Real-world Experiments Research Articles

Filtered partial differential equations: a robust surrogate constraint in physics-informed deep learning framework

Embedding physical knowledge into neural network (NN) training has been a hot topic. However, when facing the complex real world, most of the existing methods still strongly rely on the quantity and quality of observation data. Furthermore, the NNs often struggle to converge when the solution to the real equation is very complex. Inspired by large eddy simulation in computational fluid dynamics, we propose an improved method based on filtering. We analysed the causes of the difficulties in physics-informed machine learning, and proposed a surrogate constraint (filtered partial differential equation, FPDE) of the original physical equations to reduce the influence of noisy and sparse observation data. In the noise and sparsity experiment, the proposed FPDE models (which are optimized by FPDE constraints) have better robustness than the conventional PDE models. Experiments demonstrate that the FPDE model can obtain the same quality solution with 100 % higher noise and 12 % quantity of observation data of the baseline. Besides, two groups of real measurement data are used to show the FPDE improvements in real cases. The final results show that the FPDE still gives more physically reasonable solutions when facing the incomplete equation problem and the extremely sparse and high-noise conditions. The proposed FPDE constraint is helpful for merging real-world experimental data into physics-informed training, and it works effectively in two real-world experiments: simulating cell movement in scratches and blood velocity in vessels.

Efficient Workflow Scheduling for Minimizing Data Transfers and Enhancing Resource Utilization in Cloud IaaS Platforms

Cloud IaaS platforms readily provide access to homogeneous multi-core machines, whether they are physical ("bare metal") or virtual machines. Each of these machines can be equipped with high-performance SSD disks, enabling the distribution of workflow-generated files across multiple machines, which helps minimize the overhead associated with data transfers. In this paper, we propose a scheduling algorithm called SMDT-ERU (Scheduling for Minimizing Data Transfer - Enhancing Resource Utilization), designed to reduce the makespan of data-intensive workflows by minimizing data transfers between dependent tasks over the network. Intermediate files generated by tasks are stored locally on the disk of the machine where the tasks are executed. Through experimentation, we confirm that increasing the number of cores per machine reduces the additional costs caused by network data transfers. Real-world workflow experiments demonstrate the advantages of the proposed algorithm. Our data-driven scheduling approach significantly reduces execution time and the volume of data transferred over the network, outperforming one of the leading state-of-the-art algorithms, which we have adapted to fit our assumptions.

Design and Evaluation of Steganographic Channels in Fifth-Generation New Radio

Mobile communication is ubiquitous in everyday life. The fifth generation of mobile networks (5G) introduced 5G New Radio as a radio access technology that meets current bandwidth, quality, and application requirements. Network steganographic channels that hide secret message transfers in an innocent carrier communication are a particular threat in mobile communications as these channels are often used for malware, ransomware, and data leakage. We systematically analyze the protocol stack of the 5G–air interface for its susceptibility to network steganography, addressing both storage and timing channels. To ensure large coverage, we apply hiding patterns that collect the essential ideas used to create steganographic channels. Based on the results of this analysis, we design and implement a network covert storage channel, exploiting reserved bits in the header of the Packet Data Convergence Protocol (PDCP). the covert sender and receiver are located in a 5G base station and mobile device, respectively. Furthermore, we sketch a timing channel based on a recent overshadowing attack. We evaluate our steganographic storage channel both in simulation and real-world experiments with respect to steganographic bandwidth, robustness, and stealthiness. Moreover, we discuss countermeasures. Our implementation demonstrates the feasibility of a covert channel in 5G New Radio and the possibility of achieving large steganographic bandwidth for broadband transmissions. We also demonstrate that the detection of the channel by a network analyzer is possible, limiting its scope to application scenarios where operators are unaware or ignorant of this threat.

Cooperative Object Transport Via Non-Contact Prehensile Pushing by Magnetic Forces

Abstract Cooperative robot systems are an essential candidate for object transportation solutions. They offer cost-efficient and flexible operation for various types of robotic tasks. The benefits of cooperative robot systems have triggered the improvement of the object transportation field. In this study, a new way of transporting objects by cooperative robots is presented. The proposed method is performed by the pushing action of the magnetic forces of the robots. The permanent magnets mounted on the mobile robots and the cart create this repelling force. The rectangular object carrier cart equipped with passive caster wheels can be manipulated on flat terrains easily and be assigned to carry different shapes of objects. Using a carrier cart has the advantage of eliminating the vertical loads on the robots. Controlling a non-contact pushing method offers a low computational burden since simple velocity and position updates are adequate for operation management. Compared with the other methods of object transportation systems, the non-contact pushing method provides a faster operation with less sensitivity to control errors. Both simulations and real-world experiments are conducted and the performances are given comparatively with a generalized frictional contact object-pushing method. The results show that the proposed method provides 10.48% faster and 20.03% more accurate object transportation compared to the frictional contact method. It is envisioned that the presented method can be a promising candidate for object transportation tasks in the industry.

A Unified Collision Avoidance Trajectory Planning with Dual Variables for Collaborative Aerial Transportation Systems

As an essential application of unmanned aerial vehicle (UAV) systems, payload transportation has garnered significant attention in recent years. Collaborative payload transportation utilizing multiple UAVs can effectively increase the payload capacity of the transportation system. Nevertheless, the incorporation of multiple UAVs makes the dynamic model of the transportation system more complex due to the coupled UAV and payload states. In the immediate disaster relief response, the collaborative system is often required to promptly deliver supplies to the target site while avoiding obstacles to ensure the system’s safety. Consequently, devising fast delivery trajectories that avoid collisions for such complicated systems poses a considerable challenge. To this end, a novel trajectory planning method is presented for collaborative transportation systems. Specifically, the dynamic model of the collaborative transportation system is derived by utilizing the Euler–Lagrange method. Then, the trajectory planning problem is formulated as an optimization problem with considerations of dynamics, actuation, safety, and formation constraints. To expedite the optimization process, the collision avoidance safety constraint is constructed using dual variables. The efficacy of this trajectory planning approach is confirmed through multiple real-world flight experiments involving collaborative aerial transportation systems of two and three UAVs.

Structured multi-view k-means clustering

K-means is a very efficient clustering method and many multi-view k-means clustering methods have been proposed for multi-view clustering during the past decade. However, since k-means have trouble uncovering clusters of varying sizes and densities, these methods suffer from the same performance issues as k-means. Improving the clustering performance of multi-view k-means has become a challenging problem. In this paper, we propose a new multi-view k-means clustering method that is able to uncover clusters in arbitrary sizes and densities. The new method simultaneously performs three tasks, i.e., sparse connection probability matrices learning, prototypes aligning, and cluster structure learning. We evaluate the proposed new method by 5 benchmark datasets and compare it with 11 multi-view clustering methods. The experimental results on both synthetic and real-world experiments show the superiority of our proposed method.

Acoustical features as knee health biomarkers: A critical analysis

Acoustical knee health assessment has long promised an alternative to clinically available medical imaging tools, but this modality has yet to be adopted in medical practice. The field is currently led by machine learning models processing acoustical features, which have presented promising diagnostic performances. However, these methods overlook the intricate multi-source nature of audio signals and the underlying mechanisms at play. By addressing this critical gap, the present paper introduces a novel causal framework for validating knee acoustical features. We argue that current machine learning methodologies for acoustical knee diagnosis lack the required assurances and thus cannot be used to classify acoustic features as biomarkers. Our framework establishes a set of essential theoretical guarantees necessary to validate this claim. We apply our methodology to three real-world experiments investigating the effect of researchers’ expectations, the experimental protocol, and the wearable employed sensor. We reveal latent issues such as underlying shortcut learning and performance inflation. This study is the first independent result reproduction study in acoustical knee health evaluation. We conclude by offering actionable insights that address key limitations, providing valuable guidance for future research in knee health acoustics.

In-migration for transforming peripheral locations through a real-world experiment? Experiment insights on location decisions in the medium-sized city of Görlitz, Germany

A real-world experiment was implemented in the city of Görlitz to gain knowledge about the requirements for, and the opportunities of, in-migration to revitalize urban areas and increase local capacities for sustainability transformation.Many cities, in particular those in peripheral locations, are dependent on in-migration to compensate for the negative population development often experienced due to demographic and economic change. In addition, the necessary transformation towards urban sustainability requires appropriate know-how and innovation potential. A real-world experiment to stimulate the revitalization of existing urban areas and to enhance the local potential for sustainability transformation through targeted in-migration was implemented in the city of Görlitz, Germany. The aim of this article is to reflect upon the impact of such an experiment with special regard to learning effects for individuals, municipalities, and transdisciplinary collaborations. It explains that there is great value for the future development of a city in the long-term consideration and co-productive work of various stakeholders, but this is also associated with challenges.

Deep Learning-Based Emergency Rescue Positioning Technology Using Matching-Map Images

Smartphone-based location estimation technology is becoming increasingly important across various fields. Accurate location estimation plays a critical role in life-saving efforts during emergency rescue situations, where rapid response is essential. Traditional methods such as GPS often face limitations in indoors or in densely built environments, where signals may be obstructed or reflected, leading to inaccuracies. Similarly, fingerprinting-based methods rely heavily on existing infrastructure and exhibit signal variability, making them less reliable in dynamic, real-world conditions. In this study, we analyzed the strengths and weaknesses of different types of wireless signal data and proposed a new deep learning-based method for location estimation that comprehensively integrates these data sources. The core of our research is the introduction of a ‘matching-map image’ conversion technique that efficiently integrates LTE, WiFi, and BLE signals. These generated matching-map images were applied to a deep learning model, enabling highly accurate and stable location estimates even in challenging emergency rescue situations. In real-world experiments, our method, utilizing multi-source data, achieved a positioning success rate of 85.27%, which meets the US FCC’s E911 standards for location accuracy and reliability across various conditions and environments. This makes the proposed approach particularly well-suited for emergency applications, where both accuracy and speed are critical.

Chained Spatial Beam Adomian Decomposition Model: A Novel Model of Flexible Slender Beams for Large Spatial Deflections

Abstract The main element of compliant mechanisms and continuum robots is flexible slender beams. However, the modeling of beams can be complicated due to the geometric nonlinearity becoming significant at large elastic deflections. This paper presents an explicit nonlinear model called the spatial beam Adomian decomposition model (SBADM) for intermediate spatial deflections of a slender beam with uniform, bisymmetric sections subjected to general end-loading. Specifically, the elongation, bending, torsion, and shear deformations of the beams are modeled based on Timoshenko's assumptions and Cosserat rod theory. Then, the quaternion transformation and Adomian decomposition are used to solve the nonlinear governing differential equations for the beam by truncating the higher-order terms, yielding an explicit expression for spatially deflected beams within intermediate deflection ranges. Simulations demonstrate the accuracy and time-wise efficiency of the SBADM, as well as its advantages over the state-of-the-art. In addition, this paper also introduces a discretization-based scheme called the chained SBADM (CSBADM) for large spatial deflections of flexible beams. Real-world experiments with two different configurations have also been performed to validate the effectiveness of the CSBADM. The results indicate that the CSBADM can accurately calculate the load-displacement relations for large deformed beams.

Experimental and numerical investigation of heat transfer characteristics of radiator using Graphene Amine-based nano coolant

This work presents experimental and numerical investigations on enhancing the heat transfer performance of automobile radiators with nano-coolant flowing with different mass flow rate and at different inlet temperatures conditions. Thermal management is critical as it dictates a system's overall power output, performance, efficiency and energy utilisation. Radiators and Thermal management systems for batteries are vital in keeping automobiles operating under optimal conditions. Conventional coolants lack the ability to keep up with the increasing performance requirements of thermal management systems. Nanocoolants represent a cutting-edge advancement in heat transfer fluid, boosting conductivity without compromising essential fluid dynamics of the cooling solutions. Several investigators have worked on various nanocoolants, but less work is done on radiators. In view of this, experimental and numerical investigations were done on radiator heat transfer performance using Graphene Amine based nanocoolant with varied coolant temperatures (50 °C–80 °C) and flow rates (3 l/min to 9 l/min). Thermophysical properties of nanocoolants, such as density, thermal conductivity, viscosity, specific heat, and Prandtl number, were determined by experimental and mathematical modelling techniques and were comparable with an average deviation of 5 %. Compared to the base fluids of De-ionised water and ethylene glycol combinations, Graphene Amine based nanocoolant has exhibited advancements in heat transfer properties, showing elevated Nusselt numbers, heat transfer coefficients and heat transfer rates. The investigation into the thermal transfer properties of nanocoolants utilizing theoretical and real-world experiments shows consistency with an average deviation of 20 %. Maximum heat transfer enhancement of 154.3 % and maximum heat transfer coefficient of 3209.52 W/m2K was found for Graphene Amine at 80 °C with coolant flowing at 9 l/min, which is 83.1 % higher when compared to De-ionised Water.

Real-world federated learning in radiology: hurdles to overcome and benefits to gain.

Federated Learning (FL) enables collaborative model training while keeping data locally. Currently, most FL studies in radiology are conducted in simulated environments due to numerous hurdles impeding its translation into practice. The few existing real-world FL initiatives rarely communicate specific measures taken to overcome these hurdles. To bridge this significant knowledge gap, we propose a comprehensive guide for real-world FL in radiology. Minding efforts to implement real-world FL, there is a lack of comprehensive assessments comparing FL to less complex alternatives in challenging real-world settings, which we address through extensive benchmarking. We developed our own FL infrastructure within the German Radiological Cooperative Network (RACOON) and demonstrated its functionality by training FL models on lung pathology segmentation tasks across six university hospitals. Insights gained while establishing our FL initiative and running the extensive benchmark experiments were compiled and categorized into the guide. The proposed guide outlines essential steps, identified hurdles, and implemented solutions for establishing successful FL initiatives conducting real-world experiments. Our experimental results prove the practical relevance of our guide and show that FL outperforms less complex alternatives in all evaluation scenarios. Our findings justify the efforts required to translate FL into real-world applications by demonstrating advantageous performance over alternative approaches. Additionally, they emphasize the importance of strategic organization, robust management of distributed data and infrastructure in real-world settings. With the proposed guide, we are aiming to aid future FL researchers in circumventing pitfalls and accelerating translation of FL into radiological applications.

Living labs as orchestrators in the regional innovation ecosystem: a conceptual framework

ABSTRACT This research explores the conceptual integration of Living Labs (LLs) into the Regional Innovation Ecosystem (RIE) to understand their potential as facilitators of Responsible Innovation. While previous studies explored the role of Living Labs, they lack comprehensive integration strategies for addressing wicked problems. A conceptual framework is developed to understand the main role, detailed functions, and key elements for integration. Seven expert interviews were conducted to develop the conceptual framework, using the Interuniversity Microelectronics Centre (imec) within the Flemish Innovation Ecosystem as empirical context. The findings emphasise Living Labs as central orchestrators, enabling user-centric innovation, real-world experimentation, and stakeholder engagement. We identify four key elements that facilitate the proposed integration: government policy and funding, strategic integration, formal collaborations, and proof of concept. These findings have implications for advancing the theoretical understanding of Living Labs' integration into the Regional Innovation Ecosystem, and for practitioners who aim to foster Responsible Innovation.

Prediction of feed force with machine learning algorithms in boring of AISI P20 plastic mold steel

Feed force plays a critical role in machining performance, including efficiency, hole quality, energy consumption, and tool life. While feed force is typically measured using dynamometers, these devices are often impractical in real-world experiments due to their complexity and cost. In particular, boring operations, which are essential for enlarging holes within tight tolerances, depend on precise feed force control to maintain high-quality hole outcomes. Achieving this precision requires reliable pre-process feed force estimation. However, traditional analytical methods for calculating feed force in boring operations often fall short, largely due to factors such as in-hole dynamics, the inclination of the rake surface, and the slenderness of the boring bar. This study aims to apply single and hybrid machine learning methods to accurately model and predict feed force in the boring process. Unlike previous research, which focused on feature selection, we propose the use of all possible combinations of input feature subsets to determine the optimal input features. The results indicate that feed force can be predicted effectively using these optimized feature combinations. Such accurate feed force prediction is crucial for effective process optimization, material selection, and process planning.

Identifying Tampered Radio-Frequency Transmissions in LoRa Networks Using Machine Learning.

Long-range networks, renowned for their long-range, low-power communication capabilities, form the backbone of many Internet of Things systems, enabling efficient and reliable data transmission. However, detecting tampered frequency signals poses a considerable challenge due to the vulnerability of LoRa devices to radio-frequency interference and signal manipulation, which can undermine both data integrity and security. This paper presents an innovative method for identifying tampered radio frequency transmissions by employing five sophisticated anomaly detection algorithms-Local Outlier Factor, Isolation Forest, Variational Autoencoder, traditional Autoencoder, and Principal Component Analysis within the framework of a LoRa-based Internet of Things network structure. The novelty of this work lies in applying image-based tampered frequency techniques with these algorithms, offering a new perspective on securing LoRa transmissions. We generated a dataset of over 26,000 images derived from real-world experiments with both normal and manipulated frequency signals by splitting video recordings of LoRa transmissions into frames to thoroughly assess the performance of each algorithm. Our results demonstrate that Local Outlier Factor achieved the highest accuracy of 97.78%, followed by Variational Autoencoder, traditional Autoencoder and Principal Component Analysis at 97.27%, and Isolation Forest at 84.49%. These findings highlight the effectiveness of these methods in detecting tampered frequencies, underscoring their potential for enhancing the reliability and security of LoRa networks.

Inverse Kinematics of Robotic Manipulators Based on Hybrid Differential Evolution and Jacobian Pseudoinverse Approach

Robot manipulators play a critical role in several industrial applications by providing high precision and accuracy. To perform these tasks, manipulator robots require the effective computation of inverse kinematics. Conventional methods to solve IK often encounter significant challenges, such as singularities, non-linear equations, and poor generalization across different robotic configurations. In this work, we propose a novel approach to solve the inverse kinematics (IK) problem in robotic manipulators using a metaheuristic algorithm enhanced with a Jacobian step. Our method overcomes those limitations by selectively applying the Jacobian step to the differential evolution (DE) algorithm. The effectiveness and versatility of the proposed approach are demonstrated through simulations and real-world experimentation on a 5 DOF KUKA robotic arm.

Continuous planning for inertial-aided systems

Inertial-aided systems require continuous motion excitation among other reasons to characterize the measurement biases that will enable accurate integration required for localization frameworks. This paper proposes the use of informative path planning to find the best trajectory for minimizing the uncertainty of IMU biases and an adaptive traces method to guide the planner towards trajectories that aid convergence. The key contribution is a novel regression method based on Gaussian Process (GP) to enforce continuity and differentiability between waypoints from a variant of the RRT∗\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\hbox {RRT}^*$$\\end{document} planning algorithm. We employ linear operators applied to the GP kernel function to infer not only continuous position trajectories, but also velocities and accelerations. The use of linear functionals enable velocity and acceleration constraints given by the IMU measurements to be imposed on the position GP model. The results from both simulation and real-world experiments show that planning for IMU bias convergence helps minimize localization errors in state estimation frameworks.

EMSA-IK: a real-time evolutionary multi-objective semi-analytical inverse kinematics algorithm for redundant manipulators

Purpose This paper aims to propose a real-time evolutionary multi-objective semi-analytical inverse kinematics (IK) algorithm (EMSA-IK) for solving the multi-objective IK of redundant manipulators. Design/methodology/approach Within EMSA-IK, the parameterization method is applied to reduce the number of optimization variables of the evolutionary algorithm and calculate semi-analytical solutions that meet high target pose accuracy. The original evolutionary algorithm is improved with the proposed adaptive search sub-space strategy so that the improved evolutionary algorithm can be used to efficiently perform global search within the parametric joint space to obtain the global optimal parametric joint angles that satisfy multi-objective constraints. Findings Ablation experiments show the effectiveness of the improved strategy used for evolutionary algorithms. Comparative experiments on different manipulators demonstrate the advantages of EMSA-IK in terms of generalizability and balancing multiple objectives, for example, motion continuity, joint limits and obstacle avoidance. Real-world experiments further validate the effectiveness of the proposed algorithm for real-time application. Originality/value The semi-analytical IK solution that simultaneously satisfies high target pose accuracy and multi-objective constraints can be obtained in real time. Compared to existing semi-analytical IK algorithms, the proposed algorithm achieves obstacle avoidance for the first time. The proposed algorithm demonstrates superior generalizability, applicable to not only redundant manipulators with revolute joints but also those with prismatic joints.

Enhancing network robustness with structural prior and evolutionary techniques

Robustness optimization in complex networks is a critical research area due to its implications for the reliability and stability of various systems. However, existing algorithms encounter two key challenges: the lack of integration of prior network knowledge, leading to suboptimal solutions, and high computational costs, which hinder their practical application. To address these challenges, this paper introduces Eff-R-Net, an efficient evolutionary algorithm framework aimed at enhancing the robustness of complex networks through accelerated evolution. Eff-R-Net leverages global and local network information, featuring a novel three-part composite crossover operator. Prior network knowledge is incorporated in mutation and local search operators to expedite the construction of networks with superior robustness. Additionally, a simplified method for calculating robustness enhances efficiency, while adaptive hyper-parameters dynamically adjust operators execution probabilities for optimal evolution. Extensive evaluations on both synthetic (Scale-Free, Erdös-Rényi, and Small-World) and three infrastructure real-world networks demonstrate the superiority of Eff-R-Net. The algorithm improves robustness by 12.8% and reduces computational time by 25.4% compared to state-of-the-art algorithm in real-world network experiments. These findings underscore Eff-R-Net's versatility and potential in enhancing network robustness across different domains.

A comprehensive framework for 5G indoor localization

Localization inside legacy private 5G networks is a daunting task that involves solving the problem of indoor localization using commercial off-the-shelf proprietary hardware. While some previous work has focused on experimental analysis, none has undertaken to develop a realistic solution based on commercial equipment. In this study, we present the first comprehensive and concrete 5G framework that combines fingerprinting with the 3GPP Enhanced Cell ID (E-CID) approach. Our methodology consists of a machine-learning model to deduce the user’s position by comparing the signal strength received from the User Equipment (UE) with a reference radio power map. To achieve this, the 3GPP protocols and functions are improved to provide open, centralized, and universal localization functions. A new reference map paradigm named Optical Radio Power Estimation using Light Analysis (ORPELA) is introduced. Real-world experiments prove that it is reproducible and more accurate than state-of-the-art radio-planning software. Machine-learning models are then designed, trained, and optimized for an ultra-challenging radio context. Finally, a large-scale experimental campaign encompassing a wide range of cases, including line-of-sight or mobility, is being conducted to demonstrate expected location performance within realistic 5G private networks.