Perception and Navigation in Autonomous Systems in the Era of Learning: A Survey

Autonomous systems possess the features of inferring their own state, understanding their surroundings, and performing autonomous navigation. With the applications of learning systems, like deep learning and reinforcement learning, the visual-based self-state estimation, environment perception, and navigation capabilities of autonomous systems have been efficiently addressed, and many new learning-based algorithms have surfaced with respect to autonomous visual perception and navigation. In this review, we focus on the applications of learning-based monocular approaches in ego-motion perception, environment perception, and navigation in autonomous systems, which is different from previous reviews that discussed traditional methods. First, we delineate the shortcomings of existing classical visual simultaneous localization and mapping (vSLAM) solutions, which demonstrate the necessity to integrate deep learning techniques. Second, we review the visual-based environmental perception and understanding methods based on deep learning, including deep learning-based monocular depth estimation, monocular ego-motion prediction, image enhancement, object detection, semantic segmentation, and their combinations with traditional vSLAM frameworks. Then, we focus on the visual navigation based on learning systems, mainly including reinforcement learning and deep reinforcement learning. Finally, we examine several challenges and promising directions discussed and concluded in related research of learning systems in the era of computer science and robotics.

Markerless Motion Tracking With Noisy Video and IMU Data

OBJECTIVE

Marker-based motion capture, considered the gold standard in human motion analysis, is expensive and requires trained personnel. Advances in inertial sensing and computer vision offer new opportunities to obtain research-grade assessments in clinics and natural environments. A challenge that discourages clinical adoption, however, is the need for careful sensor-to-body alignment, which slows the data collection process in clinics and is prone to errors when patients take the sensors home.

METHODS

We propose deep learning models to estimate human movement with noisy data from videos (VideoNet), inertial sensors (IMUNet), and a combination of the two (FusionNet), obviating the need for careful calibration. The video and inertial sensing data used to train the models were generated synthetically from a marker-based motion capture dataset of a broad range of activities and augmented to account for sensor-misplacement and camera-occlusion errors. The models were tested using real data that included walking, jogging, squatting, sit-to-stand, and other activities.

RESULTS

On calibrated data, IMUNet was as accurate as state-of-the-art models, while VideoNet and FusionNet reduced mean ± std root-mean-squared errors by 7.6 ± 5.4 ° and 5.9 ± 3.3 °, respectively. Importantly, all the newly proposed models were less sensitive to noise than existing approaches, reducing errors by up to 14.0 ± 5.3 ° for sensor-misplacement errors of up to 30.0 ± 13.7 ° and by up to 7.4 ± 5.5 ° for joint-center-estimation errors of up to 101.1 ± 11.2 mm, across joints.

CONCLUSION

These tools offer clinicians and patients the opportunity to estimate movement with research-grade accuracy, without the need for time-consuming calibration steps or the high costs associated with commercial products such as Theia3D or Xsens, helping democratize the diagnosis, prognosis, and treatment of neuromusculoskeletal conditions.

Unified Robot and Inertial Sensor Self-Calibration

Robots and inertial measurement units (IMUs) are typically calibrated independently. IMUs are placed in purpose-built, expensive automated test rigs. Robot poses are typically measured using highly accurate (and thus expensive) tracking systems. In this paper, we present a quick, easy, and inexpensive new approach to calibrate both simultaneously, simply by attaching the IMU anywhere on the robot's end effector and moving the robot continuously through space. Our approach provides a fast and inexpensive alternative to both robot and IMU calibration, without any external measurement systems. We accomplish this using continuous-time batch estimation, providing statistically optimal solutions. Under Gaussian assumptions, we show that this becomes a nonlinear least squares problem and analyze the structure of the associated Jacobian. Our methods are validated both numerically and experimentally and compared to standard individual robot and IMU calibration methods.

LiDAR-Stabilised GNSS-IMU Platform Pose Tracking

The requirement to estimate the six degree-of-freedom pose of a moving platform frequently arises in automation applications. It is common to estimate platform pose by the fusion of global navigation satellite systems (GNSS) measurements and translational acceleration and rotational rate measurements from an inertial measurement unit (IMU). This paper considers a specific situation where two GNSS receivers and one IMU are used and gives the full formulation of a Kalman filter-based estimator to do this. A limitation in using this sensor set is the difficulty of obtaining accurate estimates of the degree of freedom corresponding to rotation about the line passing through the two GNSS receiver antenna centres. The GNSS-aided IMU formulation is extended to incorporate LiDAR measurements in both known and unknown environments to stabilise this degree of freedom. The performance of the pose estimator is established by comparing expected LiDAR range measurements with actual range measurements. Distributions of the terrain point-to-model error are shown to improve from 0.27m mean error to 0.06m when the GNSS-aided IMU estimator is augmented with LiDAR measurements. This precision is marginally degraded to 0.14m when the pose estimator is operated in an a prior unknown environment.

Enhanced Attitude and Altitude Estimation for Indoor Autonomous UAVs

In recent years the use of Unmanned Aerial Vehicles (UAVs) has considerably grown in the civil sectors, due to their high flexibility of use. Currently, two important key points are making them more and more successful in the civil field, namely the decrease of production costs and the increase in navigation accuracy. In this paper, we propose a Kalman filtering-based sensor fusion algorithm, using a low cost navigation platform that contains an inertial measurement unit (IMU), five ultrasonic ranging sensors and an optical flow camera. The aim is to improve navigation in indoor or GPS-denied environments. A multi-rate version of the Extended Kalman Filter is considered to deal with the use of heterogeneous sensors with different sampling rates, and the presence of non-linearities in the model. The effectiveness of the proposed sensor platform is evaluated by means of numerical tests on the dynamic flight simulator of a quadrotor. Results show high precision and robustness of the attitude estimation algorithm, with a reduced computational cost, being ready to be implemented on low-cost platforms.

Pursuer Assignment and Control Strategies in Multi-Agent Pursuit-Evasion Under Uncertainties

We consider a pursuit-evasion problem with a heterogeneous team of multiple pursuers and multiple evaders. Although both the pursuers and the evaders are aware of each others' control and assignment strategies, they do not have exact information about the other type of agents' location or action. Using only noisy on-board sensors the pursuers (or evaders) make probabilistic estimation of positions of the evaders (or pursuers). Each type of agent use Markov localization to update the probability distribution of the other type. A search-based control strategy is developed for the pursuers that intrinsically takes the probability distribution of the evaders into account. Pursuers are assigned using an assignment algorithm that takes redundancy (i.e., an excess in the number of pursuers than the number of evaders) into account, such that the total or maximum estimated time to capture the evaders is minimized. In this respect we assume the pursuers to have clear advantage over the evaders. However, the objective of this work is to use assignment strategies that minimize the capture time. This assignment strategy is based on a modified Hungarian algorithm as well as a novel algorithm for determining assignment of redundant pursuers. The evaders, in order to effectively avoid the pursuers, predict the assignment based on their probabilistic knowledge of the pursuers and use a control strategy to actively move away from those pursues. Our experimental evaluation shows that the redundant assignment algorithm performs better than an alternative nearest-neighbor based assignment algorithm.

Localisation of Unmanned Underwater Vehicles (UUVs) in Complex and Confined Environments: A Review

The inspection of aquatic environments is a challenging activity, which is made more difficult if the environment is complex or confined, such as those that are found in nuclear storage facilities and accident sites, marinas and boatyards, liquid storage tanks, or flooded tunnels and sewers. Human inspections of these environments are often dangerous or infeasible, so remote inspection using unmanned underwater vehicles (UUVs) is used. Due to access restrictions and environmental limitations, such as low illumination levels, turbidity, and a lack of salient features, traditional localisation systems that have been developed for use in large bodies of water cannot be used. This means that UUV capabilities are severely restricted to manually controlled low-quality visual inspections, generating non-geospatially located data. The localisation of UUVs in these environments would enable the autonomous behaviour and the development of accurate maps. This article presents a review of the state-of-the-art in localisation technologies for these environments and identifies areas of future research to overcome the challenges posed.

Sensor-to-Segment Calibration Methodologies for Lower-Body Kinematic Analysis with Inertial Sensors: A Systematic Review

Kinematic analysis is indispensable to understanding and characterizing human locomotion. Thanks to the development of inertial sensors based on microelectronics systems, human kinematic analysis in an ecological environment is made possible. An important issue in human kinematic analyses with inertial sensors is the necessity of defining the orientation of the inertial sensor coordinate system relative to its underlying segment coordinate system, which is referred to sensor-to-segment calibration. Over the last decade, we have seen an increase of proposals for this purpose. The aim of this review is to highlight the different proposals made for lower-body segments. Three different databases were screened: PubMed, Science Direct and IEEE Xplore. One reviewer performed the selection of the different studies and data extraction. Fifty-five studies were included. Four different types of calibration method could be identified in the articles: the manual, static, functional, and anatomical methods. The mathematical approach to obtain the segment axis and the calibration evaluation were extracted from the selected articles. Given the number of propositions and the diversity of references used to evaluate the methods, it is difficult today to form a conclusion about the most suitable. To conclude, comparative studies are required to validate calibration methods in different circumstances.

Contact-aided invariant extended Kalman filtering for robot state estimation

Legged robots require knowledge of pose and velocity in order to maintain stability and execute walking paths. Current solutions either rely on vision data, which is susceptible to environmental and lighting conditions, or fusion of kinematic and contact data with measurements from an inertial measurement unit (IMU). In this work, we develop a contact-aided invariant extended Kalman filter (InEKF) using the theory of Lie groups and invariant observer design. This filter combines contact-inertial dynamics with forward kinematic corrections to estimate pose and velocity along with all current contact points. We show that the error dynamics follows a log-linear autonomous differential equation with several important consequences: (a) the observable state variables can be rendered convergent with a domain of attraction that is independent of the system’s trajectory; (b) unlike the standard EKF, neither the linearized error dynamics nor the linearized observation model depend on the current state estimate, which (c) leads to improved convergence properties and (d) a local observability matrix that is consistent with the underlying nonlinear system. Furthermore, we demonstrate how to include IMU biases, add/remove contacts, and formulate both world-centric and robo-centric versions. We compare the convergence of the proposed InEKF with the commonly used quaternion-based extended Kalman filter (EKF) through both simulations and experiments on a Cassie-series bipedal robot. Filter accuracy is analyzed using motion capture, while a LiDAR mapping experiment provides a practical use case. Overall, the developed contact-aided InEKF provides better performance in comparison with the quaternion-based EKF as a result of exploiting symmetries present in system.

Accelerometer-Based Gyroscope Drift Compensation Approach in a Dual-Axial Stabilization Platform

An accelerometer-based gyro drift compensation approach in a dual-axial stabilization platform is introduced in this paper. The stabilization platform consists of platform framework, drive motor, gyro and accelerometer module and contorl board. Gyro is an angular rate detecting element to achieve angular rate and rotation angle of the dynamic platform system. However, the platform system has an unstable factor because of the drift of gyro. The main contribution of this paper is to implement a convenient gyro drift compensation approach by using the accelerometer. In contrast to a kalman filtering method, this approach is simpler and practical due to the high-precision characteristic of the accelerometer. Data filtering algorithm and limit of threshold setting of total acceleration values are applied in this approach. The validity and feasibility of the proposed approach are evaluated by four tests under various conditions.

Fusion of High-Dynamic and Low-Drift Sensors Using Kalman Filters

In practice, a high-dynamic vibration sensor is often plagued by the problem of drift, which is caused by thermal effects. Conversely, low-drift sensors typically have a limited sample rate range. This paper presents a system combining different types of sensors to address general drift problems that occur in measurements for high-dynamic vibration signals. In this paper, the hardware structure and algorithms for fusing high-dynamic and low-drift sensors are described. The algorithms include a drift state estimation and a Kalman filter based on a linear prediction model. Key issues such as the dimension of the drift state vector, the order of the linear prediction model, are analyzed in the design of algorithm. The performance of the algorithm is illustrated by a simulation example and experiments. The simulation and experimental results show that the drift can be removed while the high-dynamic measuring ability is retained. A high-dynamic vibration measuring system with the frequency range starting from 0 Hz is achieved. Meanwhile, measurement noise was improved 9.3 dB through using the linear-prediction-based Kalman filter.

Development and Performance Evaluation of Planar Travel Distance Sensors for Mobile Robots in Sandy Terrain

A planar travel distance sensor (two-dimensional sensor) was developed for a mobile robot in sandy terrain. The sensor system uses an optical flow device integrated into a small module with a simple configuration. The system achieves a high sampling rate on the order of milliseconds as well as precise measurement on a sub-millimeter order. Its performance was evaluated experimentally for measurement accuracy and repeatability, velocity response, robustness at varied heights with respect to terrain, and terrain surface characteristics. The experimental results confirm that the two-dimensional sensor system is accurate, having an error of distance traveled of less than a few percent, and that it possesses a wide dynamic range for the robot’s traveling velocity. This paper also discusses the applicability of the two-dimensional sensor for practical scenarios on the basis of the experimental findings.

Evolutionary Metric-Learning-Based Recognition Algorithm for Online Isolated Persian/Arabic Characters, Reconstructed Using Inertial Pen Signals

The development of sensors with the microelectromechanical systems technology expedites the emergence of new tools for human-computer interaction, such as inertial pens. These pens, which are used as writing tools, do not depend on a specific embedded hardware, and thus, they are inexpensive. Most of the available inertial pen character recognition approaches use the low-level features of inertial signals. This paper introduces a Persian/Arabic handwriting character recognition system for inertial-sensor-equipped pens. First, the motion trajectory of the inertial pen is reconstructed to estimate the position signals by using the theory of inertial navigation systems. The position signals are then used to extract high-level geometrical features. A new metric learning technique is then adopted to enhance the accuracy of character classification. To this end, a characteristic function is calculated for each character using a genetic programming algorithm. These functions form a metric kernel classifying all the characters. The experimental results show that the performance of the proposed method is superior to that of one of the state-of-the-art works in terms of recognizing Persian/Arabic handwriting characters.

Angular velocity estimation based on star vector with improved current statistical model Kalman filter

Angular velocity information is a requisite for a spacecraft guidance, navigation, and control system. In this paper, an approach for angular velocity estimation based merely on star vector measurement with an improved current statistical model Kalman filter is proposed. High-precision angular velocity estimation can be achieved under dynamic conditions. The amount of calculation is also reduced compared to a Kalman filter. Different trajectories are simulated to test this approach, and experiments with real starry sky observation are implemented for further confirmation. The estimation accuracy is proved to be better than 10^-4  rad/s under various conditions. Both the simulation and the experiment demonstrate that the described approach is effective and shows an excellent performance under both static and dynamic conditions.

Toward Reliable Off Road Autonomous Vehicles Operating in Challenging Environments

The DARPA PerceptOR program has implemented a rigorous evaluative test program which fosters the development of field relevant outdoor mobile robots. Autonomous ground vehicles were deployed on diverse test courses throughout the USA and quantitatively evaluated on such factors as autonomy level, waypoint acquisition, failure rate, speed, and communications bandwidth. Our efforts over the three year program have produced new approaches in planning, perception, localization, and control which have been driven by the quest for reliable operation in challenging environments. This paper focuses on some of the most unique aspects of the systems developed by the CMU PerceptOR team, the lessons learned during the effort, and the most immediate challenges that remain to be addressed.

The Estimation Theoretic Sensor Bias Correction Problem in Map Aided Localization

Simultaneous Localization and Map Building (SLAM) and Map Aided Localization (MAL) are very effective techniques employed extensively in robot navigation tasks. However, biases and drifts in both exteroceptive and proprioceptive sensors adversely impair correct localization (in MAL) and also impair map building (in SLAM). More specifically, accumulated errors as a result of biases in the sensors cause the algorithms to diverge and produce inconsistent and inaccurate results. Although offline calibration of these sensors can reduce the effects to some extent, the process results in longer setup and processing times. Moreover, during operation, the sensors’ calibration may often be subject to changes or drifts requiring regular resetting and initialization. A convenient, appropriate and effective approach to overcome problems associated with biases in sensors has been to explicitly model and estimate the bias parameters concurrently with the vehicle state online using an augmented state space approach. This paper investigates the properties of the concurrent bias estimation in MAL using an augmented, estimation theoretic state space approach for the localization of a large class of mobile robots, consisting of autonomous ground vehicles. This involves a rigorous theoretical study of the issues of observability and convergence, their interrelations and effects on the algorithm’s performance. This paper shows analytically that if sensor biases are estimated jointly with the vehicle pose in a MAL framework: 1) The uncertainties of the estimated errors in the bias parameters of both proprioceptive and exteroceptive sensors diminish in each update. 2) A derived lower bound is reached in each of these estimates. 3) The rate of convergence to this lower bound is also derived. 4) Although often neglected in the literature, observability is a major issue. From the analysis it is derived that in order to guarantee observability in MAL with bias estimation, it is necessary to observe simultaneously at least two distinct landmarks, which are not on a straight line with the vehicle position. Extensive simulations are provided to illustrate the theoretical results established for the general case of nonlinear dynamics and slowly varying sensor biases. The results are further exemplified and verified experimentally using a sophisticated MAL algorithm, utilizing a low cost inertial navigation sensor suite.

On-Line Stability Margin and Attitude Estimation for Dynamic Articulating Mobile Robots

Stability is an important concern for vehicles which move heavy loads, accelerate or brake aggressively, turn at speed, or operate on sloped terrain. In many cases, vehicles face more than one of these challenges simultaneously. Some are obliged to execute these maneuvers when their high centers of gravity leave them particularly vulnerable to tipover or rollover. A methodology is presented to estimate proximity to tipover for autonomous field robots that must be productive, effective, and self-reliant under such challenging circumstances. While the physical principles governing the computation of stability margin have been known for some time, the realization of these principles in practice raises issues which are at once similar to those of attitude estimation while contrasting heavily with inertial guidance. The problem of stability margin estimation is posed in the fairly general case of accelerated articulating motion over rough terrain. Compatibility with and distinctions from attitude estimation lead to a proposed integrated solution to both problems based on the fusion of inertial, articulation, and terrain relative velocity sensing in an optimal estimation framework. An implementation of a device targeted to an industrial lift truck is presented.

System Design of a Quadrupedal Galloping Machine

In this paper we present the system design of a machine that we have constructed to study a quadrupedal gallop gait. The gallop gait is the preferred high-speed gait of most cursorial quadrupeds. To gallop, an animal must generate ballistic trajectories with characteristic strong impacts, coordinate leg movements with asymmetric footfall phasing, and effectively use compliant members, all the while maintaining dynamic stability. In this paper we seek to further understand the primary biological features necessary for galloping by building and testing a robotic quadruped similar in size to a large goat or antelope. These features include high-speed actuation, energy storage, on-line learning control, and high-performance attitude sensing. Because body dynamics are primarily influenced by the impulses delivered by the legs, the successful design and control of single leg energetics is a major focus of this work. The leg stores energy during flight by adding tension to a spring acting across an articulated knee. During stance, the spring energy is quickly released using a novel capstan design. As a precursor to quadruped control, two intelligent strategies have been developed for verification on a one-legged system. The Levenberg-Marquardt on-line learning method is applied to a simple heuristic controller and provides good control over height and forward velocity. Direct adaptive fuzzy control, which requires no system modeling but is more computationally expensive, exhibits better response. Using these techniques we have been successful in operating one leg at speeds necessary for a dynamic gallop of a machine of this scale. Another necessary component of quadruped locomotion is high-resolution and high-bandwidth attitude sensing. The large ground impact accelerations, which cause problems for any single traditional sensor, are overcome through the use of an inertial sensing approach using updates from optical sensors and vehicle kinematics.

Jacobian-Based Iterative Method for Magnetic Localization in Robotic Capsule Endoscopy

The purpose of this study is to validate a Jacobian-based iterative method for real-time localization of magnetically controlled endoscopic capsules. The proposed approach applies finite-element solutions to the magnetic field problem and least-squares interpolations to obtain closed-form and fast estimates of the magnetic field. By defining a closed-form expression for the Jacobian of the magnetic field relative to changes in the capsule pose, we are able to obtain an iterative localization at a faster computational time when compared with prior works, without suffering from the inaccuracies stemming from dipole assumptions. This new algorithm can be used in conjunction with an absolute localization technique that provides initialization values at a slower refresh rate. The proposed approach was assessed via simulation and experimental trials, adopting a wireless capsule equipped with a permanent magnet, six magnetic field sensors, and an inertial measurement unit. The overall refresh rate, including sensor data acquisition and wireless communication was 7 ms, thus enabling closed-loop control strategies for magnetic manipulation running faster than 100 Hz. The average localization error, expressed in cylindrical coordinates was below 7 mm in both the radial and axial components and 5° in the azimuthal component. The average error for the capsule orientation angles, obtained by fusing gyroscope and inclinometer measurements, was below 5°.

Particle Filter with Novel Nonlinear Error Model for Miniature Gyroscope-Based Measurement While Drilling Navigation

The derivation of a conventional error model for the miniature gyroscope-based measurement while drilling (MGWD) system is based on the assumption that the errors of attitude are small enough so that the direction cosine matrix (DCM) can be approximated or simplified by the errors of small-angle attitude. However, the simplification of the DCM would introduce errors to the navigation solutions of the MGWD system if the initial alignment cannot provide precise attitude, especially for the low-cost microelectromechanical system (MEMS) sensors operated in harsh multilateral horizontal downhole drilling environments. This paper proposes a novel nonlinear error model (NNEM) by the introduction of the error of DCM, and the NNEM can reduce the propagated errors under large-angle attitude error conditions. The zero velocity and zero position are the reference points and the innovations in the states estimation of particle filter (PF) and Kalman filter (KF). The experimental results illustrate that the performance of PF is better than KF and the PF with NNEM can effectively restrain the errors of system states, especially for the azimuth, velocity, and height in the quasi-stationary condition.

Integrating Sensors into a Marine Drone for Bathymetric 3D Surveys in Shallow Waters

This paper demonstrates that accurate data concerning bathymetry as well as environmental conditions in shallow waters can be acquired using sensors that are integrated into the same marine vehicle. An open prototype of an unmanned surface vessel (USV) named MicroVeGA is described. The focus is on the main instruments installed on-board: a differential Global Position System (GPS) system and single beam echo sounder; inertial platform for attitude control; ultrasound obstacle-detection system with temperature control system; emerged and submerged video acquisition system. The results of two cases study are presented, both concerning areas (Sorrento Marina Grande and Marechiaro Harbour, both in the Gulf of Naples) characterized by a coastal physiography that impedes the execution of a bathymetric survey with traditional boats. In addition, those areas are critical because of the presence of submerged archaeological remains that produce rapid changes in depth values. The experiments confirm that the integration of the sensors improves the instruments’ performance and survey accuracy.

Keeping a Good Attitude: A Quaternion-Based Orientation Filter for IMUs and MARGs

Orientation estimation using low cost sensors is an important task for Micro Aerial Vehicles (MAVs) in order to obtain a good feedback for the attitude controller. The challenges come from the low accuracy and noisy data of the MicroElectroMechanical System (MEMS) technology, which is the basis of modern, miniaturized inertial sensors. In this article, we describe a novel approach to obtain an estimation of the orientation in quaternion form from the observations of gravity and magnetic field. Our approach provides a quaternion estimation as the algebraic solution of a system from inertial/magnetic observations. We separate the problems of finding the "tilt" quaternion and the heading quaternion in two sub-parts of our system. This procedure is the key for avoiding the impact of the magnetic disturbances on the roll and pitch components of the orientation when the sensor is surrounded by unwanted magnetic flux. We demonstrate the validity of our method first analytically and then empirically using simulated data. We propose a novel complementary filter for MAVs that fuses together gyroscope data with accelerometer and magnetic field readings. The correction part of the filter is based on the method described above and works for both IMU (Inertial Measurement Unit) and MARG (Magnetic, Angular Rate, and Gravity) sensors. We evaluate the effectiveness of the filter and show that it significantly outperforms other common methods, using publicly available datasets with ground-truth data recorded during a real flight experiment of a micro quadrotor helicopter.

Time- and computation-efficient calibration of MEMS 3D accelerometers and gyroscopes

We propose calibration methods for microelectromechanical system (MEMS) 3D accelerometers and gyroscopes that are efficient in terms of time and computational complexity. The calibration process for both sensors is simple, does not require additional expensive equipment, and can be performed in the field before or between motion measurements. The methods rely on a small number of defined calibration measurements that are used to obtain the values of 12 calibration parameters. This process enables the static compensation of sensor inaccuracies. The values detected by the 3D sensor are interpreted using a generalized 3D sensor model. The model assumes that the values detected by the sensor are equal to the projections of the measured value on the sensor sensitivity axes. Although this finding is trivial for 3D accelerometers, its validity for 3D gyroscopes is not immediately apparent; thus, this paper elaborates on this latter topic. For an example sensor device, calibration parameters were established using calibration measurements of approximately 1.5 min in duration for the 3D accelerometer and 2.5 min in duration for the 3D gyroscope. Correction of each detected 3D value using the established calibration parameters in further measurements requires only nine addition and nine multiplication operations.

Estimation of IMU and MARG orientation using a gradient descent algorithm

This paper presents a novel orientation algorithm designed to support a computationally efficient, wearable inertial human motion tracking system for rehabilitation applications. It is applicable to inertial measurement units (IMUs) consisting of tri-axis gyroscopes and accelerometers, and magnetic angular rate and gravity (MARG) sensor arrays that also include tri-axis magnetometers. The MARG implementation incorporates magnetic distortion compensation. The algorithm uses a quaternion representation, allowing accelerometer and magnetometer data to be used in an analytically derived and optimised gradient descent algorithm to compute the direction of the gyroscope measurement error as a quaternion derivative. Performance has been evaluated empirically using a commercially available orientation sensor and reference measurements of orientation obtained using an optical measurement system. Performance was also benchmarked against the propriety Kalman-based algorithm of orientation sensor. Results indicate the algorithm achieves levels of accuracy matching that of the Kalman based algorithm; < 0.8° static RMS error, < 1.7° dynamic RMS error. The implications of the low computational load and ability to operate at small sampling rates significantly reduces the hardware and power necessary for wearable inertial movement tracking, enabling the creation of lightweight, inexpensive systems capable of functioning for extended periods of time.

An Unscented Rauch-Tung-Striebel Smoother for a Vehicle Localization Problem

The Unscented Kalman Filter (UKF) has become relatively a new technique used in a number of nonlinear estimation problems to overcome the limitation of Taylor series linearization. It uses a deterministic sampling approach known as sigma points to propagate nonlinear systems and has been discussed in many literature. However, a nonlinear smoothing problem has received less attention than the filtering problem. Therefore, in this article an unscented smoother based on Rauch-Tung-Striebel formis examined for discretetime dynamic systems. It has advantages available in unscented transformation over approximation by Taylor expansion as well as its benefit in derivative free. To show the effectiveness of the proposed method, the unscented smoother is implemented and evaluated through a vehicle localization problem.

State derivation of a 12-axis gyroscope-free inertial measurement unit

The derivation of linear acceleration, angular acceleration, and angular velocity states from a 12-axis gyroscope-free inertial measurement unit that utilizes four 3-axis accelerometer measurements at four distinct locations is reported. Particularly, a new algorithm which derives the angular velocity from its quadratic form and derivative form based on the context-based interacting multiple model is demonstrated. The performance of the system was evaluated under arbitrary 3-dimensional motion.

3D simultaneous localization and mapping using IMU and its observability analysis

This paper investigates 3-dimensional (3D) Simultaneous Localization and Mapping (SLAM) and the corresponding observability analysis by fusing data from landmark sensors and a strap-down Inertial Measurement Unit (IMU) in an adaptive Kalman filter (KF). In addition to the vehicle's states and landmark positions, the self-tuning filter estimates the IMU calibration parameters as well as the covariance of the measurement noise. The discrete-time covariance matrix of the process noise, the state transition matrix and the observation sensitivity matrix are derived in closed form, making it suitable for real-time implementation. Examination of the observability of the 3D SLAM system leads to the the conclusion that the system remains observable, provided that at least three known landmarks, which are not placed in a straight line, are observed.

C-arm rotation encoding with accelerometers

PURPOSE

Fluoroscopic C-arms are being incorporated in computer-assisted interventions in increasing number. For these applications to work, the relative poses of imaging must be known. To find the pose, tracking methods such as optical cameras, electromagnetic trackers, and radiographic fiducials have been used-all hampered by significant shortcomings.

METHODS

We propose to recover the rotational pose of the C-arm using the angle-sensing ability of accelerometers, by exploiting the capability of the accelerometer to measure tilt angles. By affixing the accelerometer to a C-arm, the accelerometer tracks the C-arm pose during rotations of the C-arm. To demonstrate this concept, a C-arm analogue was constructed with a webcam device affixed to the C-arm model to mimic X-ray imaging. Then, measuring the offset between the accelerometer angle readings to the webcam pose angle, an angle correction equation (ACE) was created to properly tracking the C-arm rotational pose.

EXPERIMENTS AND RESULTS

Several tests were performed on the webcam C-arm model using the ACEs to tracking the primary and secondary angle rotations of the model. We evaluated the capability of linear and polynomial ACEs to tracking the webcam C-arm pose angle for different rotational scenarios. The test results showed that the accelerometer could track the pose of the webcam C-arm model with an accuracy of less than 1.0 degree.

CONCLUSION

The accelerometer was successful in sensing the C-arm's rotation with clinically adequate accuracy in the C-arm webcam model.

Classifying human leg motions with uniaxial piezoelectric gyroscopes

This paper provides a comparative study on the different techniques of classifying human leg motions that are performed using two low-cost uniaxial piezoelectric gyroscopes worn on the leg. A number of feature sets, extracted from the raw inertial sensor data in different ways, are used in the classification process. The classification techniques implemented and compared in this study are: Bayesian decision making (BDM), a rule-based algorithm (RBA) or decision tree, least-squares method (LSM), k-nearest neighbor algorithm (k-NN), dynamic time warping (DTW), support vector machines (SVM), and artificial neural networks (ANN). A performance comparison of these classification techniques is provided in terms of their correct differentiation rates, confusion matrices, computational cost, and training and storage requirements. Three different cross-validation techniques are employed to validate the classifiers. The results indicate that BDM, in general, results in the highest correct classification rate with relatively small computational cost.

Estimating Displacement of Periodic Motion With Inertial Sensors

Inertial sensors, like accelerometers and gyroscopes, are rarely used by themselves to measure displacement. Accuracy of inertial sensors is greatly handicapped by the notorious integration drift, which arises due to numerical integration of the sensors zero bias error. A solution is proposed in this paper to provide drift free estimation of displacement from inertial sensors.

Minimal-Drift Heading Measurement using a MEMS Gyro for Indoor Mobile Robots

To meet the challenges of making low-cost MEMS yaw rate gyros for the precise self-localization of indoor mobile robots, this paper examines a practical and effective method of minimizing drift on the heading angle that relies solely on integration of rate signals from a gyro. The main idea of the proposed approach is consists of two parts; 1) self-identification of calibration coefficients that affects long-term performance, and 2) threshold filter to reject the broadband noise component that affects short-term performance. Experimental results with the proposed phased method applied to Epson XV3500 gyro demonstrate that it effectively yields minimal drift heading angle measurements getting over major error sources in the MEMS gyro output.

Soil Parameter Identification and Driving Force Prediction for Wheel-Terrain Interaction

This paper considers wheeled vehicles traversing unknown terrain, and proposes an approach for identifying the unknown soil parameters required for vehicle driving force prediction (drawbar pull prediction). The predicted drawbar pull can potentially be employed for traversability prediction, traction control, and trajectory following which, in turn, improve overall performance of off-road wheeled vehicles. The proposed algorithm uses an approximated form of the wheel-terrain interaction model and the Generalized Newton Raphson method to identify terrain parameters in real-time. With few measurements of wheel slip, i, vehicle sinkage, z, and drawbar pull, DP, samples, the algorithm is capable of identifying all the soil parameters required to predict vehicle driving forces over an entire range of wheel slip. The algorithm is validated with experimental data from a wheel-terrain interaction test rig. The identified soil parameters are used to predict the drawbar pull with good accuracy. The technique presented in this paper can be applied to any vehicle with rigid wheels or deformable wheels with relatively high inflation pressure, to predict driving forces in unknown environments.

Heads up and camera down: a vision-based tracking modality for mobile mixed reality

Anywhere Augmentation pursues the goal of lowering the initial investment of time and money necessary to participate in mixed reality work, bridging the gap between researchers in the field and regular computer users. Our paper contributes to this goal by introducing the GroundCam, a cheap tracking modality with no significant setup necessary. By itself, the GroundCam provides high frequency, high resolution relative position information similar to an inertial navigation system, but with significantly less drift. We present the design and implementation of the GroundCam, analyze the impact of several design and run-time factors on tracking accuracy, and consider the implications of extending our GroundCam to different hardware configurations. Motivated by the performance analysis, we developed a hybrid tracker that couples the GroundCam with a wide area tracking modality via a complementary Kalman filter, resulting in a powerful base for indoor and outdoor mobile mixed reality work. To conclude, the performance of the hybrid tracker and its utility within mixed reality applications is discussed.

Development of a real-time three-dimensional spinal motion measurement system for clinical practice

This paper presents an inertial based sensing system for real-time three-dimensional measurement of human spinal motion, in a portable and non-invasive manner. Applications of the proposed system range from diagnosis of spine injury to postural monitoring, on-field as well as in the lab setting. The system is comprised of three inertial measurement sensors, respectively attached and calibrated to the head, torso and hips, based on the subject's anatomical planes. Sensor output is transformed into meaningful clinical parameters of rotation (twist), flexion-extension and lateral bending of each body segment, with respect to calibrated global reference space. Modeling the spine as a compound flexible pole model allows dynamic measurement of three-dimensional spine motion, which can be animated and monitored in real-time using our interactive GUI. The accuracy of the proposed sensing system has been verified with subject trials using a VICON optical motion measurement system. Experimental results indicate an error of less than 3.1 degrees in segment orientation tracking.