Robust Localization Research Articles

Conformational transitions of the Spindly adaptor underlie its interaction with Dynein and Dynactin.

Cytoplasmic Dynein 1, or Dynein, is a microtubule minus end-directed motor. Dynein motility requires Dynactin and a family of activating adaptors that stabilize the Dynein-Dynactin complex and promote regulated interactions with cargo in space and time. How activating adaptors limit Dynein activation to specialized subcellular locales is unclear. Here, we reveal that Spindly, a mitotic Dynein adaptor at the kinetochore corona, exists natively in a closed conformation that occludes binding of Dynein-Dynactin to its CC1 box and Spindly motif. A structure-based analysis identified various mutations promoting an open conformation of Spindly that binds Dynein-Dynactin. A region of Spindly downstream from the Spindly motif and not required for cargo binding faces the CC1 box and stabilizes the intramolecular closed conformation. This region is also required for robust kinetochore localization of Spindly, suggesting that kinetochores promote Spindly activation to recruit Dynein. Thus, our work illustrates how specific Dynein activation at a defined cellular locale may require multiple factors.

Robust AOA-Based Target Localization for Uniformly Distributed Noise via ℓp-ℓ1 Optimization.

This paper addresses the problem of robust angle of arrival (AOA) target localization in the presence of uniformly distributed noise which is modeled as the mixture of Laplacian distribution and uniform distribution. Motivated by the distribution of noise, we develop a localization model by using the -norm with as the measurement error and the -norm as the regularization term. Then, an estimator for introducing the proximal operator into the framework of the alternating direction method of multipliers (POADMM) is derived to solve the convex optimization problem when . However, when , the corresponding optimization problem is nonconvex and nonsmoothed. To derive a convergent method for this nonconvex and nonsmooth target localization problem, we propose a smoothed POADMM estimator (SPOADMM) by introducing the smoothing strategy into the optimization model. Eventually, the proposed algorithms are compared with some state-of-the-art robust algorithms via numerical simulations, and their effectiveness in uniformly distributed noise is discussed from the perspective of root-mean-squared error (RMSE). The experimental results verify that the proposed method has more robustness against outliers and is less sensitive to the selected parameters, especially the variance of the measurement noise.

CrowdLoc: Robust image indoor localization with edge-assisted crowdsensing

Image indoor localization, which is advantageous in infrastructure-less deployment and high positioning accuracy among various indoor localization schemes, is usually deployed in a lightweight end (shooting picture) and heavyweight server (searching image for location in the database) manner. Such a scheme is suffered from two major drawbacks: (1) The uploading and searching of the shooting pictures cause a severe burden to the server. (2) The image map will become defected as time elapse, such as missing features in initialization maps and continuous changes in the scene. To keep the image map up-to-date, a labor-intensive site survey is required with dedicated devices. The frequent updating will further exhaust the limited resources in the server. These two drawbacks have greatly hindered the image-based indoor localization from large-scale deployment. To this end, we study how to use crowdsourcing data to mitigate map defects and optimize location computation and map storage through edge architecture. In this paper, we propose CrowdLoc, an edge-assisted image localization architecture with lightweight multi-view localization on clients and crowdsourcing map-cache on the edge. Our method is mainly innovative in three parts. Firstly, CrowdLoc provides a lightweight multi-view localization on the client-side and the low-confidence localization can be offloaded to the edge server to use the crowdsourcing map for more accurate calculation. Secondly, CrowdLoc uses a map defect perception algorithm to recognize the update of the scene during user localization and update the crowdsourcing map-cache automatically on the edge side. Thirdly, in order to improve cache utilization, we propose a cache elimination strategy that runs periodically on the edge side. It can classify cache data according to the frequency of cache hits on a regular basis, remove invalidation data, and persist long-term updated data to the client map to reduce clients’ computing offload. We evaluate the performance of our method based on the real data in the lobby of an experimental building with a map missing rate of 20% and a map update rate of 40%. The results show that the mean localization error of CrowdLoc is 1.54 meters, which is 59.4% lower than that of multi-view localization without crowdsourcing maps, and only 0.56 meters higher than localization in non-defective maps.

Semi-Supervised Deep Adversarial Forest for Cross-Environment Localization

Extracting channel state information (CSI) from WiFi signals is of proved high-effectiveness in locating human locations in a device-free manner. However, existing localization/positioning systems are mainly trained and deployed in a fixed environment, and thus they are likely to suffer from substantial performance declines when immigrating to new environments. In this paper, we address the fundamental problem of WiFi-based cross-environment indoor localization using a semi-supervised approach, in which we only have access to the annotations of the source environment while the data in the target environments are un-annotated. This problem is of high practical values in enabling a well-trained system to be scalable to new environments without tedious human annotations. To this end, a deep neural forest is introduced which unifies the ensemble learning with the representation learning functionalities from deep neural networks in an end-to-end trainable fashion. On top of that, an adversarial training strategy is further employed to learn environment-invariant feature representations for facilitating more robust localization. Extensive experiments on real-world datasets demonstrate the superiority of the proposed methods over state-of-the-art baselines. Compared with the best-performing baseline, our model excels with an average 12.7% relative improvement on all six evaluation settings.

IoT Sensor Selection for Target Localization: A Reinforcement Learning based Approach

Internet of things (IoT) is a key enabler for target localization , where IoT-based sensors work towards identifying target's location in an area of interest (AoI). Appropriate selection of IoT sensors, while considering their energy constraints and confidence in the data they provide, is crucial to preserve the network lifetime and improve the efficiency of the localization task. The contemporary localization systems focus on optimizing nodes’ placement, and their selection. However, these systems can prove inefficient when deployed in new environments, since they lack adaptability and dynamicity. Machine learning , specifically reinforcement learning (RL), is well-suited to build robust target localization systems. Nevertheless, there exist some limitations in the current RL-based localization systems, which include– 1) they are application specific and require prior knowledge about the source characteristics (source type and strength), 2) they are geared towards a fixed environment (area size and number of sensors), 3) they use sensor-equipped agent traversing the AoI, which lacks scalability. Therefore, this work presents a comprehensive active sensor selection mechanism, coupled with a data-driven Q-learning approach, for target localization. The trained RL model guides the localization task to find a source with unknown location and intensity, irrespective of the environment or the source characteristics without retraining. A state-space reduction technique is proposed to diminish the curse of dimensionality and enhance the generalization ability of the RL model, thus realizing a scalable and adaptable system. The proposed approach is compared with a benchmark which is the only other work that utilizes sensor selection in localization tasks, where the results indicate that a more efficient, scalable, and adaptable localization is achieved.

Deep learning-based robust positioning for all-weather autonomous driving

Interest in autonomous vehicles (AVs) is growing at a rapid pace due to increased convenience, safety benefits and potential environmental gains. Although several leading AV companies predicted that AVs would be on the road by 2020, they are still limited to relatively small-scale trials. The ability to know their precise location on the map is a challenging prerequisite for safe and reliable AVs due to sensor imperfections under adverse environmental and weather conditions, posing a formidable obstacle to their widespread use. Here we propose a deep learning-based self-supervised approach for ego-motion estimation that is a robust and complementary localization solution under inclement weather conditions. The proposed approach is a geometry-aware method that attentively fuses the rich representation capability of visual sensors and the weather-immune features provided by radars using an attention-based learning technique. Our method predicts reliability masks for the sensor measurements, eliminating the deficiencies in the multimodal data. In various experiments we demonstrate the robust all-weather performance and effective cross-domain generalizability under harsh weather conditions such as rain, fog and snow, as well as day and night conditions. Furthermore, we employ a game-theoretic approach to analyse the interpretability of the model predictions, illustrating the independent and uncorrelated failure modes of the multimodal system. We anticipate our work will bring AVs one step closer to safe and reliable all-weather autonomous driving.

Topological edge states of a long-range surface plasmon polariton at the telecommunication wavelength.

Confining light by plasmonic waveguides is promising for miniaturizing optical components, while topological photonics has been explored for robust light localization. Here we propose combining the two approaches into a simple periodically perforated plasmonic waveguide (PPW) design exhibiting robust localization of long-range surface plasmon polaritons. We predict the existence of a topological edge state originating from a quantized topological invariant, and numerically demonstrate the viability of its excitation at telecommunication wavelength using near-field and waveguide-based approaches. Strong modification of the radiative lifetime of dipole emitters by the edge state, and its robustness to disorder, are demonstrated.

PSE-Match: A Viewpoint-Free Place Recognition Method With Parallel Semantic Embedding

Accurate localization on the autonomous driving cars is essential for autonomy and driving safety, especially for complex urban streets and search-and-rescue subterranean environments where high-accurate GPS is not available. However current odometry estimation may introduce the drifting problems in long-term navigation without robust global localization. The main challenges involve scene divergence under the interference of dynamic environments and effective perception of observation and object layout variance from different viewpoints. To tackle these challenges, we present PSE-Match, a viewpoint-free place recognition method based on parallel semantic analysis of isolated semantic attributes from 3D point-cloud models. Compared with the original point cloud, the observed variance of semantic attributes is smaller. PSE-Match incorporates a divergence place learning network to capture different semantic attributes parallelly through the spherical harmonics domain. Using both existing benchmark datasets and two in-field collected datasets, our experiments show that the proposed method achieves above 70% average recall with top one retrieval and above 95% average recall with top ten retrieval cases. And PSE-Match has also demonstrated an obvious generalization ability with limited training dataset.

G-VIDO: A Vehicle Dynamics and Intermittent GNSS-Aided Visual-Inertial State Estimator for Autonomous Driving

This paper proposes G-VIDO, a vehicle dynamics, and intermittent Global Navigation Satellite System (GNSS)-aided visual-inertial state estimator, to address the state estimation problem of autonomous vehicle localization (i.e., position and orientation estimation in the global coordinate system) under various GNSS states. A dynamics pre-integration theory is proposed on the basis of a two-degree-of-freedom (DOF) vehicle dynamics model, and dynamics constraints are built in the optimization back-end, considering the unobservable problem of the monocular visual-inertial system under degenerate motions. The proposed highly nonlinear system can be robustly initialized by loosely aligning the monocular structure from motion (SfM) results, pre-integrated IMU measurements, and vehicle motion information. GNSS is used for reference frame transformation and constraint construction in the sliding window. The cumulative error can be corrected with the aid of GNSS, and the vehicle’s position in the global coordinate system can be determined. A GNSS anomaly detection algorithm is proposed to improve the system robustness under intermittent GNSS. Experiments have shown that G-VIDO can provide real-time, robust, and seamless localization in multiple GNSS states, with an RMSE of less than 30 cm (with GNSS). Moreover, we proved that the initialization and local odometry modules in G-VIDO outperform several state-of-the-art VIO systems and our preliminary work VINS-Vehicle.

Real-Time Localization and Mapping Utilizing Multi-Sensor Fusion and Visual–IMU–Wheel Odometry for Agricultural Robots in Unstructured, Dynamic and GPS-Denied Greenhouse Environments

Autonomous navigation in greenhouses requires agricultural robots to localize and generate a globally consistent map of surroundings in real-time. However, accurate and robust localization and mapping are still challenging for agricultural robots due to the unstructured, dynamic and GPS-denied environmental conditions. In this study, a state-of-the-art real-time localization and mapping system was presented to achieve precise pose estimation and dense three-dimensional (3D) point cloud mapping in complex greenhouses by utilizing multi-sensor fusion and Visual–IMU–Wheel odometry. In this method, measurements from wheel odometry, an inertial measurement unit (IMU) and a tightly coupled visual–inertial odometry (VIO) are integrated into a loosely coupled framework based on the Extended Kalman Filter (EKF) to obtain a more accurate state estimation of the robot. In the multi-sensor fusion algorithm, the pose estimations from the wheel odometry and IMU are treated as predictions and the localization results from VIO are used as observations to update the state vector. Simultaneously, the dense 3D map of the greenhouse is reconstructed in real-time by employing the modified ORB-SLAM2. The performance of the proposed system was evaluated in modern standard solar greenhouses with harsh environmental conditions. Taking advantage of measurements from individual sensors, our method is robust enough to cope with various challenges, as shown by extensive experiments conducted in the greenhouses and outdoor campus environment. Additionally, the results show that our proposed framework can improve the localization accuracy of the visual–inertial odometry, demonstrating the satisfactory capability of the proposed approach and highlighting its promising applications in autonomous navigation of agricultural robots.

More comprehensive proprioceptive stimulation of the hand amplifies its cortical processing.

Corticokinematic coherence (CKC) quantifies the phase coupling between limb kinematics and cortical neurophysiological signals reflecting proprioceptive feedback to the primary sensorimotor (SM1) cortex. We studied whether the CKC strength or cortical source location differs between proprioceptive stimulation (i.e., actuator-evoked movements) of right-hand digits (index, middle, ring, and little). Twenty-one volunteers participated in magnetoencephalography measurements during which three conditions were tested: 1) simultaneous stimulation of all four fingers at the same frequency, 2) stimulation of each finger separately at the same frequency, and 3) simultaneous stimulation of the fingers at finger-specific frequencies. CKC was computed between MEG responses and accelerations of the fingers recorded with three-axis accelerometers. CKC was stronger (P < 0.003) for the simultaneous (0.52 ± 0.02) than separate (0.45 ± 0.02) stimulation at the same frequency. Furthermore, CKC was weaker (P < 0.03) for the simultaneous stimulation at the finger-specific frequencies (0.38 ± 0.02) than for the separate stimulation. CKC source locations of the fingers were concentrated in the hand region of the SM1 cortex and did not follow consistent finger-specific somatotopic order. Our results indicate that proprioceptive afference from the fingers is processed in partly overlapping cortical neuronal circuits, which was demonstrated by the modulation of the finger-specific CKC strengths due to proprioceptive afference arising from simultaneous stimulation of the other fingers of the same hand as well as overlapping cortical source locations. Finally, comprehensive simultaneous proprioceptive stimulation of the hand would optimize functional cortical mapping to pinpoint the hand region, e.g., prior brain surgery.NEW & NOTEWORTHY Corticokinematic coherence (CKC) can be used to study cortical proprioceptive processing and localize proprioceptive hand representation. Our results indicate that proprioceptive stimulation delivered simultaneously at the same frequency to fingers (D2-D4) maximizes CKC strength allowing robust and fast localization of the human hand region in the sensorimotor cortex using MEG.

Leveraging Local Planar Motion Property for Robust Visual Matching and Localization

One primary difficulty preventing the visual localization for service robots is the robustness against changes, including environmental changes and perspective changes. In recent years, learning-based feature matching methods have been widely studied and effectively verified in practical applications. Learning-based feature matching effectively solves the problem of environmental changes, including illumination changes and man-made changes. However, there is still room for improvement dealing with large perspective changes. In this letter, we leverage local planar motion property to simplify the affine transform and propose an augmentation-based feature matching method that greatly enhances the robustness to perspective changes. The proposed feature matching approach maintains low matching costs as the augmentation is performed on the simplified affine matrix space. Combined with the motion property aided minimal solution for pose estimation, an end-to-end robust visual localization system is proposed which is shown to bring 67% improvement in localization performance under large perspective changes in publicly available OpenLORIS dataset, while increasing computational cost by only 20% by using batch processing techniques with a single GPU. In addition, a guide for map frame selection is presented to support robust localization with very sparse map frames in storage. Experiments on the classified dataset with environmental changes and perspective changes validate the effectiveness of the proposed system.

Simultaneous Localization of Rail Vehicles and Mapping of Environment With Multiple LiDARs

Precise and real-time rail vehicle localization as well as railway environment monitoring is crucial for railroad safety. In this letter, we propose a multi-LiDAR based simultaneous localization and mapping (SLAM) system for railway applications. Our approach starts with measurements preprocessing to denoise and synchronize multiple LiDAR inputs. Different frame-to-frame registration methods are used according to the LiDAR placement. In addition, we leverage the plane constraints from extracted rail tracks to improve the system accuracy. The local map is further aligned with global map utilizing absolute position measurements. Considering the unavoidable metal abrasion and screw loosening, online extrinsic refinement is awakened for long-during operation. The proposed method is extensively verified on datasets gathered over 3000 km. The results demonstrate that the proposed system achieves accurate and robust localization together with effective mapping for large-scale environments. Our system has already been applied to a freight traffic railroad for monitoring tasks.

Combined Usage of Hydraulic Model Calibration Residuals and Improved Vector Angle Method for Burst Detection and Localization in Water Distribution Systems

This paper proposes a combined usage of the hydraulic model calibration residuals and an improved vector angle method for the Battle of the Leakage Detection and Isolation Methods (BattLeDIM). The proposed method starts with the hydraulic model decomposition and followed by data partition, with both procedures aimed at simplifying the BattLeDIM problem. Next, a calibration residuals–based burst detection approach is used, in which the nodal demands of the hydraulic model are calibrated in real-time using a newly developed algorithm based on prior information. The burst detection is achieved by identifying the anomalous calibration residuals. Finally, the locations of pipe bursts are approximated by an improved vector angle method, which compares the angle between each pipe’s sensitivity vector and the vector of calibration residuals. Application results to the BattLeDIM problem indicate the proposed method is effective and efficient in burst detection and localization, with nearly all abrupt bursts (9 out of the total 19 bursts) in 2019 being successfully detected within 5–10 min after their occurrence, and 7 out of 10 detected bursts being localized with a spatial difference less than 300 m from the true burst pipes. Performance comparisons with the Kalman filtering–based burst detection method have been conducted, and the obtained results indicate the proposed method has better performance under both consistent and varying operation conditions, as well as in detectability for small bursts. Lessons learned from this battle and future works are summarized towards the development of a more effective and robust burst detection and localization method.

Robust Multispectral Visual-Inertial Navigation With Visual Odometry Failure Recovery

Besides the large amount of information multispectral imaging offers, multispectral visual odometry remains overlooked due to the dissimilarity between modalities. In order to tackle the challenging feature matching between multispectral stereo images and to overcome the lack of robust multispectral visual localisation solutions, a novel approach is proposed in this paper. It consists in tracking features in each modality simultaneously, in a monocular manner and then, estimating motion in a windowed bundle adjustment framework and using the geometry of the stereo setup to recover the missing scale. The estimation is robustified by selecting adequate keyframes based on feature parallax and by maximising the mutual information between all the features reprojected in the stereo pair. Furthermore, the proposed multispectral visual odometry solution is integrated in an error-state Kalman filter framework to deal with challenging environments, where the quality of images is reduced. Two measurements models, using absolute and relative camera poses, are presented. The superiority of relative poses is then shown by providing a failure recovery algorithm which relies on inertial data when visual data are not accessible. The algorithm was tested on innovative series of visible-thermal multispectral datasets, acquired from a car with real driving conditions. An overall error of around 2% of the travelled distance was achieved on these datasets.

Abstract 1494: 5-methyltetrahydrofolate (5-MTHF) as a superior ligand for delivery of small RNAs for cancer therapy

Abstract MicroRNAs (miRNAs) are small noncoding RNAs that regulate a wide array of genes. Because miRNA downregulate translation proteins associated with cancer progression, miRNA-based therapeutics have great potential as anticancer drugs. Nonetheless, the therapeutic potential of miRNA replacement is limited due to lack of safe and efficient delivery vehicles, and inability to target miRNAs to the intended cells/tissues. Thus, identifying molecules that are able to deliver therapeutic miRNA is of critical need. This could be achieved by targeting receptors that are overexpressed on target tumor cells. For example, the folate receptor (FR) is overexpressed on various tumor types including breast, lung, ovarian and brain tumors. Our laboratory has previously developed a unique strategy that directly links a miRNA mimic to the high affinity FR ligand, folate (FolamiRs). In the present work, to further improve the potency of FolamiRs, we introduce the reduced folate metabolite, 5-methyltetrahydrofolate (5-MTHF) that maintains nanomolar affinity for the FR at pH 7, but disengages from the FR at a significantly higher pH than that of folate, which is important when including an endosomal escape agent that typically disrupts endosomal acidification. We synthesize and conjugate 5-MTHF ligand to the potent tumor suppressor miRNA-34a (5-MTHF-miR-34a). We show that 5-MTHF-miR-34a uptake is specific to the cells overexpressing FR and represses target gene expression. Our data supports the use of 5-MTHF in combination with the endosomal escape agent, nigericin, to achieve robust cytosolic localization of therapeutically relevant miRNAs such as miR-34a. This work provides the foundation for future studies using 5-MTHF- conjugated therapeutic agents for targeting additional FR expressing cells, including myeloid derived suppressor cells (MDSCs) and macrophages. Collectively these studies will pave the way for developing 5-MTHF-based small RNA conjugates for treating FR-expressing cancers. The details of these intriguing results will be discussed in the poster. Citation Format: Sudarsan R. Kasireddy, Ahmed M. Abdelaal, Philip S. Low, Andrea L. Kasinski. 5-methyltetrahydrofolate (5-MTHF) as a superior ligand for delivery of small RNAs for cancer therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1494.

Reconfigurable Topological States in Arrays of Bianisotropic Particles

Topological photonics promotes a novel approach to resilient light manipulation by exploiting spatio-temporal symmetries of the system and dual symmetry of electromagnetic field. Various prospective device applications pose the need to flexibly control robust field localization associated with topological modes. Here we design a topological array of resonant dielectric meta-atoms with bianisotropic response induced by a spatial symmetry reduction. Mutual orientation of the designed meta-atoms encodes a staggered bianisotropy pattern capable of trapping topological states in a one-dimensional array containing a small number of particles. We show that the topological interface state can be tailored by the rotation of coaxial bianisotropic particles arranged in an equidistant lattice. Our experimental implementation based on ceramic horseshoe-shaped disks demonstrates remarkable reconfigurability of electromagnetic topological states.

Research on rapid location method of mobile robot based on semantic grid map in large scene similar environment

AbstractAiming at the problem that adaptive Monte Carlo localization (AMCL) algorithm is difficult to localize in large scenes and similar environments. This paper uses a semantic information-assisted approach to improve the AMCL algorithm. This method realizes the robust localization of the robot in the large scenes and similar environments. Firstly, the 2D grid map created by simultaneous localization and mapping using lidar can obtain highly accurate indoor environmental contour information. Secondly, the semantic object capture is achieved by using a depth camera combined with an instance segmentation algorithm. Then, the semantic grid map is created by mapping the semantic point cloud through the back-projection process of the pinhole camera. Finally, semantic grid maps are used as a priori information to assist in localization, which will be used to improve the initial particle swarm distribution in the global localization of the AMCL algorithm and thus will solve the robot localization problem in this environment. The experimental evidence shows that the semantic grid map solves the environmental information degradation problem caused by 2D lidar as well as improves the robot’s perception of the environment. In addition, this paper improves the localization robustness of the AMCL algorithm in large scenes and similar environments, resulting in an average localization success rate of about 90% or even higher, and further reduces the number of iterations. The global localization problem of robots in large scenes and similar environments is effectively solved.

Performance evaluation of deep transfer learning on multi-class identification of common weed species in cotton production systems

Precision weed management offers a promising solution for sustainable cropping systems through the use of chemical-reduced/non-chemical robotic weeding techniques, which apply suitable control tactics to individual weeds or small clusters. Therefore, accurate identification of weed species plays a crucial role in such systems to enable precise, individualized weed treatment. Despite recent progress, the development of a robust weed identification and localization system in the presence of unstructured field environments remains a serious challenge, requiring supervised modeling using large volumes of annotated data. This paper makes a first comprehensive evaluation of deep transfer learning (DTL) for identifying common weed species specific to cotton (Gossypium hirsutum L.) production systems in southern United States (U.S.). A new dataset for weed identification was created, consisting of 5187 color images of 15 weed classes collected under natural light conditions and at varied weed growth stages, in cotton fields (primarily in Mississippi and North Carolina) during the 2020 and 2021 growth seasons. We evaluated 35 state-of-the-art deep learning models through transfer learning with repeated holdout validations and established an extensive benchmark for the considered weed identification task. DTL achieved high classification accuracy of F1 scores exceeding 95%, requiring reasonably short training time (less than 2.5 h) across models. ResNeXt101 achieved the best overall F1-score of 98.93 ± 0.34%, whereas 10 out of the 35 models achieved F1 scores near or above 98.0%. However, the performance on minority weed classes with few training samples was less satisfactory for models trained with a conventional, unweighted cross entropy loss function. To address this issue, a weighted cross entropy loss function was adopted, which achieved substantially improved accuracies for minority weed classes (e.g., the F1-scores for Xception and MnasNet on the Spurred Anoda weed increased from 48% to 90% and 50% to 82%, respectively). Furthermore, a deep learning-based cosine similarity metric was employed to analyze the similarity among weed classes, assisting in the interpretation of classifications. Both the codes (https://github.com/Derekabc/CottonWeeds) for model benchmarking and the weed dataset (https://www.kaggle.com/yuzhenlu/cottonweedid15) of this study are made publicly available, which expect to be a valuable resource for future research on weed identification and beyond.

Robust Localization in Cellular Networks via Reinforced Iterative M-Estimation and Fuzzy Adaptation

We consider robust mobile location estimation using range measurements from multiple base stations under probabilistically mixed line-of-sight (LOS)/non-LOS (NLOS) conditions in a cellular network, where the NLOS statistics and occurrence probabilities (NOPs) are unknown. We develop new algorithms for achieving desirable performance under these challenging conditions. First, we propose a measurement preprocessing (MPP) scheme for dynamically bounding the NLOS influence on raw range measurements. The mobile positioning problem is cast as the state estimation of an MPP-based system model. Next, we introduce extended Kalman filter (EKF)-based reinforced robust regression and develop a gradient descent-type iterative M-estimation procedure with ensured stability, resulting in a reinforced M-estimation-based robust EKF (M-REKF). Subsequently, we present a fuzzy-tuning M-REKF (FTM-REKF) incorporating adequate fuzzy adaptation of measurement noise covariances in the M-REKF. A performance analysis provides insight into how adequate fuzzy rules can be established methodically for the FTM-REKF. Simulations demonstrate that despite varying NOPs and NLOS statistics, the FTM-REKF and M-REKF stably yield satisfactory location estimates of moving and stationary targets, with the former outperforming the latter, and they can meet the Enhanced 911 positioning requirements. Moreover, the proposed algorithms significantly outperform several salient iterative M-estimation-based EKFs, and their MPP-based counterparts.