High Position Accuracy Research Articles

SIRN: An iterative reasoning network for transmission lines based on scene prior knowledge

High-accuracy positioning and identification of transmission line components is the premise for their status detection and fault diagnosis. However, due to the limitations of imbalance object scale and distribution in aerial images, the problem of detecting dense-tiny objects still needs to be solved. Considering the prior knowledge of fixed connection scenes, we proposed a novel method named scene iterative reasoning network (SIRN) based on self-adaptive clustering and local scene knowledge. It consists of the coarse-grained detector (CGD), scene self-adaptive clustering (SSAC) subnetwork, and scene structure iterative reasoning (SSIR) subnetwork. The CGD was proposed to obtain global coarse detection results. Then, the SSAC subnetwork utilized unsupervised self-adaptive clustering to obtain accurate dense regions based on the CGD results. Finally, the SSIR subnetwork was designed to extract and fuse the scene semantic structure information of local regions for realizing accurate dense-tiny object detection. In summary, the SIRN model utilizes an iterative inference strategy to detect dense regions accurately. By leveraging scene structure information, the model effectively transforms the challenge of detecting dense objects into an advantage, resulting in precise detection. The SIRN model is compatible with single-stage and two-stage object detection models, achieving a notable mean average precision (mAP) improvement ranging from 4.4% to 12.2% compared to the baseline models. Compared to other state-of-the-art object detection models, SIRN demonstrates significant advantages in accuracy and error rate metrics. In addition, qualitative and quantitative experiments show that the SIRN model considerably enhanced the detection of dense-tiny objects. The code is available at https://github.com/CharmingWang/SIRN.

Deep learning and hybrid approach for particle detection in defocusing particle tracking velocimetry

The present work aims at the improvement of particle detection in defocusing particle tracking velocimetry (DPTV) by means of a novel hybrid approach. Two deep learning approaches, namely faster R-CNN and RetinaNet are compared to the performance of two benchmark conventional image processing algorithms for DPTV. For the development of a hybrid approach with improved performance, the different detection approaches are evaluated on synthetic and images from an actual DPTV experiment. First, the performance under the influence of noise, overlaps, seeding density and optical aberrations is discussed and consequently advantages of neural networks over conventional image processing algorithms for image processing in DPTV are derived. Furthermore, current limitations of the application of neural networks for DPTV are pointed out and their origin is elaborated. It shows that neural networks have a better detection capability but suffer from low positional accuracy when locating particles. Finally, a novel Hybrid Approach is proposed, which uses a neural network for particle detection and passes the prediction onto a conventional refinement algorithm for better position accuracy. A third step is implemented to additionally eliminate false predictions by the network based on a subsequent rejection criterion. The novel approach improves the powerful detection performance of neural networks while maintaining the high position accuracy of conventional algorithms, combining the advantages of both approaches.

Non-line-of-sight imaging and location determination using deep learning

Non-line-of-sight (NLOS) imaging methods have been proposed and rapidly developed to address the challenge of detecting objects hidden from the direct line of sight. One critical issue in NLOS imaging is acquiring the spatial position of hidden objects. Active NLOS imaging methods solve this problem using complex designs and sophisticated equipment while the conventional passive NLOS imaging scheme, which does not have a controllable light source and time-gating detector, rarely captures the spatial information of objects. Using speckle patterns and the assistance of a deep learning method, we experimentally demonstrated that the passive NLOS imaging system could simultaneously visualize and locate objects hidden in corners like an active NLOS imaging system could. High-fidelity reconstructed images and high-accuracy location recognition were realized using the proposed network. The network could provide 99% recognition accuracy for the object's position with an axis spacing of 36 μm. These results were significant for NLOS imaging with high-accuracy positioning requirements.

Two-stage clustering for improve indoor positioning accuracy

High accurate positioning, as a key factor for the most resource management algorithms, has attracted a lot of attention due to its crucial role in some new applications such as smart factories, IoT, and autonomous cars. Indoor positioning, as another major category of positioning in addition to outdoor positioning, is a very complicated process and achieving high accuracy is a tough process. This paper proposes a two-stage clustering-based approach (TSCA) for indoor positioning to deal with highly complex data sets which was collected from a large-scale indoor radio system. In the first stage, a new clustering algorithm, called group matching method is developed, which divides the dataset into several groups (sub-datasets) according to the different reference signal received power (RSRP) values of the real-world dataset. In the second stage, KNN and its variants, are used to match and evaluate the location of each device in one of sub-datasets instead of the entire dataset, which can increase the accuracy of the positioning. This method can perfectly solve the problem of uneven distribution of reference point data in the process of data acquisition, which is a popular challenge for most real scene data acquisition. The proposed method is compared with several state-of-the-art ML methods such as Support Vector Regression (SVR), and clustering methods such as K-means. The results indicate a high positioning accuracy improvement of more than 55% compared to a modified KNN method use our own RSRP fingerprint dataset, and a 2D accuracy improvement of 13.36% and a 3D accuracy improvement of 10.3% compared to a traditional KNN method use a Wi-Fi Received Signal Strength fingerprint.

High-Accuracy Positioning in GNSS-Blocked Areas by Using the MSCKF-Based SF-RTK/IMU/Camera Tight Integration

The integration of global navigation satellite system (GNSS) single-frequency (SF) real-time kinematics (RTKs) and inertial navigation system (INS) has the advantages of low-cost and low-power consumption compared to the multiple-frequency GNSS RTK/INS integration system. However, due to the vulnerability of GNSS signal reception, the application of the GNSS SF-RTK/INS integration is limited in complex environments. To improve the positioning accuracy of SF-RTK/INS integration in GNSS-blocked environments, we present a low-cost tight integration system based on BDS/GPS SF-RTK, a low-cost inertial measurement unit (IMU), and a monocular camera. In such a system, a multi-state constraint Kalman filter (MSCKF) is adopted to integrate the single-frequency pseudo-range, phase-carrier, inertial measurements, and visual data tightly. A wheel robot dataset collected under satellite signal-blocked conditions is used to evaluate its performance in terms of position, attitude, and run time, respectively. Results illustrated that the presented model can provide higher position accuracy compared to those provided by the RTK/INS tight integration system and visual-inertial tight integration system. Moreover, the average running time presents the potential of the presented method in real-time applications.

Deep Learning-Based Cow Tail Detection and Tracking for Precision Livestock Farming

Cow tail detection and tracking in videos provide valuable information for individual identification, calving process, behavior analysis, and body condition monitoring. Although deep learning-based detection methods have demonstrated good performance, many of them have high complexity and still require an improvement in video object detection by reducing the computation time and false positives. To fully exploit the interframe information and achieve a real-time online detection, the optimized cow tail detection and tracking method is proposed based on an improved single shot multibox detector (SSD) and Kalman filter. Here, our improved SSD-integrated DenseNet and Inception-v4 reduce detection information loss and network parameters. Then, an improved window function-based Kalman filter and Hungarian are adopted to remove error detections and enhance the cow tail tracking accuracy. Experiments on our acquired rear-view videos show that the proposed approach achieved fast tail detection with an accuracy of 96.97%, a speed of 96 fps, a smaller model size of 25 MB, and higher position accuracy with an average 6.45 pixels deviation. The proposed approach outperformed the region-based R-CNN models and other tracking methods (e.g., particle filter), which provides a new solution to automatic cow detection and tracking in smart livestock farming.

Application of immuno- and affinity labeling with fluorescent dyes to in-resin CLEM of Epon-embedded cells

In-resin CLEM (Correlative Light and Electron Microscopy) of Epon-embedded cells involves correlating fluorescence microscopy with electron microscopy in the same Epon-embedded ultrathin section. This method offers the advantage of high positional accuracy compared to standard CLEM. However, it requires the expression of recombinant proteins. In order to detect the localization of endogenous target(s) and their localized ultrastructures of Epon-embedded samples using in-resin CLEM, we investigated whether immunological and affinity-labeling using fluorescent dyes applied to in-resin CLEM of Epon-embedded cells. The orange fluorescent (λem ∼550 nm) and far-red (λem ∼650 nm) fluorescent dyes examined maintained a sufficient level of fluorescent intensity after staining with osmium tetroxide and subsequent dehydration treatment with ethanol. Immunological in-resin CLEM of mitochondria and the Golgi apparatus was achieved using anti-TOM20, anti-GM130 antibodies, and fluorescent dyes. Two-color in-resin CLEM revealed that wheat germ agglutinin-puncta showed the ultrastructures of multivesicular body-like structures. Finally, taking the advantage of high positional accuracy, volume in-resin CLEM of mitochondria in the semi-thin section (2 μm thick) of Epon-embedded cells was performed by focused ion beam scanning electron microscopy. These results suggested that the application of immunological reaction and affinity-labeling with fluorescent dyes to in-resin CLEM of Epon-embedded cells is suitable for analyzing the localization of endogenous targets and their ultrastructures by scanning and transmission electron microscopy.

High-Resolution Mobile Mapping Platform Using 15-mm Accuracy LiDAR and SPAN/TerraStar C-PRO Technologies

Nowadays, most of the mobile mapping systems use GNSS/INS positioning technology and two-dimensional sensors to construct maps, self-localize, and gather environmental information, as well. Several problems can arise with traditional architectures of these systems, especially in situations where the GNSS signal is unavailable or multiple paths are involved, such as reliability issues and poor accuracy. Moreover, their cost of up to 2 million USD still poses a significant challenge for the development of new GIS applications. This paper proposes a new design of a mobile mapping system that incorporates a 1.5 cm accurate 3D LiDAR sensor and a high accuracy positioning system based on SPAN/TerraStar C-PRO technologies. The Extended Kalman Filter was used in this research to reduce the impact of GNSS signal loss by combining the SLAM method with SPAN/TerraStar C-PRO technologies. In the experiments, the concept of our mobile mapping platform was validated using the simulation environment Gazebo. So as to evaluate the proposed platform, a real dataset was collected from a complex environment where the GNSS signal is rarely available, exactly, from the campus of Moncton -Université de Moncton. The obtained results disclosed that the proposed platform proves its performance in terms of accuracy and reliability. Due to the integration of SLAM algorithm with SPAN/TerraStarC-PRO technologies, the generated 3D point cloud map includes a number of 285 million points with an mean accuracy 0.28 m even in the case of GNSS signal loss.

Autonomous Needle Navigation in Retinal Microsurgery: Evaluation in ex vivo Porcine Eyes.

Important challenges in retinal microsurgery include prolonged operating time, inadequate force feedback, and poor depth perception due to a constrained top-down view of the surgery. The introduction of robot-assisted technology could potentially deal with such challenges and improve the surgeon's performance. Motivated by such challenges, this work develops a strategy for autonomous needle navigation in retinal microsurgery aiming to achieve precise manipulation, reduced end-to-end surgery time, and enhanced safety. This is accomplished through real-time geometry estimation and chance-constrained Model Predictive Control (MPC) resulting in high positional accuracy while keeping scleral forces within a safe level. The robotic system is validated using both open-sky and intact (with lens and partial vitreous removal) ex vivo porcine eyes. The experimental results demonstrate that the generation of safe control trajectories is robust to small motions associated with head drift. The mean navigation time and scleral force for MPC navigation experiments are 7.208 s and 11.97 mN, which can be considered efficient and well within acceptable safe limits. The resulting mean errors along lateral directions of the retina are below 0.06 mm, which is below the typical hand tremor amplitude in retinal microsurgery.

Improving the Spatial Accuracy of UAV Platforms Using Direct Georeferencing Methods: An Application for Steep Slopes

The spatial accuracy of unmanned aerial vehicles (UAVs) and the images they capture play a crucial role in the mapping process. Researchers are exploring solutions that use image-based techniques such as structure from motion (SfM) to produce topographic maps using UAVs while accessing locations with extremely high accuracy and minimal surface measurements. Advancements in technology have enabled real-time kinematic (RTK) to increase positional accuracy to 1–3 times the ground sampling distance (GSD). This paper focuses on post-processing kinematic (PPK) of positional accuracy to achieve a GSD or better. To achieve this, precise satellite orbits, clock information, and UAV global navigation satellite system observation files are utilized to calculate the camera positions with the highest positional accuracy. RTK/PPK analysis is conducted to improve the positional accuracies obtained from different flight patterns and altitudes. Data are collected at altitudes of 80 and 120 meters, resulting in GSD values of 1.87 cm/px and 3.12 cm/px, respectively. The evaluation of ground checkpoints using the proposed PPK methodology with one ground control point demonstrated root mean square error values of 2.3 cm (horizontal, nadiral) and 2.4 cm (vertical, nadiral) at an altitude of 80 m, and 1.4 cm (horizontal, oblique) and 3.2 cm (vertical, terrain-following) at an altitude of 120 m. These results suggest that the proposed methodology can achieve high positional accuracy for UAV image georeferencing. The main contribution of this paper is to evaluate the PPK approach to achieve high positional accuracy with unmanned aerial vehicles and assess the effect of different flight patterns and altitudes on the accuracy of the resulting topographic maps.

Moment‐Based Shape‐Learning Holography for Fast Classification of Microparticles

Compact computational in‐line holography based on deep learning is an attractive single‐shot approach to image microparticles dispersed in 3D volume. The particle shape contains valuable information for species classification, but the dataset acquisition process for network training suffers from labor consuming and low efficiency due to the complex and varied 2D shapes of different particle species. Herein, a moment‐based shape‐learning holography (MSLH) is proposed where the shape of a microparticle is mathematically characterized using varying weights of Zernike moments. By decomposing and recombining feature shapes of pollen particles into numerous new characteristic shapes as incremental data, MSLH improves the efficiency of dataset preparation. The depth‐encoded shape‐learning is achieved using a U‐net with self‐attention mechanism, which enables fast axial depth determination. The shape reconstruction uses a wavelet‐based method with more explicit physical meanings, making MSLH a hybrid data‐and‐model driven approach that requires fewer primary data. Validation results show that MSLH achieves high accuracy in axial position and shape reconstruction, while maintaining good classification effectiveness. It is believed that MSLH is an easy‐to‐setup, efficient‐to‐construct, and fast‐to‐output approach for shape‐based classifications of 3D distributed microparticles in dynamic fluid.

Comprehensive Evaluations of NLOS and Linearization Errors on UWB Positioning

Currently, ultra-wide band (UWB) is adopted as a useful high-accuracy positioning technique in satellite-blocked areas. However, UWB’s positioning performance would be limited significantly because of non-line of sight (NLOS) errors. Additionally, the truncation errors in these linearization-based adjustments such as least squares (LS) and extended Kalman filter (EKF) would also visibly degrade UWB positioning accuracy. To overcome the impacts of NLOS errors and truncation errors, this paper introduced a robust-theory-based particle filter (RPF) into UWB positioning. In such a method, the IGG-III model and PF were adopted to limit the impacts of NLOS errors and truncation errors, respectively, by introducing a weight inflation factor and particle group. For comparison, the Bancroft, LS, EKF, unscented Kalman filter (UKF), cubature Kalman filter (CKF), PF, and RPF were also presented. Here, the influences of truncation errors were analyzed by comparing the results based on LS and EKF with those calculated by UKF, CKF, and PF. The impacts of NLOS errors were evaluated by making a comparison between the results of PF and RPF. Results based on a set of simulated UWB data and a group of experiment UWB data demonstrated that the RPF can significantly avoid the positioning errors caused by both truncation errors and NLOS errors. In general, position improvements percentages of 57.2%, 52.7%, 39.6%, 38.2%, 26.6%, and 20.4% can be obtained by RPF compared to those calculated by Bancroft, LS, EKF, UKF, CKF, and PF, respectively. As a comparison, the truncation error would lead to about 8.1%, 10.1%, and 33.2% accuracy decrease in the north, east, and vertical directions on average. Such accuracy-decrease rates caused by NLOS were 6.1%, 5.2%, and 25%.

NSGA-II based optimal Wi-Fi access point placement for indoor positioning: A BIM-based RSS prediction

One of the most basic systems required in smart cities are indoor positioning systems that allow positioning of people and objects inside buildings to provide them with suitable location-based services. Due to the existence of Wi-Fi technology infrastructure in most buildings and the fact that most mobile devices are equipped with this technology, indoor positioning systems based on Wi-Fi technology and fingerprinting method are quite popular. Therefore, the optimal placement of Wi-Fi Access Points (APs) is important to maximize the accuracy of indoor positioning. This paper presents a method to generate virtual fingerprints inside buildings by predicting Wi-Fi RSS values using integration of BIM and signal propagation models. The proposed method allows the optimization of the spatial distribution and number of Wi-Fi APs. The experiments reveal improvement in signal propagation modelling to measure RSS, efficient RSS fingerprinting-based IPS in the offline phase, decrease in the number of measurement efforts, and optimization of the spatial distribution of Wi-Fi APs, that finally was resulted in high overall accuracy of the indoor positioning.

A novel knowledge-driven flexible human–robot hybrid disassembly line and its key technologies for electric vehicle batteries

Based on the unique problems and challenges in the disassembly scenario of waste electric vehicle batteries (EVBs), we propose a knowledge-driven flexible human–robot hybrid disassembly line. The disassembly line can not only split the EVB disassembling tasks layer by layer to the primitive-level subtasks based on knowledge but also intelligently generate the optimal disassembling primitive sequence to complete the disassembling task. To further elaborate on the key technologies of the above-mentioned disassembly line, we focus on the implementation of the high-accuracy screw disassembly and disassembly planning system based on NeuroSymbolic. Implementation of the high-accuracy screw disassembly is realized by a new design of the disassembly actuator, high-accuracy screw position and pose estimation based on visual perception, and robust insertion based on force perception. The disassembly planning system is constructed by defining disassembly primitives and introducing neural predicates which can take the sensor data as input and output the symbolic states. Experiments show that our system can effectively solve the screw disassembling task in complex disassembly scenarios.

Capacitive Linear Displacement Sensors With High Accuracy and Long Range Absolute Positioning Capability Based on a Splicing Technique

Considerable progress has been made in the development of absolute capacitive linear displacement sensors based on time grating that are capable of providing highly accurate absolute position measurements in conjunction with easy fabrication using standard printed circuit board (PCB) technology. The working range of currently available sensor designs has, however, been rather limited. The present work addresses this issue by greatly expanding the measurement range via a technique for splicing multiple identical stationary rulers with individual measurement lengths of 600 mm. Here, absolute positioning is realized in a manner similar to that of Vernier calipers using stationary segments composed of periodic rectangular excitation electrodes with orthogonal excitation signals and a moving component composed of sinusoidal-shaped induction electrodes. The sensor wiring is greatly simplified by transferring the induction signals output by the moving ruler to the stationary rulers via special transmitting and receiving electrodes. Splicing errors are eliminated by applying two sets of induction electrodes, including a front sensing unit (FSU) and a rear sensing unit (RSU) along the measurement direction on the moving ruler. The performance of the proposed design is evaluated experimentally using a two-segment prototype sensor fabricated by PCB technology. The results demonstrate that interference between the FSU and RSU generates a first harmonic error. This is then addressed by increasing the distance between the FSU and RSU, and enhancing the shielding layer between the two units. This yields a measurement accuracy of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\pm 3.5~\mu \text{m}$ </tex-math></inline-formula> over the full 988 mm measurement range of the sensor.

The Development of Piezoelectric Inchworm Actuator Clamped With Magnetorheological Elastomer and Its Potential Application in Brain–Computer Interface Implantation

The piezoelectric inchworm actuator with high-positioning accuracy and large output force is a potential device to implant microelectrode in the invasive brain–computer interface (BCI). However, the traditional piezoelectric inchworm actuator adopts multichannel electrical control signals from external equipment and rigid-to-rigid friction contact to drive, resulting in complex control and uneven friction. To solve the above problems, in this article, magnetorheological elastomer (MRE) is first proposed as the clamping material of the piezoelectric inchworm actuator. An MRE–capillary–cover sandwich structure is adopted to achieve rigid-to-elastomeric clamping. The driving and clamping actions of the actuator are controlled by two cam mechanisms mounted on the same shaft, and a battery can feed the actuator. A kinetic model is established to analyze and predict the actuator motion. Several crucial clamping parameters are explored, and the performance experiments are conducted to evaluate actuator characteristics. A verification experiment is carried out to prove the feasibility of the microelectrode implantation in the pig brain (0.6% agarose gel). The actuator can reach the maximum velocity of 1250.4 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">μ</i> m/s and the maximum output force of 420 mN; in the forward/backward motion, the minimum step is 0.5837/0.5912 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">μ</i> m, and the repeatability is 0.0465/0.0469 <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">μ</i> m; thus, the actuator has great application potential in BCI.

Underwater Acoustic Positioning for Close Sources Using Time Division and Code Division Multiplexing

Underwater acoustic positioning system (UAPS) is used to know the positions of underwater robots and underwater structures. In ultra-short baseline (USBL) acoustic positioning systems, the three-dimensional position is determined by measuring the time difference of arrival (TDOA). In this paper, we investigate the acoustic positioning system targeting multiple sound sources and propose a simultaneous multi-point measurement method using time division and code division multiplexing (TD-CDM). TD-CDM provides higher position accuracy than code division multiplexing (CDM) and has a much shorter positioning time than time division and multiplexing (TDM). The effectiveness of TD-CDM has been proven by the results of the water tank experiment and simulation.

Clinical value of styrofoam fixation in intracranial tumor radiotherapy.

To analyze the application value of two postural fixation techniques.(styrofoam combined with head mask and fixed headrest combined with head mask) in intracranial tumor radiotherapy via cone beam computed tomography (CBCT). This study included 104 patients with intracranial tumors undergoing radiotherapy. The patients were divided into two groups: Group A (54 cases with styrofoam fixation) and Group B (50 cases with fixed headrest fixation). The positional deviation in 3D space between the two groups was compared using CBCT. The set-up errors were expressed as median (25th percentile, 75th percentile)or M(p25, p75) since the set-up errors in all directions were not normally distributed,The Mann-Whitney U test was performed. The age and gender of patients in the two groups were not significantly different. The set-up errors of A in lateral (X), longitudinal (Y), vertical (Z), and yaw(Rtn) axes were 1.0 (0,1) mm, 1.0 (0,1) mm, 1.0 (0,2) mm, and 0.4 (0.1, 0.8) degrees, respectively while the set-up errors of B were 1.0 (0,1) mm, 1.0 (1,2) mm, 1.0 (0,2) mm, and 0.5 (0.15,0.9) degrees, respectively. Moreover, patients in the styrofoam group had significantly smaller set-up errors in the Y-axis than patients in the headrest group (p=0.001). However, set-up errors in the X, Z, and Rtn axes were not significantly different between the two groups. The expansion boundaries of the target area in the X, Y, and Z directions were 1.77 mm, 2.45 mm, and 2.47 mm, respectively. The outer expansion boundaries of the headrest group were 2.03 mm, 3.88 mm, and 2.57 mm in X, Y, and Z directions, respectively. The set-up times of groups A and B were (32.71 ± 5.21) seconds and (46.57 ± 6.68) seconds, respectively (p=0.014). Patients in group A had significantly better comfort satisfaction than patients in group B (p=0.001). Styrofoam plus head thermoplastic mask body fixation technique has a higher positional accuracy in intracranial tumor radiotherapy than headrest plus head thermoplastic mask fixation. Besides, styrofoam plus head thermoplastic mask body fixation technique is associated with improved positioning efficiency, and better comfort than headrest plus head thermoplastic mask fixation, and thus can be effectively applied for intracranial tumor radiotherapy positioning.

Blockchain-Based Deep CNN for Brain Tumor Prediction Using MRI Scans.

Brain tumors are nonlinear and present with variations in their size, form, and textural variation; this might make it difficult to diagnose them and perform surgical excision using magnetic resonance imaging (MRI) scans. The procedures that are currently available are conducted by radiologists, brain surgeons, and clinical specialists. Studying brain MRIs is laborious, error-prone, and time-consuming, but they nonetheless show high positional accuracy in the case of brain cells. The proposed convolutional neural network model, an existing blockchain-based method, is used to secure the network for the precise prediction of brain tumors, such as pituitary tumors, meningioma tumors, and glioma tumors. MRI scans of the brain are first put into pre-trained deep models after being normalized in a fixed dimension. These structures are altered at each layer, increasing their security and safety. To guard against potential layer deletions, modification attacks, and tempering, each layer has an additional block that stores specific information. Multiple blocks are used to store information, including blocks related to each layer, cloud ledger blocks kept in cloud storage, and ledger blocks connected to the network. Later, the features are retrieved, merged, and optimized utilizing a Genetic Algorithm and have attained a competitive performance compared with the state-of-the-art (SOTA) methods using different ML classifiers.

Research on Position Sensorless Control of RDT Motor Based on Improved SMO with Continuous Hyperbolic Tangent Function and Improved Feedforward PLL

With the increasing use of electric propulsion ships, the emergence of the shaftless rim-driven thruster (RDT) as a revolutionary integrated motor thruster is gradually becoming an important development direction for green ships. The shaftless structure of RDTs leads to their dependence on position sensorless control techniques. In this study, a novel control algorithm using a composite sliding mode observer (SMO) with a modified feed-forward phase-locked loop (PLL) is presented for achieving high accuracy position and speed control of shaftless RDT motors. The deviation between the observed and actual currents is exploited to develop a current SMO to extract back electromotive force (back-EMF) errors. On this basis, a back-EMF observer is established to achieve accurate estimation of the back-EMF. The basic structure of the PLL was modified and incorporates a speed feedforward mechanism, which enhances the performance of rotor position estimation and facilitates bidirectional rotation. The stability of the algorithm has been verified in Matlab/Simulink for a range of steady-state, dynamic, and ship propeller loading conditions. Remarkably, the control algorithm boasts an impressive adjustment time of approximately 0.006 s and its position estimation error may be as low as 0.03 rad. Simulation results highlight the performance of the algorithm to achieve bidirectional rotation, while exhibiting fast convergence, minimal vibration, exceptional control accuracy, and robustness.