Average Position Error Research Articles

Light-Emitting Diode-Based Optical Localization of a Robot in Continuous Motion Using Dynamic Prediction

Abstract For mobile robots, localization is essential for navigation and spatial correlation of its collected data. However, localization in Global Positioning System-denied environments such as underwater has been challenging. Light-emitting diode (LED)-based optical localization has been proposed in the literature, where the bearing angles extracted from the line-of-sight of the robot viewed from a pair of base nodes (also known as beacon nodes) are used to triangulate the position of the robot. The state-of-the-art in this approach uses a stop-and-go motion for the robot in order to ensure an accurate position measurement, which severely limits the mobility of the robot. This work presents an LED-based optical localization scheme for a mobile robot undergoing continuous motion, despite the two angles in each measurement cycle being captured at different locations of the robot. In particular, the bearing angle measurements are captured by the robot one at a time and are properly correlated with respect to the base nodes by utilizing the velocity prediction from Kalman filtering. The proposed system is evaluated in simulation and experiments, with its performance compared to the traditional state-of-the-art approach where the two angle measurements in each cycle are used directly to compute the position of the robot. In particular, the experimental results show that the average position and velocity estimation errors are reduced by 55% and 38%, respectively, when comparing the proposed method to the state-of-the-art.

Single NV centers array preparation and static magnetic field detection.

To solve the problem of static magnetic field detection accuracy and consistency, we prepared an array of single NV centers for static magnetic field vector and gradient detection using the femtosecond laser direct writing method. The prepared single NV centers are characterized by fewer impurity defects and good stress uniformity, with an average spatial positioning error of only 0.2 µm. This array of single NV centers can achieve high accuracy magnetic field vector and gradient measurement with GBZ≈-0.047 µT/µm in the Z-axis. This result provides a new idea for large-range, high-precision magnetic field vector and gradient measurements.

Application of Initial Bias Estimation Method for Inertial Navigation System (INS)/Doppler Velocity Log (DVL) and INS/DVL/Gyrocompass Using Micro-Electro-Mechanical System Sensors.

This article proposes a novel initial bias estimation method using a trajectory generator (TG). The accuracy of attitude and position estimation in navigation after using the inertial navigation system/Doppler velocity log (INS/DVL) and INS/DVL/gyrocompass (IDG) for 1 h were evaluated, and the results were compared to those obtained using the conventional Kalman filter (KF) estimation method. The probability of a horizontal position error < 1852 m (1 nautical mile) with a bias interval > 400 s was 100% and 9% for the TG and KF, respectively. In addition, the IDG average horizontal position errors over 1 h were 493 m and 507 m for the TG and KF, respectively. Moreover, the amount of variation was 2 m and 27 m for the TG and the KF, respectively. Thus, the proposed method is effective for initial bias estimation of INS/DVL and IDG using micro-electro-mechanical system sensors on a constantly moving vessel.

The Effect of Force Ratio Multiplier on A Control System for Surfing Problem Simulation

A surfing problem is a control problem of the surfing board to maintain its position on top of an ocean wave as long as possible. There are some physical and mathematical problems regarding a surfing problem that have not yet been solved. One of them is on translating a target inclination problem from an ordinary differential equation (ODE) control system to the inclination control system via the distribution of a surfer’s weight. To move the surfing board swiftly, a correct value of the multiplier, which is notated by σ, is needed on the weight distribution system. In this work, an investigation has been done on the effect of the multiplier in an attempt to help moves the surfing board fulfils the target inclination angle needed by using a smoothed particle hydrodynamics (SPH) simulation. The result from this work shows that the best value for the multiplier is σ=10 that gives the smallest average positional error for some variations of a given target position. This work gives a contribution on an attempt to model a surfer’s phenomenon mathematically.

An Image Reconstruction Algorithm for Electrical Impedance Tomography Using Measurement Estimation of Virtual Electrodes

For biomedical imaging in mini-scale, the inverse problem of electrical impedance tomography (EIT) is severely ill-conditioned due to the number of electrodes is very limited. In this paper, a novel difference imaging algorithm for 2D-EIT using measurement estimation of virtual electrodes is proposed. The proposed reconstruction algorithm breaks though the limitation of micro EIT sensor’s structure and tackles the problem of low spatial resolution in mini-scale by introducing virtual electrodes. The inverse problem of EIT is decomposed into two separately tasks: estimation potentials of virtual electrodes with proper priors, determination of the conductivity distribution using both virtual and real electrodes. It is formulated as a possibility model using virtual electrodes’ potentials as latent variables. Real electrodes’ potentials are regarded as the observable variables. Conductivity distribution is predicted by a maximum likelihood of the possibility model by using EM algorithm. The proposed algorithm is verified by both simulation and experiment. In experiment, medaka fish embryo which is about 1mm in diameter is reconstructed. Comparing with Tikhonov and NOSER regularization method, average ICC is improved 11.54% from 0.7620 to 0.8499 in simulation and improved 5.79% from 0.6978 to 0.7382 in experiment. The algorithm not only performs well in conductivity reconstruction, but also in shape reconstruction. Average position error is decreased 47.57%, average shape error is decreased 24.44% in experiment. It is very suitable for image reconstruction in mini-scale.

An Improved Mixture Density Network for 3D Human Pose Estimation with Ordinal Ranking

Estimating accurate 3D human poses from 2D images remains a challenge due to the lack of explicit depth information in 2D data. This paper proposes an improved mixture density network for 3D human pose estimation called the Locally Connected Mixture Density Network (LCMDN). Instead of conducting direct coordinate regression or providing unimodal estimates per joint, our approach predicts multiple possible hypotheses by the Mixture Density Network (MDN). Our network can be divided into two steps: the 2D joint points are estimated from the input images first; then, the information of human joints correlation is extracted by a feature extractor. After the human pose feature is extracted, multiple pose hypotheses are generated via the hypotheses generator. In addition, to make better use of the relationship between human joints, we introduce the Locally Connected Network (LCN) as a generic formulation to replace the traditional Fully Connected Network (FCN), which is applied to a feature extraction module. Finally, to select the most appropriate 3D pose result, a 3D pose selector based on the ordinal ranking of joints is adopted to score the predicted pose. The LCMDN improves the representation capability and robustness of the original MDN method notably. Experiments are conducted on the Human3.6M and MPII dataset. The average Mean Per Joint Position Error (MPJPE) of our proposed LCMDN reaches 50 mm on the Human3.6M dataset, which is on par or better than the state-of-the-art works. The qualitative results on the MPII dataset show that our network has a strong generalization ability.

Positioning performance of the Neustrelitz total electron content model driven by Galileo Az coefficients

To aid single-frequency GNSS users, the Neustrelitz Total Electron Content Model (NTCM) has been proposed as an adequate solution to mitigate propagation errors besides the GPS Klobuchar and Galileo NeQuick models. Using the three effective ionization coefficients broadcast in the Galileo navigation message as driver parameters, the version NTCM-GlAzpar, in general, performs equal to or better than NeQuickG when compared to the total electron content domain. In this work, we performed a global statistical validation of the NTCM-GlAzpar model in the position domain by comparing its results with the results of the Klobuchar and NeQuickG models for the first time. For this purpose, we used the GNSS analysis tool gLAB and its customization capabilities in the Standard Point Positioning mode. The data used for model validation correspond to a one-month period of perturbed solar and geomagnetic activity (December 2014) and another one-month period of quiet conditions (December 2019). The data have a worldwide coverage with up to 73 IGS stations. The statistical analysis of the hourly average 3D position error shows that the root mean squared (RMS) values of the Klobuchar model are 6.71 and 2.75 m for the perturbed and quiet conditions, respectively, whereas the NeQuickG model has RMS values of 4.61 and 2.35 m. In comparison, the corresponding RMS values of 4.36 and 2.32 m of the NTCM-GlAzpar model confirm its better positioning performance for both periods. However, we identify also that the performance of NTCM-GlAzpar slightly worsens toward higher latitudes and at night local time. Simple software adaptations and a low computational cost make NTCM-GlAzpar an alternative practicable algorithm to improve the accuracy of GNSS single-frequency applications.

A YOLOv3-based computer vision system for identification of tea buds and the picking point

Famous tea industry which need to harvest tea buds has great economic benefits. However, the harvesting is time-consuming and labor-intensive, especially with the shortage of labor currently, an intelligent tea bud picking robot is urgently needed. The vision system is a precursor to the development of a tea bud picking robot. To resolve such issues, we applied robotics and deep learning technologies to develop a computer vision system for intelligent picking of tea buds. The system was designed to recognize tea buds and extract their picking points. A method for locating the picking points was proposed based on a combination of YOLO-v3 algorithm, semantic segmentation algorithm, skeleton extraction and minimum bounding rectangle. An intelligent tea end-effector based on Personal Computer and microcontroller collaborative control was designed to solve the picking problem like complex shading and easy breakage. Thus, the picking rate of the overall system was improved. Based on Openmv smart camera embedded mobilenet_v2 algorithm as the visual model of the classification device, so that the quality of tea buds was preliminatively classified. Finally, the effects of different shooting angles and shooting methods as well as the accuracy of target detection and semantic segmentation algorithms on the extraction of tea bud picking points were investigated. The results show that the average accuracy of YOLO-v3 for identification of tea buds is 71.96% and the average horizontal positioning error of the robotic arm is 2.4 mm. Also, the average depth positioning error is 4.2 mm and the accuracy of tea bud picking point extraction is 83%. After the test, the successful picking rate of tea buds is 80% by this computer vision system of robot. The results of this study is potential to develop a machine-based tea picking system for industry and would contribute to the development of precision agriculture.

Tropical Cyclone Track Prediction with an Encoding-to-Forecasting Deep Learning Model

Abstract Recently, with the accumulation of remote sensing data, the traditional tropical cyclone (TC) track prediction methods (e.g., dynamic methods and statistical methods) have limitations in prediction efficiency and accuracy when dealing with a large amount of data. However, deep learning methods begin to show their advantages to capture the complex spatiotemporal features in high-dimensional data. The task of TC track prediction based on remote sensing images can be formulated as a spatiotemporal sequence-to-sequence problem. Therefore, a novel encoding-to-forecasting model with convolutional long short-term memory (ConvLSTM) and spatial attention network (SAN-EFSModel) was proposed to predict TC tracks in this paper, which can fully extract the long-term spatial and temporal features. The proposed model was evaluated on the real remote sensing images in the western North Pacific Ocean to forecast 24-h TC tracks. Compared with ECMWF-HRES model and NCEP-GFS model, the proposed method has a better prediction accuracy in the testing set with an average position error about 30 km less. Compared with the deep learning methods, the proposed method also has the best performance. Significance Statement Tropical cyclones have a great impact on human life and natural environment due to their high frequency of occurrence, heavy degree of harm, wide impact range, and long disaster chain. In this article, we propose an encoding-to-forecasting model with convolutional long short-term memory (ConvLSTM) and spatial attention network (SAN-EFSModel) to predict tropical cyclone tracks, in which the SAN module and the convolution module of ConvLSTM have the ability to extract spatial features, and the long short-term memory (LSTM) module of ConvLSTM can extract temporal features from remote sensing images and tropical cyclone tracks. The results show that the average position error (APE) of our proposed method has about 30% improvement in a 24-h forecast compared with prevailing numerical weather prediction models.

Pedestrian Inertial Navigation with Building Floor Plans for Indoor Environments via Particle Filter

&lt;p&gt;With the rapid development of smart city, the indoor positioning services became more and more important. During the existing solutions, inertial measurement unit (IMU) with pedestrian dead reckoning (PDR) was a promising scheme since they did not require external equipment in the environment. However, the orientation drift of low-cost IMU limited their application in practical. To address this problem, a zero-velocity update (ZUPT) framework included Kalman filter and particle filter is designed based on the foot-based low-cost IMU and digital floor plan to provide the service of personal navigation. In the designed Smoothing for ZUPT-aided INSs framework, the Kalman filter is used to estimate the position and attitude by zero velocity correction technique. Then, the particle filter is used to improve the localization and heading accuracy by map matching. The position estimation presented in this study achieves an average position error of 1.16 m. The experimental results show that the designed framework can solve the personal navigation problem in the case of building plan information assistance and help improve the accuracy and reliability of continuous position determination of personal navigation systems effectively. &lt;/p&gt; &lt;p&gt;&amp;nbsp;&lt;/p&gt;

LIDAR-BASED INITIAL GLOBAL LOCALIZATION USING IMPERFECT ARCHITECTURAL SKELETON INFORMATION

Abstract. Initial global localization of a mobile robotic platform is the foundation for its navigation and mapping, especially when the platform enters into unknown environments. In GNSS-denied indoor scenes, LiDAR is widely used for robot localization, especially in indoor scenes with poor lighting. In most existing LiDAR-based initial global localization methods, it is necessary to build the point cloud reference map in advance, which costs a large quantity of manpower and time. For this reason, a LiDAR-based initial global localization method using imperfect architectural skeleton information is proposed in this work. Firstly, we propose a lightweight management scheme for collected imperfect architectural information, which is convenient for efficient registration with real scans. Secondly, we extract architectural skeletons (stable man-made structures such as walls and columns) from both architectural information and real scans, and design them as line pairs feature patterns like P-LP, V-LP and C-LP. Thirdly, we propose a matrix descriptor for line pairs feature patterns description and initial matching. Finally, we construct error equations to estimate the pose by initial matching line pairs, and acquire the optimal localization results with the highest hit ratio on architectural grid map. A mobile robotic platform with the 16 beam LiDAR is experimented in typical indoor scenes such as rooms, corridors and undergrounding parking lots. Experiments show that the success rate of initial global localization reaches 80%, the average position error is about 0.10m and the running time is about 400ms per 1000 scans, which meet the requirements of indoor autonomous driving.

Wearable Permanent Magnet Tracking System for Wireless Capsule Endoscope

Permanent magnet tracking technology is a reliable solution used to localize the wireless capsule endoscope (WCE) with an embedded permanent magnet. However, in wearable permanent magnet tracking systems, compensation for the geomagnetic field plays a key role in the localization accuracy. The geomagnetic field superimposed on the sensor measurement will change in accordance with human body posture changes, thus causing a reduction in localization accuracy. In this study, a wearable permanent magnet tracking system and a corresponding calibration method are proposed to compensate for the geomagnetic field interference dynamically. The proposed system overcomes the shortcomings of existing systems that prevent patients from freely changing posture. In addition, a nine-axis inertial measurement unit is adopted to obtain a robust posture estimation that contributes to the compensation of the geomagnetic field. This innovation allows the localization accuracy to be maintained over a longer period of time than with current technology. The experiments show the effectiveness and stability of the wearable permanent magnet tracking system for realizing the geomagnetic field compensation. Within a tracking region of 46 cm <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times34$ </tex-math></inline-formula> cm <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times28$ </tex-math></inline-formula> cm, the proposed method achieves a 15-Hz update rate. Furthermore, the average position error caused by postural changes is reduced from 9.9 mm to 1.8 mm, and the average orientation error is reduced from 25.4° to 5.1°.

Calibration and Localization of Optically Pumped Magnetometers Using Electromagnetic Coils.

In this paper, we propose a method to estimate the position, orientation, and gain of a magnetic field sensor using a set of (large) electromagnetic coils. We apply the method for calibrating an array of optically pumped magnetometers (OPMs) for magnetoencephalography (MEG). We first measure the magnetic fields of the coils at multiple known positions using a well-calibrated triaxial magnetometer, and model these discreetly sampled fields using vector spherical harmonics (VSH) functions. We then localize and calibrate an OPM by minimizing the sum of squared errors between the model signals and the OPM responses to the coil fields. We show that by using homogeneous and first-order gradient fields, the OPM sensor parameters (gain, position, and orientation) can be obtained from a set of linear equations with pseudo-inverses of two matrices. The currents that should be applied to the coils for approximating these low-order field components can be determined based on the VSH models. Computationally simple initial estimates of the OPM sensor parameters follow. As a first test of the method, we placed a fluxgate magnetometer at multiple positions and estimated the RMS position, orientation, and gain errors of the method to be 1.0 mm, 0.2°, and 0.8%, respectively. Lastly, we calibrated a 48-channel OPM array. The accuracy of the OPM calibration was tested by using the OPM array to localize magnetic dipoles in a phantom, which resulted in an average dipole position error of 3.3 mm. The results demonstrate the feasibility of using electromagnetic coils to calibrate and localize OPMs for MEG.

An accurate 3D augmented reality navigation system with enhanced autostereoscopic display for oral and maxillofacial surgery.

Oral and maxillofacial surgery demands high precision navigation to achieve optimal surgical outcomes. Augmented reality (AR) has been a promising solution for intuitive and accurate guidance, whereas existing systems have room for improvement. We propose a high quality and accurate in situ AR navigation system. Enhanced image quality is achieved by deploying eye tracking in lenticular-based autostereoscopic display. Viewpoint tracking and optical tracking are integrated to assure accurate in situ images. The proposed system can provide in situ AR images in real-time, with a viewing angle of 59.5°×49.1°, 3D image resolution of 0.35×0.21mm, and image accuracy of 0.99±0.70mm. A maxillary drilling experiment obtains average position error of 1.12±0.30mm and localization time of 8.7±3.9s, significantly better than conventional 2D navigation system. The proposed system can display high-quality AR images and therefore reduce positioning error and operating time.

Reliable SINS/SAR/GNSS Integrated Navigation System for the Supersonic Vehicle

In the GNSS-challenged/denied environments, compared with other aided navigation systems, synthetic aperture radar- (SAR-) aided navigation systems can achieve all-weather, all-day, and global positioning. This paper is aimed at designing a reliable navigation system for the supersonic vehicle in the whole flight phase. Considering the SAR is of great significance for the reliable and precise navigation of the supersonic vehicle, we introduce it to the terminal phase to cope with the interference of the global navigation satellite system (GNSS). At the same time, to accelerate the speed of SAR positioning, a positioning method based on the area feature and stack sequential decoding algorithm (SSDA) is designed. Then, applying the SAR positioning under the framework of the Kalman filter (KF), we propose a tightly coupled strapdown inertial navigation system (SINS)/SAR/GNSS integration algorithm, which can achieve positioning without a complicated fusion method between the front phase and terminal phase. The flight experiment results show that the proposed method can correct the average position error to 1.7 m within 1 cycle.

Characterization of Soft 3D Printed Actuators for Parallel Networks

Soft pneumatic actuators allow compliant force application and movement for a variety of tasks. While most soft actuators have compliance in directions perpendicular to their direction of force application, they are most often analyzed only in their direction of actuation. In this work, we show a characterization of a soft 3D printed bellows actuator that considers shear and axial deformations, modeling both active and passive degrees of freedom. We build a model based on actuator geometry and a parallel linear and torsional spring system which we fit to experimental data in order to obtain the model constants. We demonstrate this model on a soft delta mechanism and show how this single actuator model can be used to predict parallel structures of actuators with average positional errors of 4.0%. These results verify that the presented model and modeling approach can be used to speed up the design and simulation of more complex soft robot models by characterizing both active and passive forces of their one degree-of-freedom soft actuators.

Comparing Complementary Kalman Filters Against SLAM for Collaborative Localization of Heterogeneous Multirobot Teams

Abstract This paper presents an investigation of collaborative localization for heterogeneous robots, resulting in a scheme for relative localization of a heterogeneous team of low-cost mobile robots. A novel complementary Kalman filter (CKF) approach is presented to address collaborative localization and mapping by optimally estimating the error states of the team. This indirect filter optimally combines the inertial/visual proprioceptive measurements of each vehicle with stereoscopic measurements made by unmanned ground vehicles (UGVs). An analysis is presented for both CKF and simultaneous localization and mapping (SLAM) approaches on maps containing randomly placed obstacles. In both simulation and experiments, we demonstrate the proposed methodology with a heterogeneous robot team. Six behavioral strategies, specifying the role and behavior of each robot, are simulated and evaluated for both CKF and SLAM approaches on maps containing randomly placed obstacles. Results show that the sources of error, image quantization, asynchronous sensors, and a nonstationary bias were sufficiently modeled to estimate the pose of the aerial robot. The results demonstrate localization accuracies of 2–4 cm, on average, while the robots operate at a distance from each other between 1 m and 4 m. The best performing behavior for the CKF approach maintained an average positional error of 2.2 cm and a relative error of 0.30% of the distance traveled for the entire team at the conclusion of maneuvers. For all multi-UGV strategies, the CKF approach outperformed the best SLAM results by a 6.7 cm mean error (0.48% of distance traveled).

Automatic geo‐localization framework without GNSS data

The information unavailability of a global navigation satellite system (GNSS) for geo-localization in many military and tourism applications implies the need to employ other applicable types of information, such as automatic image-based location recognition. Unlike most existing localization methods that use GNSS information to locate an initial position, this paper proposes an automatic geo-localization framework that does not require GNSS data. The proposed framework is a two-stage pipeline that uses a query image and digital elevation model (DEM) data as input. The authors frame automatic geo-localization without GNSS recordings as a skyline matching problem. By extracting the skyline from the DEM data and query images, the query image can be localized by matching the query skyline feature to the DEM skyline database. It has been demonstrated that this low-cost approach can perform efficiently in mountainous or hilly areas to produce reliable localization results. The system was tested on 50 testing site points within a large-scale area (China, 202.6 km2), and an average position error of 43.13 m was detected within 4.5 s.

Computer Vision-Based Localization With Visible Light Communications

Visible light positioning and computer vision-based localization have the potential to be cost-effective technologies for accurate indoor localization. However, the feasibility of existing methods in this domain is limited. In this paper, a novel visible light communication (VLC)-assisted perspective-four-line algorithm (V-P4L) is proposed for practical indoor localization. The basic idea of V-P4L is to jointly use VLC and computer vision techniques to achieve high localization accuracy regardless of LED height differences. In particular, the space-domain information is first exploited to estimate the orientation and coordinate information of a single rectangular LED luminaire in the camera coordinate system based on plane geometry theory and solid geometry theory. Then, by using time-domain information transmitted by VLC and the estimated luminaire information, the proposed V-P4L can estimate the position and pose of the camera using single-view geometry theory and the linear least square (LLS) method. To further mitigate the effect of height differences among LEDs on localization accuracy, a correction algorithm based on the LLS method and a simple optimization method is proposed. Due to the combination of time- and space-domain information, V-P4L can achieve accurate localization using a single luminaire without limitation on the correspondences between the features and their projections in conventional perspective-n-line (PnL) algorithms. Simulation results show that the position error caused by the proposed V-P4L algorithm is always less than 15 cm and the orientation error is always less than 4° using popular indoor luminaires. Experimental results with real hardware show that the average position error is less than 3 cm under both similar and different heights for the LEDs.

Wireless Position Method for Indoor Personnel Based on Improved Neural Network by Chaotic Whale Optimization Algorithm

The positioning accuracy of the indoor personnel positioning system is low, which cannot meet the needs of an intelligent transformer substation. Therefore, a wireless positioning method for indoor personnel based on the Elman neural network (ENN) using chaotic whale optimization algorithm (CWOA) was proposed. Firstly, a certain number of sample fingerprint databases were collected by wireless terminals in the wireless network on indoor personnel. Secondly, CWOA was used to optimize the weights and self-connected feedback gain factor of the ENN. Thirdly, the fingerprint database was used to train and test the optimized ENN, and the position algorithm model of the neural network was established. Finally, the real-time position was achieved with the neural network location algorithm model through the fingerprint data of the position point collected by the wireless terminal. The results show that the average position error of this indoor wireless positioning method is 1.24 m; ENN position algorithm by the CWOA has a stronger global search ability and is more suitable for the application in the time-varying indoor environment.