedure type and surgical adjuncts can alter EPR significantly.

Gait features for discriminating between mobility-limiting musculoskeletal disorders: Lumbar spinal stenosis and knee osteoarthritis

BackgroundFunctional ambulation limitations are features of lumbar spinal stenosis (LSS) and knee osteoarthritis (OA). With numerous validated walking assessment protocols and a vast number of spatiotemporal gait parameters available from sensor-based assessment, there is a critical need for selection of appropriate test protocols and variables for research and clinical applications. Research questionIn patients with knee OA and LSS, what are the best sensor-derived gait parameters and the most suitable clinical walking test to discriminate between these patient populations and controls? MethodsWe collected foot-mounted inertial measurement unit (IMU) data during three walking tests (fast-paced walk test-FPWT, 6-min walk test– 6MWT, self-paced walk test – SPWT) for subjects with LSS, knee OA and matched controls (N = 10 for each group). Spatiotemporal gait characteristics were extracted and pairwise compared (Omega partial squared – ωp2) between patients and controls. ResultsWe found that normal paced walking tests (6MWT, SPWT) are better suited for distinguishing gait characteristics between patients and controls. Among the sensor-based gait parameters, stance and double support phase timing were identified as the best gait characteristics for the OA population discrimination, whereas foot flat ratio, gait speed, stride length and cadence were identified as the best gait characteristics for the LSS population discrimination. SignificanceThese findings provide guidance on the selection of sensor-derived gait parameters and clinical walking tests to detect alterations in mobility for people with LSS and knee OA.

Accurate recognition of lower limb ambulation mode based on surface electromyography and motion data using machine learning

Background and Objective The lower limb activity of recognition of the elderly, the weak, the disabled and the sick is an irreplaceable role in the caring of daily life. The main purpose of this study is to assess the feasibility of using the surface electromyography (sEMG) signal and inertial measurement units (IMUs) data to determine the optimal fusion features and classifier for the study of daily ambulation mode recognition.Methods We have carried out several steps of experiments to obtain and test the optimal combination of the sEMG data and the body motion data at the feature level and the most suitable machine learning classification algorithm. Firstly, the sEMG and IMUs signals of eighteen participants performing four different ambulatory activities have recorded using wearable sensors. Secondly, several features of the sEMG sensors and IMU data were extracted and tested by the Markov Random Field based Fisher–Markov feature selector. Finally, four ML classifiers with several feature combinations were estimated with sensitivity, precision and recognition accurate rate of ambulatory activity classification.Results The results of this work showed that all selected features were significantly statistical difference in four ambulation modes. The principal component analysis was used to reduce the dimension of selected sEMG features and IMU features to form a fusion feature input support vector machine classifier, which could predict ambulatory activities with good classification performance.Conclusions It is concluded that the results demonstrate the feasibility of the ML classification model, which could provide a more novel way to guarantee the recognition rate and effectiveness of monitor daily ambulatory activity.

Dual-sensor fusion based attitude holding of a fin-actuated robotic fish

In nature, the lateral line system (LLS) is a critical sensor organ of fish for rheotaxis in complex environments. Inspired by the LLS, numbers of artificial lateral line systems (ALLSs) have been designed to the fish-like robots for flow field perception, assisting the robots to be stable in the face of flow disturbances. However, almost all pressure sensor based ALLSs face the challenge of the low signal to noise ratio (SNR), resulting in inaccurate perception information. To solve this problem, this paper describes a dual-sensor fusion method by integrating the ALLSs with the inertial measurement unit (IMU), and shows the excellent performance by a higher precision and lower latency attitude holding of robotic fish. First, low-pass filtering is performed on ALLS data with low-SNR. Second, the ALLS data is mapped to the angle of attack based on an artificial neural network. Finally, a fusion perception method is established based on the time correlation between ALLS and IMU. To demonstrate the efficacy of our proposed method, we compare the result of attitude holding by three methods (dual-sensor fusion method, IMU based method, and ALLS based method). Furthermore, dual-sensor fusion method is tested at varied flow velocities and varied desired angles of attack, indicating that the algorithm can enable the robotic fish to perform dynamic movements in the incoming flow. This work provides a method for the attitude control of autonomous underwater vehicles (AUVs) by fusing the sensory data of ALLS and IMU, which is also applicable to other flow sensors and IMU.

Wearable Sensors Quantify Mobility in People With Lower Limb Amputation During Daily Life.

It is necessary to effectively assess functional mobility for appropriate prosthetic prescription and post-amputation rehabilitation. As part of this process, patients' ability for variable cadence and community ambulation are assessed in-clinic, often through visual assessments and without objective standards. The purpose of this study was to explore the clinical viability of using wearable sensors to characterize the functional mobility of people with lower limb amputation. We collected inertial measurement unit (IMU) and global positioning system (GPS) data over two weeks, from 17 individuals with lower limb amputation and 14 healthy non-amputee controls. We calculated stride-by-stride cadence, walking speed and stride lengths, along with whether they occurred in or out of the home. Self-selected walking speed was also assessed in the lab. Compared to the lab, both groups walked slower and with a lower cadence during their daily lives. There were no differences in cadence variability between groups or between strides taken in and out of the home. Both groups walked faster and with greater stride lengths away from the homes. The results suggest that functional capacity measured in the lab was not necessarily reflected in routine walking during daily life. The walking measures derived in this approach can be used to aid in the prosthetic prescription process or in the assessment of different interventions.

A Machine Learning Approach to Estimate Hip and Knee Joint Loading Using a Mobile Phone-Embedded IMU

Hip osteoarthritis patients exhibit changes in kinematics and kinetics that affect joint loading. Monitoring this load can provide valuable information to clinicians. For example, a patient's joint loading measured across different activities can be used to determine the amount of exercise that the patient needs to complete each day. Unfortunately, current methods for measuring joint loading require a lab environment which most clinicians do not have access to. This study explores employing machine learning to construct a model that can estimate joint loading based on sensor data obtained solely from a mobile phone. In order to learn such a model, we collected a dataset from 10 patients with hip osteoarthritis who performed multiple repetitions of nine different exercises. During each repetition, we simultaneously recorded 3D motion capture data, ground reaction force data, and the inertial measurement unit data from a mobile phone attached to the patient's hip. The 3D motion and ground reaction force data were used to compute the ground truth joint loading using musculoskeletal modeling. Our goal is to estimate the ground truth loading value using only the data captured by the sensors of the mobile phone. We propose a machine learning pipeline for learning such a model based on the recordings of a phone's accelerometer and gyroscope. When evaluated for an unseen patient, the proposed pipeline achieves a mean absolute error of 29% for the left hip and 36% for the right hip. While our approach is a step in the direction of using a minimal number of sensors to estimate joint loading outside the lab, developing a tool that is accurate enough to be applicable in a clinical context still remains an open challenge. It may be necessary to use sensors at more than one location in order to obtain better estimates.

Latency Compensated Visual-Inertial Odometry for Agile Autonomous Flight.

In visual-inertial odometry (VIO), inertial measurement unit (IMU) dead reckoning acts as the dynamic model for flight vehicles while camera vision extracts information about the surrounding environment and determines features or points of interest. With these sensors, the most widely used algorithm for estimating vehicle and feature states for VIO is an extended Kalman filter (EKF). The design of the standard EKF does not inherently allow for time offsets between the timestamps of the IMU and vision data. In fact, sensor-related delays that arise in various realistic conditions are at least partially unknown parameters. A lack of compensation for unknown parameters often leads to a serious impact on the accuracy of VIO systems and systems like them. To compensate for the uncertainties of the unknown time delays, this study incorporates parameter estimation into feature initialization and state estimation. Moreover, computing cross-covariance and estimating delays in online temporal calibration correct residual, Jacobian, and covariance. Results from flight dataset testing validate the improved accuracy of VIO employing latency compensated filtering frameworks. The insights and methods proposed here are ultimately useful in any estimation problem (e.g., multi-sensor fusion scenarios) where compensation for partially unknown time delays can enhance performance.

Crater Navigation System for Autonomous Precision Landing on the Moon

Interest in autonomous planetary precision landing missions has been increasing in the scientific and engineering community, and is likely to continue to do so for the foreseeable future. As an enabling technology in the context of lunar landing, DLR, German Aerospace Center has been developing a terrain absolute navigation system that matches craters detected in image data to globally available lunar crater maps. The proposed Crater Navigation (CNav) system is adaptive, comprising three different crater matching methods that are specifically tailored to different navigation conditions encountered during the vehicle descent, so that it may be used as a stand-alone navigation sensor that can be closely integrated with a lander guidance, navigation, and control system to enable reliable absolute navigation throughout the entire descent phase of a mission. As robustness is a vital aspect to mission success, the CNav system includes verification mechanisms that ensure high dependability of the resulting navigation solution. This robustness is verified separately for all of the three different matching techniques presented in this paper. Closed-loop performance of the matchers is demonstrated as well, both for simulated image data sets, as for navigation camera images acquired during the Chinese Chang’e-3 landing mission. Successful uninterrupted estimation of the entire Chang’e-3 kinematic vehicle state during the powered descent until a final altitude of 350 m above ground, with neither known camera calibration nor inertial measurement unit data available, showcases the potential of the CNav system.