Low Localization Accuracy Research Articles

A Looseness Detection Method for Railway Catenary Fasteners Based on Reinforcement Learning Refined Localization

Brace sleeve (BS) fasteners, i.e., nut and bolt, are small components but play essential roles in fixing BS and cantilever in railway catenary system. They are commonly inspected by onboard cameras using computer vision to ensure the safety of railway operation. However, most BS fasteners cannot be directly localized because they are too small in the inspection images. Instead, the BS is first localized for detecting the BS fastener. This leads to a new problem that the localized BS boxes may not contain the complete BS fasteners due to low localization accuracy, making it infeasible to further diagnose the fastener conditions. To tackle this problem, this article proposes a novel pipeline for BS fastener looseness diagnosis. First, the competitive deep learning model Faster RCNN ResNet101 is used to coarsely localize BSs. Second, an action-driven reinforcement learning agent is adopted to refine the coarse-localized boxes through a dynamic position searching process. Then, BS fasteners are extracted from the refined localized BS image by the deep segmentation model YOLACT++, which is fast and interpretable. Finally, a looseness diagnosis criterion based on segmented information are proposed. We evaluate the performance of submodels independently and the overall performance of the whole model on a real-life catenary image dataset collected from a high-speed line in China. The test results show that the proposed method is effective for BS looseness detection in railway catenary.

Design of a New Lightweight Accurate Grid-Based Localization Method for WASN

Existing solutions designed for acoustic target localization based on the received signal energy in WSN suffer either from low localization accuracy due to suboptimal convergence or high computations requirements due to the application of complex algorithms. The main contribution of this article is the design of a new lightweight energy-based acoustic source localization solution intended for implementation in wireless sensors. The novelty of this solution resides in devising lightweight computation of the cost function in the location estimator algorithm and the application of the Tabu metaheuristic method to enhance the localization accuracy. In terms of localization accuracy, the proposed scheme outperforms many existing solutions with reduced localization error by up to 88% while using only three energy measurements, which saves around 85% of node energy. Furthermore, the performance analysis results showed a stable behavior of the proposed algorithm across a wide range of parameters variations of the energy-decay model, which demonstrates the robustness of the scheme.

Online glass confidence map building using laser rangefinder for mobile robots

Accurate localization and mapping are essential for mobile robots. Using laser rangefinders (LRFs), current state-of-the-art indoor Simultaneous Localization and Mapping (SLAM) can provide accurate real-time localization and mapping in most environments. An exemption are those where glass is predominant, as LRFs can not properly detect glass due to glass' transparency and reflectiveness. With such buildings becoming more common, this has become an important issue to address. Failure to detect glass causes two problems for SLAM: incorrectly mapping glass as open space; and, lower localization accuracy due to mismatches between measured and expected range data. In this paper, we propose a glass confidence map that correctly maps glass as occupied, as well as the probability of an object to be glass/non-glass. Our approach consists of four steps: (i) map all objects, even potential dynamic obstacles, as occupied, (ii) compute the probability of scanned objects to be glass/non-glass using a neural network, (iii) online map updates by matching scanned objects to probability map, and (iv) filter dynamic obstacles and noise. We validated our approach in an office with large glass areas, achieving more than 95% of glass areas correctly mapped as occupied with less than 5% glass/non-glass classification error.

Block-Sparse Coding-Based Machine Learning Approach for Dependable Device-Free Localization in IoT Environment

Device-free localization (DFL) locates targets without equipping with wireless devices or tag under the Internet-of-Things (IoT) architectures. As an emerging technology, DFL has spawned extensive applications in the IoT environment, such as intrusion detection, mobile robot localization, and location-based services. Current DFL-related machine learning (ML) algorithms still suffer from low localization accuracy and weak dependability/robustness because the group structure has not been considered in their location estimation, which leads to an undependable process. To overcome these challenges, we propose in this work a dependable block-sparse scheme by particularly considering the group structure of signals. An accurate and robust ML algorithm named block-sparse coding with the proximal operator (BSCPO) is proposed for DFL. In addition, a severe Gaussian noise is added in the original sensing signals for preserving network-related privacy as well as improving the dependability of the model. The real-world data-driven experimental results show that the proposed BSCPO achieves robust localization and signal-recovery performance even under severely noisy conditions and outperforms state-of-the-art DFL methods. For single-target localization, BSCPO retains high accuracy when the signal-to-noise ratio exceeds −10 dB. BSCPO is also able to localize accurately under most multitarget localization test cases.

A Hybrid Range-free Algorithm Using Dynamic Communication Range for Wireless Sensor Networks

Location plays a backbone role in networks, since it will great influence basic wireless sensor networks (WSNs) architecture. Distance-Vector Hop (DV-Hop) is a representative range-free localization algorithm, which is widely utilized to locate node position in location-based application. However, with poor localization accuracy, it cannot satisfy precise location-based application requirement. Consequently, we proposed a hybrid range-free algorithm depends on dynamic communication range to address low localization accuracy problem, named as DCDV-Hop. Firstly, we applied statistical methods to analyze the relationship between location error and hop count under different communication ranges. Thereafter, we employed centroid algorithm to calculate target node coordinate based on hop threshold. Finally, a weighted least square is applied to locate remaining target nodes. We conducted considerable experiments, the results demonstrated that our proposed algorithm DCDV-Hop can effectively reduce accumulate localization error and improve localization accuracy of target nodes, with stable performance. Moreover, maximum localization accuracy reached up to 91.35% and localization error reduced more than 50%, compared with DV-Hop algorithm.

An Efficient Indoor Wi-Fi Positioning Method Using Virtual Location of AP

Wi-Fi fingerprinting has been widely used for indoor localization because of its good cost-effectiveness. However, it suffers from relatively low localization accuracy and robustness owing to the signal fluctuations. Virtual Access Points (VAP) can effectively reduce the impact of signal fluctuation problem in Wi-Fi fingerprinting. Current techniques normally use the Log-Normal Shadowing Model to estimate the virtual location of the access point. This would lead to inaccurate location estimation due to the signal attenuation factor in the model, which is difficult to be determined. To overcome this challenge, in this study, we propose a novel approach to calculating the virtual location of the access points by using the Apollonius Circle theory, specifically the distance ratio, which can eliminate the attenuation parameter term in the original model. This is based on the assumption that neighboring locations share the same attenuation parameter corresponding to the signal attenuation caused by obstacles. We evaluated the proposed method in a laboratory building with three different kinds of scenes and 1194 test points in total. The experimental results show that the proposed approach can improve the accuracy and robustness of the Wi-Fi fingerprinting techniques and achieve state-of-art performance.

AGV Localization System Based on Ultra-Wideband and Vision Guidance

Aiming at the problems of low localization accuracy and complicated localization methods of the automatic guided vehicle (AGV) in the current automatic storage and transportation process, a combined localization method based on the ultra-wideband (UWB) and the visual guidance is proposed. Both the UWB localization method and the monocular vision localization method are applied to the indoor location of the AGV. According to the corner points of an ArUco code fixed on the AGV body, the monocular vision localization method can solve the pose information of the AGV by the PnP algorithm in real-time. As an auxiliary localization method, the UWB localization method is called to locate the AGV coordinates. The distance from the tag on the AGV body to the surrounding anchors is measured by the time of flight (TOF) ranging algorithm, and the actual coordinates of the AGV are calculated by the trilateral centroid localization algorithm. Then, the localization data of the UWB is corrected by the mean compensation method to obtain a consistent and accurate localization trajectory. The experiment result shows that this localization system has an error of 15mm, which meets the needs of AGV location in the process of automated storage and transportation.

A Node Location Algorithm Based on Node Movement Prediction in Underwater Acoustic Sensor Networks

Aiming at the problems of the low mobility, low location accuracy, high communication overhead, and high energy consumption of sensor nodes in underwater acoustic sensor networks, the MPL (movement prediction location) algorithm is proposed in this article. The algorithm is divided into two stages: mobile prediction and node location. In the node location phase, a TOA (time of arrival)-based ranging strategy is first proposed to reduce communication overhead and energy consumption. Then, after dimension reduction processing, the grey wolf optimizer (GWO) is used to find the optimal location of the secondary nodes with low location accuracy. Finally, the node location is obtained and the node movement prediction stage is entered. In coastal areas, the tidal phenomenon is the main factor leading to node movement; thus, a more practical node movement model is constructed by combining the tidal model with node stress. Therefore, in the movement prediction stage, the velocity and position of each time point in the prediction window are predicted according to the node movement model, and underwater location is then completed. Finally, the proposed MPL algorithm is simulated and analyzed; the simulation results show that the proposed MPL algorithm has higher localization performance compared with the LSLS, SLMP, and GA-SLMP algorithms. Additionally, the proposed MPL algorithm not only effectively reduces the network communication overhead and energy consumption, but also improves the network location coverage and node location accuracy.

A visual SLAM method based on point-line fusion in weak-matching scene

Visual simultaneous localization and mapping (SLAM) is well-known to be one of the research areas in robotics. There are many challenges in traditional point feature-based approaches, such as insufficient point features, motion jitter, and low localization accuracy in low-texture scenes, which reduce the performance of the algorithms. In this article, we propose an RGB-D SLAM system to handle these situations, which is named Point-Line Fusion (PLF)-SLAM. We utilize both points and line segments throughout the process of our work. Specifically, we present a new line segment extraction method to solve the overlap or branch problem of the line segments, and then a more rigorous screening mechanism is proposed in the line matching section. Instead of minimizing the reprojection error of points, we introduce the reprojection error based on points and lines to get a more accurate tracking pose. In addition, we come up with a solution to handle the jitter frame, which greatly improves tracking success rate and availability of the system. We thoroughly evaluate our system on the Technische Universität München (TUM) RGB-D benchmark and compare it with ORB-SLAM2, presumably the current state-of-the-art solution. The experiments show that our system has better accuracy and robustness compared to the ORB-SLAM2.

Research on robotic automatic machining for welding groove of complex integral impeller

Robots used in machining processes are more prone to inaccurate motion due to low location accuracy and asymmetrical structure. In this paper, a new method for improving robot motion precision was proposed to correct machining trajectory and improve stability of the robotic system on machining welding groove of complex integral impeller. First, the flange curves are extracted and equidistantly discretized; the corresponding discrete points on the central axis are obtained according to impeller 3D model. Then, according to the central point and symmetrical mapping theory, the corresponding discrete points of the central axis and the cover tray curve are obtained in turn, and the discrete points of the cover tray curve are smoothly connected to form the cover curve, which is the theoretical curve after processing. The distance between the cover curve and the flange curve is the machining allowance. Second, a unification of different coordinates of platform system is obtained based on coordinates transformation, and gravity compensation is accomplished to obtain the contact force. Third, the cutter tool keeps close contact with the flange curve along the feed motion and reads the actual position of the flange curve by in situ force/position control. The actual trajectory of the flange curve and the deviation between theoretical and actual trajectory was achieved. Finally, the actual cover tray curve is achieved by offsetting the machining allowance along the cutting depth with the actual trajectory. In this way, the positioning accuracy of the robot can be controlled by controlling the repetitive positioning accuracy, and the subsequent variable depth cutting can be realized. Experiments were performed to verify the benefits of correcting machining trajectory to achieve the machining of variable cutting depth on welding groove of blade and improve machining accuracy.

Localization and Data Collection in AUV-Aided Underwater Sensor Networks: Challenges and Opportunities

With the fast growing demand for underwater applications such as marine environmental monitoring, undersea resource exploration, disaster prevention and monitoring, assisted localization and navigation, and security monitoring, the IoUT is proposed to enable a new network framework to connect underwater smart things in rivers and oceans. Conventional UWSNs are perceived as the fundamental infrastructure of IoUT. However, the high cost of underwater devices, high energy consumption of data aggregation, and low localization accuracy limit their further development. AUV brings the mobility property into network designs, which improves localization accuracy, data transmission rate, and data aggregation efficiency. However, it also brings new challenges for localization, path planning and coordination of AUVs. In this article, we briefly introduce the architecture of AUV-aided UWSNs and summarize their advantages based on the current research. Several significant issues when designing localization algorithms and coordination schemes for AUV-aided UWSNs are investigated in detail. We also analyze the main challenges under different scenarios such as the interaction between AUV and sensor nodes and communications among multiple AUVs. Based on these discussions, we conclude the article with the future research directions of AUV-aided UWSNs.

Indoor Localization System Based on Hybrid Wi-Fi/BLE and Hierarchical Topological Fingerprinting Approach

Wi-Fi-based localization using received signal strength (RSS) with pedestrian dead reckoning (PDR) algorithm is widely used to track pedestrians in indoor environments. However, the unsatisfactory deployment of Wi-Fi access points (APs) in buildings and the unstable performance of PDR are still key problems that lead to low localization accuracy. In this paper, we make contributions on proposing a hybrid Wi-Fi and Bluetooth Low Energy (BLE) indoor localization system (ILS) based on an efficient BLE deployment strategy and hierarchical topological fingerprinting (HTF). For the BLE deployment strategy, we deploy BLE beacons in places that do not have clear Wi-Fi signals for localization. This efficiently increases the localization accuracy. For HTF, we hierarchically localize targets based on a topological fingerprint (TF) map. First of all, we quickly localize the room in which the target is located by Dendogram-based support vector machine (DSVM). Then, the specific position of the target is estimated by fusing Wi-Fi and BLE signals with the TF map. The new BLE-based fingerprinting algorithm is used to localize targets in environments sparsely populated by BLE beacons. We conduct physical experiments in a real building. The experimental results demonstrate that the beacons deployed based on our proposed deployment strategy results in greater localization accuracy. Furthermore, the HTF approach performs better than the other commonly used localization methods.

Three-Dimensional Localization Algorithm of WSN Nodes Based on RSSI-TOA and Single Mobile Anchor Node

Aimed at the shortcomings of low localization accuracy of the fixed multianchor method, a three-dimensional localization algorithm for wireless sensor network nodes is proposed in this paper, which combines received signal strength indicator (RSSI) and time of arrival (TOA) ranging information and single mobile anchor node. A mobile anchor node was introduced in the proposed three-dimensional localization algorithm for wireless sensor networks firstly, and the mobile anchor node moves according to the Gauss–Markov three-dimensional mobility model. Then, based on the idea of using RSSI ranging in the near end and TOA ranging in the far end, a ranging method combining RSSI and TOA ranging information is proposed to obtain the precise distance between the anchor node and the unknown node. Finally, the maximum-likelihood estimation method is used to estimate the position of unknown nodes based on the obtained ranging values. The MATLAB simulation results show that the proposed algorithm had a higher localization accuracy and lower localization energy consumption compared with the traditional RSSI localization method or TOA localization method.

An indoor Wi-Fi access points localization algorithm based on improved path loss model parameter calculation method and recursive partition

The localization accuracy of the existing methods for indoor Wi-Fi access points ranging-based localization depends on the accuracy of the received signal strength measured. Because the existing ranging-based methods are interfered by various indoor environmental factors, it is difficult to accurately measure the received signal strength, which leads to the problem of low localization accuracy of the indoor Wi-Fi access points. An indoor Wi-Fi access points localization algorithm based on improved path loss model parameter calculation method and recursive partition is proposed in this article. The algorithm recursively partitions the region where the target Wi-Fi access points are located according to the idea of quadtree partition, and partitions it into same sub-grids, which is sequentially performed until the sub-grids are smaller than the set threshold. The detection device is used at the detection location of the grids to measure the received signal strength, which is from the detection points to the target access point, the grid center point is used as the location of the candidate target access point, the parameters in the path loss model are calculated by using the signal strength differences between the detection points, and then the distances between the detection points and the target access point are calculated by using the signal strength values from the detection points to the target access point. Finally, the location of the target access point is estimated by executing a localization algorithm, and the location of the grid center point closest to the target access point is taken as the location of the target access point. The experimental results show that under the premise that the target access point can be found, the proposed algorithm reduces the use of the device and improves the localization accuracy compared with the typical localization method.

Robust WiFi Localization by Fusing Derivative Fingerprints of RSS and Multiple Classifiers

It is notable that localization accuracy using received signal strength (RSS) fingerprints solely is very vulnerable to dynamic environments. Utilizing multiple fingerprints gleaned from RSS for localization is a propitious strategy to overcome the RSS susceptibility. Brimful utilization via fusing multiple fingerprint functions which supplement each other are not harnessed by existing fusion-based techniques, resulting in low localization accuracy. This paper presents a novel and robust WiFi localization modus operandi by fusing DerIvative Fingerprints of RSS with MultIple Classifiers (DIFMIC). DIFMIC first constructs a multiple fingerprints group by gleaning hyperbolic location fingerprint (HLF) and signal strength differences fingerprint (DIFF) from RSS fingerprints. Then, it obtains Multiple Fingerprints Trained Classifiers (MFTCs) via training each basic classifier with each fingerprint. To fully leverage the inherent supplementation among fingerprints and classifiers, a two-layer fusion profile (weights) joint optimization algorithm with multiple constraints is proposed. We also propose a Fusion Profile Selection (FPS) algorithm to intelligently choose fusion weights from the two-layer fusion profile for a more accurate localization. DIFMIC shows more leverage in combining multiple information, thus exhibiting better robustness in WiFi positioning. Results from our experiments reflect that DIFMIC performs better than other existing methods in real environments.

An Energy Efficient Scheme using Mobility Prediction for Localization of Wireless Sensor Nodes

Location estimation of sensor nodes in Wireless Sensor Networks is very essentials because without information of location the information is meaningless. Most of the range-free algorithms have low localization accuracy, low cost, and applications limited to indoor uses only. This paper proposes a Mobility Prediction localization algorithm using the Link expiration time estimation method, this concept brings continuous link among the anchors and the mobile sensor nodes. This gives more accurate position estimation (3.2% of R) and employs fewer samples (average 20.58 per slot) for the task, hence results in less energy consumption than a Sequential Monte Carlo localization scheme. Both the algorithms are studied, analyzed and compared with speed of the mobile nodes in terms of localization error (6.38% to 6.55% better on different anchor density) communication cost (51.72% high), the number of samples taken per slot (average 58.8% less) and residual energy profile.

A DW-SAPSO Localization Method based on Correctional RSSI Ranging Model

Aiming at the problem of low localization accuracy caused by the ranging error generated by the RSSI ranging model with empirical parameters in different indoor scenarios, a simulated annealing-based particle swarm optimization localization algorithm with decreasing inertia weight (DW-SAPSO) is proposed. In the RSSI ranging phase, the parameters of ranging model are estimated and corrected by the difference between different reference points. In the localization phase, the nonlinear cost function is constructed according to the measurable quantity. Simultaneously the simulated annealing and decreasing inertia weight mechanism are introduced to the standard particle swarm optimization (PSO) localization algorithm, which effectively improves the algorithm’s global optimization ability and local search accuracy. Experimental results show that compared with an existing method and the PSO localization algorithm, the modified DW-SAPSO localization algorithm has better convergence performance and can improve the average localization accuracy of the latter by about 32.7%, which has some practical value.

PRDL: Relative Localization Method of RFID Tags via Phase and RSSI Based on Deep Learning

Ultra-high frequency radio frequency identification (UHF RFID) technology has been widely used in many areas, and RFID localization becomes a research hotspot. There are many kinds of research on absolute localization; however, due to some disadvantages of absolute localization, relative localization is more effective in some situations. At present, there are some problems with relative localization: existing methods have low localization accuracy, and it is difficult for them to deal with high-density tags. Aiming at these problems, this paper proposes PRDL: relative localization method of RFID tags via phase and RSSI based on deep learning. By using deep learning, the variation characteristics of RFID phase and RSSI are extracted with limited data accuracy conditions. On this basis, we can infer the relative positional relationship of RFID tags with high accuracy, and design the corresponding sorting algorithm to obtain the sequence arrangement. PRDL has experimented with bare tags and actual books, and the experimental results show that PRDL can achieve better results than the traditional relative localization methods. A series of tests also showed that PRDL has good robustness and generalization ability.

ITL-MEPOSA: Improved Trilateration Localization With Minimum Uncertainty Propagation and Optimized Selection of Anchor Nodes for Wireless Sensor Networks

For wireless sensor networks, sensing data without location information are useless for future processing and decision-making systems. A wireless localization method can provide critical position information for the sensing data in the coverage field. Trilateration localization is applied widely in many systems due to its low cost and widespread availability. However, there are many negative factors that lead to low localization accuracy. Many improved versions of trilateration localization, including the weighted trilateration localization, the trilateration centroid localization, and other related methods, have been proposed to improve localization accuracy. However, these methods do not consider the influence mechanism of the negative factors, and the localization results are unsatisfactory. In this paper, considering the uncertainty propagation mechanism during trilateration localization, we propose an improved trilateration localization method with minimum uncertainty propagation and optimized selection of anchor nodes (ITL-MEPOSA). In this method, we consider the uncertainty propagation from distance estimation to localization calculation and evaluate the uncertainty propagation of each distance estimation result. Then, we utilize a single scan-sliding window with an optimized algorithm to select anchor nodes with minimum uncertainty propagation. Finally, based on the selected anchor nodes and their corresponding distance estimation results, an accurate localization result can be obtained through the least square criterion. The simulations and experimental results show that the proposed method can obtain high localization accuracy with acceptable efficiency.

Data Fusion in Wireless Communication Network Node Positioning

Aiming at the problems of low localization accuracy, high energy consumption and delay of transmission in wireless sensor networks, a wireless sensor network data fusion method is proposed. Combined with the theory of D-S argument, the wireless sensor network system model is first constructed. Then based on the analysis of ACO (Ant Colony Optimization) and MST (Minimum Spanning Tree) methods, a wireless sensor network data fusion method is proposed. The simulation results show that the larger the number of nodes in WSN is, the larger the energy consumption and delay are. As the moving speed of Sink node increases, the average transmission delay decreases and the useful data rate increases. The average energy The consumption is also gradually increasing; the performance of proposed WSN data fusion method is better than that of ACO and MST.