Efficient Mapping Research Articles

A Characteristic Mapping Method for the three-dimensional incompressible Euler equations

We propose an efficient semi-Lagrangian Characteristic Mapping (CM) method for solving the three-dimensional (3D) incompressible Euler equations. This method evolves advected quantities by discretizing the flow map associated with the velocity field. Using the properties of the Lie group of volume preserving diffeomorphisms SDiff, long-time deformations are computed from a composition of short-time submaps which can be accurately evolved on coarse grids. This method is a fundamental extension to the CM method for two-dimensional incompressible Euler equations [1]. We take a geometric approach in the 3D case where the vorticity is not a scalar advected quantity, but can be computed as a differential 2-form through the pullback of the initial condition by the characteristic map. This formulation is based on the Kelvin circulation theorem and gives point-wise a Lagrangian description of the vorticity field. We demonstrate through numerical experiments the validity of the method and show that energy is not dissipated through artificial viscosity and small scales of the solution are preserved. We provide error estimates and numerical convergence tests showing that the method is globally third-order accurate.

LiDAR Road-Atlas: An Efficient Map Representation for General 3D Urban Environment

In this work, we propose the LiDAR Road-Atlas, a compact and efficient 3D map representation, for autonomous robot or vehicle navigation in a general urban environment. The LiDAR Road-Atlas can be generated by an online mapping framework which incrementally merges local 2D occupancy grid maps (2D-OGMs). Specifically, the contributions of our method are threefold. First, we solve the challenging problem of creating local 2D-OGMs in nonstructured urban scenes based on a real-time delimitation of traversable and curb regions in a LiDAR point cloud. Second, we achieve accurate 3D mapping in multiple-layer urban road scenarios by a probabilistic fusion scheme. Third, we achieve a very efficient 3D map representation of a general environment thanks to the automatic local-OGM-induced traversable-region labeling and a sparse probabilistic local point-cloud encoding. Given the LiDAR Road-Atlas, one can achieve accurate vehicle localization, path planning, and some other tasks. Our map representation is insensitive to dynamic objects which can be filtered out in the resulting map based on a probabilistic fusion. Empirically, we compare our map representation with a couple of popular map representations in robotics society, and our map representation is more favorable in terms of efficiency, scalability, and compactness. Additionally, we also evaluate localization performance given the LiDAR Road-Atlas representations on two public datasets. With a 16-channel LiDAR sensor, our method achieves an average global localization error of 0.26 m (translation) and 1.07 (rotation) on the Apollo dataset, and 0.89 m (translation) and 1.29 (rotation) on the MulRan dataset, respectively, at 10 Hz, which validates its promising performance. The code for this work is open-sourced at https://github.com/IMRL/Lidar-road-atlas.

Supervised versus Semi-Supervised Urban Functional Area Prediction: Uncertainty, Robustness and Sensitivity

To characterize a community-scale urban functional area using geo-tagged data and available land-use information, several supervised and semi-supervised models are presented and evaluated in Hong Kong for comparing their uncertainty, robustness and sensitivity. The following results are noted: (i) As the training set size grows, models’ accuracies are improved, particularly for multi-layer perceptron (MLP) or random forest (RF). The graph convolutional network (GCN) (MLP or RF) model reveals top accuracy when the proportion of training samples is less (greater) than 10% of the total number of functional areas; (ii) With a large amount of training samples, MLP shows the highest prediction accuracy and good performances in cross-validation, but less stability on same training sets; (iii) With a small amount of training samples, GCN provides viable results, by incorporating the auxiliary information provided by the proposed semantic linkages, which is meaningful in real-world predictions; (iv) When the training samples are less than 10%, one should be cautious using MLP to test the optimal epoch for obtaining the best accuracy, due to its model overfitting problem. The above insights could support efficient and scalable urban functional area mapping, even with insufficient land-use information (e.g., covering only ~20% of Beijing in the case study).

A framework for efficient soil architecture mapping using ground-penetrating radar

Ground-penetrating radar (GPR) is more and more prevailing in soil survey as a geophysical method due to its non-invasive and high efficiency. For subsurface soil architecture mapping, its quality relies not upon only the accuracy of GPR evaluation itself but also on often overlooked survey strategy. This study focuses on the latter one and proposed a new survey strategy to promote the commonly-used survey strategy that subjectively employing multiple parallel two-dimensional GPR acquisition lines. A preliminary survey roughly capturing the spatial variability of soil architecture is suggested to provide two key parameters, the spacing and the direction of the parallel survey lines, for the subsequent commonly-used survey design. Thus, the final survey can not only take advantage of the dense space sampling rate of GPR itself along the survey direction with the highest variability of the subsurface soil architecture, but also well controls the mapping quality and efficiency. Field experiments and numerical tests demonstrate that the new survey strategy could effectively reduce mapping uncertainties originating from unmatching sampling rate and interpolation algorithm selection. Besides, a suitable mapping quality and efficiency can be guaranteed but preventing unnecessary too dense sampling rate, namely low mapping efficiency. This indicates the proposed strategy would be useful for relevant soil surveys using geophysical methods.

Design of a Quasi-Direct Drive Actuator with Embedded Pulley for a Compact, Lightweight, and High-Bandwidth Exosuit

Although exosuits have several advantages compared to exoskeleton type of wearable robots, they have limitations, such as bulkiness and low control performance. This study addresses the design and evaluation of a compact, lightweight, and highly responsive actuator to be used for exosuits, based on the Quasi-Direct Drive (QDD) actuation. The design requirements of the actuator were set based on the actuation system used in the state-of-the-art exosuit from Harvard University (HE) so that it could be an improvement compared to HE. Several design concepts were comparatively evaluated to select the optimal design, and a design for the pulley embedded QDD (PEQDD) actuator was selected. The PEQDD was fabricated using mechanical components selected based on the design constraints or designed through mechanical analysis. Using a dynamometer, the efficiency map of the PEQDD was drawn. The control bandwidth comparison test with the motor originally used for HE showed improved bandwidth from 6.25 Hz to 20 Hz. Preliminary testing was done in walking and running conditions using an exosuit utilizing PEQDD. The test results showed that the actuator performance met all the design requirements.

Efficient 2D Mapping for Reversible Data Hiding

After the prediction, how to modify the 2D prediction error histogram (PEH) for reversible data hiding (RDH) is an important issue. In this letter, we propose an efficient 2D mapping strategy by considering the error pairs on two diagonal lines of the 2D PEH. The proposed 2D mapping can quickly adapt to the frequencies of the error pairs for different mapping area sizes. As a bonus, a satisfactory embedding performance can be achieved since in each group of two symmetrical error pairs the one with larger frequencies has priority to select more mapping ways. Experimental results have shown that it is low time-consuming and achieves better embedding performance in comparison with several classical and advanced 2D mappings.

Resource Allocation Based on Digital Twin-Enabled Federated Learning Framework in Heterogeneous Cellular Network

Federated learning (FL) allows user devices (UDs) to upload local model parameters to participate in a global model training, which protects UDs' data privacy. Nevertheless, FL still faces challenges such as core network congestion, UDs' limited resources and less efficient mapping between devices and cyber systems. Therefore, in this article, we integrate the digital twin (DT) and the mobile edge computing (MEC) technologies into a hierarchical FL framework in the heterogeneous cellular network scenario. When the UDs are not in the service range of the small base stations (SBSs), the framework allows macro base stations to assist UDs' local computation, thus reducing the transmission delay. It also protects the user privacy and allows more users to join in the training in order to improve the FL accuracy. In addition, we propose a deep reinforcement learning-based scheme to solve the joint optimization problem of dynamic UDs-stations association and resource allocation, thereby minimizing the energy consumption within a limited time delay. Simulation results show that our proposed scheme not only effectively reduces the task transmission failure rate and energy consumption compared with the baseline scheme, but also saves the communication cost through the DT network.

A neural network-based data-driven local modeling of spotwelded plates under impact

Solving large structural problems with multiple complex localized behaviors is extremely challenging. To address this difficulty, both intrusive and non-intrusive Domain Decomposition Methods (DDM) have been developed in the past, where the refined model (local) is solved separately in its own space and time scales. In this work, the Finite Element Method (FEM) at the local scale is replaced with a data-driven Reduced Order Model (ROM) to further decrease computational time. The reduced model aims to create a low-cost, accurate and efficient mapping from interface velocities to interface forces and enable the prediction of their time evolution. The present work proposes a modeling technique based on the Physics-Guided Architecture of Neural Networks (PGANNs), which incorporates physical variables other than input/output variables into the neural network architecture. We develop this approach on a 2D plate with a hole as well as a 3D case with spot-welded plates undergoing fast deformation, representing nonlinear elastoplasticity problems. Neural networks are trained using simulation data generated by explicit dynamic FEM solvers. The PGANN results are in good agreement with the FEM solutions for both test cases, including those in the training dataset as well as the unseen dataset, given the loading type is present in the training set.

Efficient tone mapping via macro and micro information enhancement and color correction

Efficient tone mapping via macro and micro information enhancement and color correction

The CTH Network: An NoC Platform for Scalable and Energy Efficient Application Mapping Solution

Multi-threaded design of the applications has given a new dimension to the execution concurrency. The Network-on-Chip (NoC) infrastructure has evolved as the communication backbone in multiprocessor System-on-Chips (MPSoCs) to exploit Thread-Level Parallelism (TLP). Mapping the threads of various applications on the processing cores is crucial in utilizing the resources of a many-core system. An application's performance depends primarily on the mapping strategy. However, the performance of an application gets influenced significantly by the underlying network architecture of the NoC. The present work proposes a run-time application mapping technique based on the Cube-Tree-Hybrid (CTH) topology focusing on the traffic load balancing across the network. The mapping overhead decreased significantly by exploiting the low network diameter of CTH with its high scalability and path diversity, projecting a 31% reduction in normalized execution time over the prevalent algorithms on the existing mapping platforms. Rigorous experimentation with varying application and network sizes is carried out across mesh, ZMesh, torus, and Mest-of-Tree (MoT) networks. The experimental observations project that the proposed algorithm results in sustainable application performance with 15%-21% gain in the overall energy consumption over the existing state-of-the-art techniques across various benchmark workloads.

Design and Analysis of an Asymmetric Spoke and Delta-Shape Interior Permanent Magnet Synchronous Machine

This paper proposes a design of an asymmetric spoke and delta-shape interior permanent magnet (AIPM-P) synchronous machine for electric vehicles that use the magnetic-field-shifting (MFS) technique to improve the average torque and reduce torque ripples. The asymmetry of the AIPM-P is due to the asymmetric rotor structure, asymmetric permanent magnet (PM) placement, and asymmetric flux barriers (AFB). Due to this asymmetry, the MFS is utilized, which decreases the current angle difference between the peak PM torque ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$T_{PM}$ </tex-math></inline-formula> ) and reluctance torque ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$T_{r}$ </tex-math></inline-formula> ) components. In the delta-shape, the bar magnet helps reduce the d-axis inductance, which helps in increasing the reluctance torque, ultimately increasing synthetic torque. Furthermore, the combination of delta-shape and spoke shape PM makes the novel hybrid layer structure that improves the average torque. The proposed AIPM-P is then compared with the conventional designs in the literature having the same stator and rotor radius, with a 4-pole and 24-slot combination. The open circuit, on-load characteristics, efficiency, copper, and iron loss maps of the proposed and conventional designs are investigated and compared. Compared to the conventional designs, the proposed AIPM-P synchronous machine has high efficiency and a good torque profile, which is suitable for EV applications.

Large-Scale and Efficient Texture Mapping Algorithm via Loopy Belief Propagation

Texture mapping as a fundamental task in 3D modeling has been well established for well-acquired aerial assets under consistent illumination, yet it remains a challenge when it is scaled to large datasets with images under varying views and illuminations. A well-performed texture mapping algorithm must be able to efficiently select views, fuse and map textures from these views to mesh models, at the same time, achieve consistent radiometry over the entire model. Existing approaches achieve efficiency either by limiting the number of images to one view per face, or simplifying global inferences to only achieve local color consistency. In this paper, we break this tie by proposing a novel and efficient texture mapping framework that allows the use of multiple views of texture per face, at the same time to achieve global color consistency. The proposed method leverages a loopy belief propagation algorithm to perform an efficient and global-level probabilistic inferences to rank candidate views per face, which enables face-level multi-view texture fusion and blending. The texture fusion algorithm, being non-parametric, brings another advantage over typical parametric post color correction methods, due to its improved robustness to non-linear illumination differences. The experiments on three different types of datasets (i.e. satellite dataset, unmanned-aerial vehicle dataset and close-range dataset) show that the proposed method has produced visually pleasant and texturally consistent results in all scenarios, with an added advantage of consuming less running time as compared to the state of the art methods, especially for large-scale dataset such as satellite-derived models.

High-Dynamic-Range 3-D Shape Measurement With Adaptive Speckle Projection Through Segmentation-Based Mapping

The 3-D shape measurement of high-dynamic-range (HDR) scenes based on structured light projection is challenging because the conventional projection without considering the reflection distribution of the tested surface usually causes over-exposure and over-dark areas simultaneously in the captured images, which will lead to the failure or low accuracy of 3-D reconstruction. In this article, we propose a novel HDR 3-D surface measurement method based on adaptive speckle projection with fully considering the reflection distribution. An efficient segmentation-based mapping strategy is established to speed up the generation of adaptive speckle patterns, which segments the estimated mask with optimal projection intensity into multiple subregions by the eight-connected standard to raise mapping accuracy and project gray-coded and sinusoidal fringe patterns to build the pixelwise mapping relationship between binocular cameras and the projector for each subregion, respectively. Then the pixelwise intensity of projected speckle patterns can be accurately adjusted according to optimal intensity. The generated adaptive speckle pattern is used to scan the tested object, and high signal-to-noise-ratio (SNR) speckle images can be obtained for accurate and complete 3-D measurement. Experimental results verify that the point cloud produced by our method has obvious advantages compared with the traditional methods in terms of measurement accuracy, whose coverage is as large as 92.26% and the average distance (AD) relative to ground truth (GT) is as low as 0.128 mm. Meanwhile, the number of projected patterns for mapping is decreased by a third compared with the widely used mapping strategy.

XLA-NDP: Efficient Scheduling and Code Generation for Deep Learning Model Training on Near-Data Processing Memory

Deep learning (DL) model training must address the memory bottleneck to continue scaling. Processing-in-memory approaches can be a viable solution as they move computations near or into the memory, reducing substantial data movement. However, to deploy applications on such hardware, end-to-end software support is crucial for efficient computation mapping and scheduling as well as extensible code generation, but no consideration has been made for DL training workloads. In this paper, we propose XLA-NDP, a compiler and runtime solution for NDPX, a near-data processing (NDP) architecture integrated with an existing DL training framework. XLA-NDP offloads NDPX kernels and schedules them to overlap with GPU kernels to maximize parallelism based on GPU and NDPX costs, while providing a template-based code generator with low-level optimizations. The experiments showed that XLA-NDP provides up to 1.41x speedup (1.24x on average) over the GPU baseline for four DL model training.

Mapping of soil properties using geographic information systems (on the example of Tashkent region)

The article analyzes the options for mapping soil properties using GIS systems. The perspectives on the use of GIS in the efficient mapping of soil mechanical content, agrochemical, and salinization characteristics are given. The study is conducted for irrigated sierozem soils of the Parkent district of Tashkent region. The basis of the research is the analysis of soil map data of the study areas, and the generalization of the results of soil-cartographic, laboratory, and cameral-analytical studies, which were conducted by commonly use methodological instructions. As a result of the research, digital thematic maps were created for the selected areas, such as various thematic maps, such as soil mechanical composition, soil nutrient status, and salinity maps.

Discrete Mixed-Integer Shooting (DMIS): Algorithm and Application to Plug-In Hybrid Electric Vehicle Energy Management Accounting for Fuel Cranking and Actual Powertrain Efficiency Maps

This article presents an algorithm for mixed-integer nonlinear optimal control problems (OCPs) formulated directly in discrete time. In our current application, we address the energy management of a plug-in hybrid electric vehicle (PHEV) seeking to achieve minimum trip fuel consumption under a terminal state-of-charge (SOC) equality constraint while avoiding powertrain efficiency map approximations in order to maximize its fuel economy (FE). An additional engine cranking state with a nonsmooth state and control transition cost is then incorporated to prevent undesired engine on/off behaviors. We pursue a single shooting-based numerical algorithm, motivated by the success of the computationally efficient Pontryagin’s Minimum Principle (PMP)-based single shooting in achieving similar FE results as dynamic programming (DP) for the power-split optimization problem without the additional engine cranking state shown in prior work. The Discrete Mixed-Integer Shooting (DMIS) algorithm, with costate backward-in-time propagation, is presented here, avoiding unstable shooting iterations that are typically encountered in PMP-based single shooting. When applying DMIS to solve the power-split problem with cranking fuel consideration, we show through simulation that it leads to better FE results than the PMP-based single shooting without the cranking state, thus demonstrating the effectiveness of the DMIS algorithm. Finally, experimental dynamometer test results with the model predictive powertrain controller solved by DMIS for minimum trip fuel consumption are presented and discussed, demonstrating its real-time execution and the cranking fuel reduction capability.

Efficient Approach for Braille Conversion of Multilingual Text: Strategic Mapping Solutions

In the today’s modern society, people can easily access information from various sources like the internet and newspapers. However, individuals with visual impairments encounter difficulties in harnessing this wealth of information, unlike their sighted counterparts who effortlessly stay informed about current events and knowledge. To bridge this disparity,there is a vital necessity to develop a system that streamlines the conversion of natural language text into Braille, thereby offering enhanced learning prospects for the visually impaired.Visually impaired people face challenges in benefiting from this wealth of information, unlike sighted individuals who can effortlessly stay updated with day-to-day news and knowledge. To bridge this gap, there is a crucial need to develop a system that facilitates the conversion of natural language text into Braille, thereby providing enhanced learning opportunities for the visually impaired. This research paper presents strategic mapping solutions for Braille conversion, with the primary aim of optimizing the translation process of multilingual text into Braille characters. The proposed approach incorporates the techniques from natural language processing (NLP) and address the challenges associated with converting diverse languages into the tactile writing system of Braille. The focus is on designing an efficient mapping system that takes into consideration the linguistic characteristics of different languages, ensuring accurate and contextually meaningful representation of Braille.

Metaheuristic Based Clustering with Deep Learning Model for Big Data Classification

Recently, a massive quantity of data is being produced from a distinct number of sources and the size of the daily created on the Internet has crossed two Exabytes. At the same time, clustering is one of the efficient techniques for mining big data to extract the useful and hidden patterns that exist in it. Density-based clustering techniques have gained significant attention owing to the fact that it helps to effectively recognize complex patterns in spatial dataset. Big data clustering is a trivial process owing to the increasing quantity of data which can be solved by the use of Map Reduce tool. With this motivation, this paper presents an efficient Map Reduce based hybrid density based clustering and classification algorithm for big data analytics (MR-HDBCC). The proposed MR-HDBCC technique is executed on Map Reduce tool for handling the big data. In addition, the MR-HDBCC technique involves three distinct processes namely pre-processing, clustering, and classification. The proposed model utilizes the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) technique which is capable of detecting random shapes and diverse clusters with noisy data. For improving the performance of the DBSCAN technique, a hybrid model using cockroach swarm optimization (CSO) algorithm is developed for the exploration of the search space and determine the optimal parameters for density based clustering. Finally, bidirectional gated recurrent neural network (BGRNN) is employed for the classification of big data. The experimental validation of the proposed MR-HDBCC technique takes place using the benchmark dataset and the simulation outcomes demonstrate the promising performance of the proposed model interms of different measures.

A Less-Disturbed Ecological Driving Strategy for Connected and Automated Vehicles

This paper proposes a less-disturbed ecological driving strategy for connected and automated vehicles (CAVs). The proposed strategy integrates the offline planning and the online tracking. In offline planning, an energy efficient reference speed is created based on traffic information (such as the average traffic speed) and characteristics of the vehicle (such as the engine efficiency map) via dynamic programming. The consideration of average traffic speed in speed planning avoids selfish optimisations. In online tracking, model predictive control is employed to update the vehicle speed in real-time to track the reference speed. A key challenge in applying ecological driving strategies in real driving is that the vehicle has to consider other traffic participants when tracking the reference speed. Therefore, this paper combines both longitudinal control and lateral control to achieve better speed tracking by overtaking the preceding vehicle when necessary. The proposed less-disturbed ecological driving strategy has been evaluated in simulations in both single road segment scenario and real traffic environment. Comparisons of the proposed method with benchmark strategies and human drivers are made. The results demonstrate that the proposed less-disturbed ecological driving strategy is more effective in energy saving. Compared to human drivers, the less-disturbed eco-driving strategy improves the fuel efficiency of CAVs by 4.53%.

Utilizing ISA-95 in an Industrial Knowledge Graph for Material Flow Simulation - Semantic Model Extensions and Efficient Data Integration

Despite access to emerging technologies for effective data representation and interoperability between different types of data, many enterprises struggle to use the information and knowledge available in manufacturing data to increase the efficiency and quality of their production. This is especially true for knowledge graphs and semantic web technologies. Nevertheless, breaking down the rigid and historical manufacturing data systems and transferring them to knowledge graphs often involves a lot of manual work for cleaning, mapping, and transforming the data. To address this issue, we have developed a methodology to efficiently utilize ISA-95 in knowledge graphs for manufacturing. The methodology is based on two main building blocks. First, we built an ontology based on the ISA-95 standard and extended it with additional semantic concepts. The second building block is an extract, transform, and load pipeline, to allow an efficient mapping of different data formats into the concepts specified by the ontology, and consequently a fast data integration into the knowledge graph. The created knowledge graph serves as basis for efficient management of material flow simulations.