Autonomous Solution Research Articles

A contextual account of digital servitization through autonomous solutions: Aligning a digital servitization process and a maritime service ecosystem transformation to autonomous shipping

This study focuses on digital servitization (DS) through autonomous solutions by building on a service ecosystems perspective. The rise of autonomous solutions exemplifies the ongoing digitalization and societal transformation and therefore integrative theoretical perspectives are needed to complement the dominant focal actor perspective in extant DS research. The study presents a longitudinal case of a solution provider's DS process to demonstrate how transformation towards autonomous shipping was driven in the maritime sector. An empirically enriched framework communicates DS process as aligned changes in value propositions, resource configurations and institutional arrangements within the service ecosystem. The study offers academic contributions and practical implications on managing DS through autonomous solutions as a strategic reorientation of a firm in the multi-level context of service ecosystem transformation.

Teleoperation for Urban Search and Rescue Applications

An important application of field robotics research is robotic assistance for search and rescue operations. The problem of robotic search and rescue requires techniques to map, navigate, and search unknown complex environments. In subterranean domains such as tunnels, caves, and underground urban environments these activities are made more difficult due to communication constraints and unavailability of global positioning systems. We present here Coordinated Robotics participation in the Urban Circuit of the Defense Advanced Research Projects Agency (DARPA) Subterranean Challenge which addresses these problems in the underground urban environment. Our Teleoperation strategy serves as a baseline approach by which to compare autonomous solutions. Our aim is to provide insight into our system design and our lessons learned from the competition.

COVID-rate: an automated framework for segmentation of COVID-19 lesions from chest CT images

Novel Coronavirus disease (COVID-19) is a highly contagious respiratory infection that has had devastating effects on the world. Recently, new COVID-19 variants are emerging making the situation more challenging and threatening. Evaluation and quantification of COVID-19 lung abnormalities based on chest Computed Tomography (CT) images can help determining the disease stage, efficiently allocating limited healthcare resources, and making informed treatment decisions. During pandemic era, however, visual assessment and quantification of COVID-19 lung lesions by expert radiologists become expensive and prone to error, which raises an urgent quest to develop practical autonomous solutions. In this context, first, the paper introduces an open-access COVID-19 CT segmentation dataset containing 433 CT images from 82 patients that have been annotated by an expert radiologist. Second, a Deep Neural Network (DNN)-based framework is proposed, referred to as the text {COVID-Rate}, that autonomously segments lung abnormalities associated with COVID-19 from chest CT images. Performance of the proposed text {COVID-Rate} framework is evaluated through several experiments based on the introduced and external datasets. Third, an unsupervised enhancement approach is introduced that can reduce the gap between the training set and test set and improve the model generalization. The enhanced results show a dice score of 0.8069 and specificity and sensitivity of 0.9969 and 0.8354, respectively. Furthermore, the results indicate that the text {COVID-Rate} model can efficiently segment COVID-19 lesions in both 2D CT images and whole lung volumes. Results on the external dataset illustrate generalization capabilities of the text {COVID-Rate} model to CT images obtained from a different scanner.

On the impact of load profile data on the optimization results of off-grid energy systems

Access to electricity via large scale power grids is seen as one of the solutions for a fully renewable power system. However, it remains a huge technical, economical, and geopolitical challenge. In the meantime, millions of people across the world have none or limited access to electricity and quite often to rely on autonomous solutions such as diesel generators. With the decreasing cost of renewable energy generation technologies in recent years, one could observe a simultaneous increase in studies dedicated to optimal sizing of renewable off-grid systems. Many of these studies rely on the usage of typical daily load profiles to model the electricity demand, sometimes enhanced with seasonal or random components. Such approaches tend to overlook the existing potential case-specific correlation between availability of renewable energy and energy demand and in particular the natural variability of the load in terms of its extreme values or ramp rates. The objective of this study is to investigate the impact of different types of load input data (for instance real load, monthly adjusted typical load, and typical daily load) on the cost of energy provided by off-grid PV-battery systems supplying various loads with different reliability levels. For this purpose, we determine the optimal capacity of PV-battery systems based on commonly used energy management strategies and optimization algorithms. The analysis of the obtained results indicates that, on average, using daily load profiles tends to underestimate the cost by 1.2% points (pp) for a system with 100% reliability and by over 5 pp for a system characterized by 95% reliability. Using monthly adjusted typical daily load profiles leads to slight differences compared to the results obtained by using real load as input. Although the obtained average values indicate a tendency of underestimating the energy cost, some outliers have been also observed reaching values of up to 15% of overestimating the cost of energy.

Design and implementation of a hybrid IoT system of data collection and visualization

To carry out a scientific study on a given problem (medical, environmental, etc.), it is often necessary to have recourse to a reliable dataset. In Africa, access to data is generally a problem for the scientific community. Most of the existing datasets are detained by some tiers organization who often do not give access to it. To overcome these problems related to access to data, scientists generally use traditional methods (surveys, on-site readings, etc.) for their study. However, these methods are not only expensive, but also require a regular visit to the site for updates. In this article, we propose an autonomous solution for collecting and analyzing data based on the Internet of Things (IoT). The solution can be used to collect data of different kinds. Unlike traditional data collection methods, the proposed solution is much cheaper and allows real time data collection. To demonstrate its proper functioning, a practical case on the environmental data collection will be studied in this paper.

On Supervisees’ Sense of Agency and Safety

To be alive and maintain an analytic stance amid transference–countertransference challenges, therapists need to experience both agency and safety. Their sense of safety increases when they connect with their internal analytic community’s representations and their sense of agency increases when they disengage from these representations. When therapists maintain the dialectical tension between these positions, they can choose a third position, to move between connecting and disengaging from the internal representations, in tune with the clinical context. Supervisors help their supervisees to contain this dialectical tension by manifesting their own sense of safety and agency when taking responsibility for their failures to understand the supervisees’ and the patients’ experiences. Furthermore, supervisors help foster the supervisees’ agency and safety separately. They help foster agency by finding autonomous and spontaneous solutions to clinical issues and help foster safety by drawing on the analytic community’s knowledge, to anchor the supervisees’ intuitive discoveries in a conceptual network, and clarify and stabilize the reality and boundaries in supervised therapies.

Adaptive Stretch-Forming Process: A Computer Vision and Statistical Analysis Approach

An industrial process is defined through its quality of parts and their production costs. Labour-intensive operations must be applied to produce high-quality components with inexpensive resources. Recent development in dedicated software allows the industrial sector to rely on more and more autonomous solutions to obtain an optimum ratio between part quality and cost. The stretch forming process is an operation that has a high degree of difficulty, due to the process parameters and the spring-back effect of materials. Our approach to solving several of the shortcomings of this process was to develop a self-adaptive algorithm with computer vision capabilities that adapts to the process in real-time. This experimental study highlights the results obtained using this method, as well as a comparison to a classical method for the stretch-forming process (SFP). The results have noted that the stretch-forming algorithm improves the process, while adapting its decisions with each step.

Towards Autonomous Drone Racing without GPU Using an OAK-D Smart Camera.

Recent advances have shown for the first time that it is possible to beat a human with an autonomous drone in a drone race. However, this solution relies heavily on external sensors, specifically on the use of a motion capture system. Thus, a truly autonomous solution demands performing computationally intensive tasks such as gate detection, drone localisation, and state estimation. To this end, other solutions rely on specialised hardware such as graphics processing units (GPUs) whose onboard hardware versions are not as powerful as those available for desktop and server computers. An alternative is to combine specialised hardware with smart sensors capable of processing specific tasks on the chip, alleviating the need for the onboard processor to perform these computations. Motivated by this, we present the initial results of adapting a novel smart camera, known as the OpenCV AI Kit or OAK-D, as part of a solution for the ADR running entirely on board. This smart camera performs neural inference on the chip that does not use a GPU. It can also perform depth estimation with a stereo rig and run neural network models using images from a 4K colour camera as the input. Additionally, seeking to limit the payload to 200 g, we present a new 3D-printed design of the camera’s back case, reducing the original weight 40%, thus enabling the drone to carry it in tandem with a host onboard computer, the Intel Stick compute, where we run a controller based on gate detection. The latter is performed with a neural model running on an OAK-D at an operation frequency of 40 Hz, enabling the drone to fly at a speed of 2 m/s. We deem these initial results promising toward the development of a truly autonomous solution that will run intensive computational tasks fully on board.

A robust fuzzy approach for gene expression data clustering

In the big data era, clustering is one of the most popular data mining method. Most clustering algorithms have complications like automatic cluster number determination, poor clustering precision, inconsistent clustering of various datasets and parameter-dependent, etc. A new fuzzy autonomous solution for clustering named Meskat-Mahmudul (MM) clustering algorithm was proposed to overcome the complexity of parameter-free automatic cluster number determination and clustering accuracy. The Meskat-Mahmudul clustering algorithm finds out the exact number of clusters based on the average silhouette method in multivariate mixed attribute dataset, including real-time gene expression dataset and missing values, noise, and outliers. Meskat-Mahmudul Extended K-Means (MMK) clustering algorithm enhances the K-Means algorithm, which serves the purpose of automatic cluster discovery and runtime cluster placement. Several validation methods are used to evaluate clusters and certify optimum cluster partitioning and perfection. Some datasets are used to assess the performance of the proposed algorithms to other algorithms in terms of time complexity and clustering efficiency. Finally, Meskat-Mahmudul clustering and Meskat-Mahmudul Extended K-Means clustering algorithms were found superior over conventional algorithms.

Digital Transformation of Transportation in the Age of COVID-19

Purpose: To highlight the key trends in the transportation industry and the role of emerging technologies and digital transformation in acceleration of value creation. The conceptual framework suggests how such transformation could be implemented. Design/methodology/approach: Research in a focus group was carried out where selected digital transformation frameworks were reviewed, and one was selected as the most applicable to the transportation industry. The selected framework was adopted for the transportation industry. Findings: 1. Value creation should be positioned as the key objective of transformation. 2. Digital transformation is not so much about technology as it is about people. 3. The staged approach towards transformation allows it to be paced with account being taken of the maturity of technology as well as the maturity of competencies across the organization. For a successful digital transformation, the organization invests in development of digital capabilities. Moving towards digital means a mindset shift, cultural change and adoption of new methods of working. Research limitations/implications: Transportation companies are at the early stage of transformation at such a scale. The case studies on successful transformations are limited. Therefore, experience is gathered mainly by the test and learn process. It is recommended that findings are validated with a wider group of experts when transformation matures. Comparing the same framework usage in other industries can also offer additional learnings to be considered by change leaders. Originality/value: Transportation industry is in a process of transformation towards connected, electric and autonomous solutions, accelerated during the COVID-19 pandemic. The presented framework, adopted for the transportation industry, offers a practical tool for executives and change leaders to lead the transformation.

Learning-‘N-Flying: A Learning-Based, Decentralized Mission-Aware UAS Collision Avoidance Scheme

Urban Air Mobility, the scenario where hundreds of manned and Unmanned Aircraft Systems (UASs) carry out a wide variety of missions (e.g., moving humans and goods within the city), is gaining acceptance as a transportation solution of the future. One of the key requirements for this to happen is safely managing the air traffic in these urban airspaces. Due to the expected density of the airspace, this requires fast autonomous solutions that can be deployed online. We propose Learning-‘N-Flying (LNF), a multi-UAS Collision Avoidance (CA) framework. It is decentralized, works on the fly, and allows autonomous Unmanned Aircraft System (UAS)s managed by different operators to safely carry out complex missions, represented using Signal Temporal Logic, in a shared airspace. We initially formulate the problem of predictive collision avoidance for two UASs as a mixed-integer linear program, and show that it is intractable to solve online. Instead, we first develop Learning-to-Fly (L2F) by combining (1) learning-based decision-making and (2) decentralized convex optimization-based control. LNF extends L2F to cases where there are more than two UASs on a collision path. Through extensive simulations, we show that our method can run online (computation time in the order of milliseconds) and under certain assumptions has failure rates of less than 1% in the worst case, improving to near 0% in more relaxed operations. We show the applicability of our scheme to a wide variety of settings through multiple case studies.

Inherent paradoxes in the shift to autonomous solutions provision: a multilevel investigation of the shipping industry

Transforming a traditional industry by adopting autonomous solutions is complex, generating paradoxical tensions on multiple aggregate levels. We undertake an in-depth case study of a leading maritime autonomous solutions provider and its ecosystem partners. We apply the paradox lens using thematic analysis. Our research contributes to the digital servitization literature by identifying six paradoxes inherent in the shift to autonomous solutions, nested in the micro, meso, and macro levels. We develop a multilevel framework of organizational paradoxes, delineating cascading effects of paradoxes across levels. We offer valuable insights for providers to integrate multilevel perspectives into the shift to autonomous solutions.

What law should apply then? The implications of the successful invocation of the public policy defence in EU private international law in family matters

The EU legislation in the area of private international law addresses explicitly the “negative” aspect of public policy, i.e. the non-application of the otherwise applicable law on the ground that it is at variance with the fundamental values of the forum. By contrast, the legislative measures adopted so far remain silent as to the law or rules that one should apply as a result of the successful invocation of the public policy defence. The paper aims, first, to assess the approach whereby the latter issue should be decided in accordance with the private international law rules of the forum. Secondly, the paper contends that an autonomous solution to the issue of the subsidiarily applicable law should mirror the goals pursued by the EU legislator – namely autonomy, flexibility, proximity and foreseeability –, and enshrined in the already adopted instruments dealing with the conflict of laws, rather than following the more widely known and endorsed approaches either not ensuring foreseeability and legal certainty, or leading to the immediate application of the lex fori. The focus will be on conflict-of-law rules in family matters, although similar patterns can be exported to other areas of the judicial cooperation in civil and commercial matters.

An integrated safety and security analysis for cyber-physical harm scenarios

Increasing digitalization and autonomous solutions in physical systems promise to enhance their performance, cost-efficiency and reliability. However, the integration of novel information technologies with safety-related systems also brings new vulnerabilities and risks that challenge the traditional field of safety analysis. Particularly, cyber security threats are becoming key factors in complex accident scenarios in cyber-physical systems (CPSs), where unintentional errors and design flaws overlap with cyber security vulnerabilities that could lead to harm to humans and assets. This overlap between safety and security analysis is still a loosely defined domain without established theories and methods, leading to complications during the risk analysis of CPSs. In this paper, we first describe how the domain of safety science increasingly overlaps with security analysis. Subsequently, based on this overlapping, we illustrate and complement an integrated method for the identification of harm scenarios in CPSs. This method, coined Uncontrolled Flows of Information and Energy (UFoI-E), offers a distinct theoretical foundation rooted in accident causation models and a framework to design diagrammatic representations of CPSs during the analysis. After summarizing these features of the UFoI-E method, we present our original contribution to the method, which is a new practical toolkit for risk identification composed of an ontology of harm scenarios and a database of checklists built from lessons learned analysis and expert knowledge. Finally, we demonstrate an application of the method in an illustrative case and show representative fields for future work.

Autonomic cloud computing based management and security solutions: State‐of‐the‐art, challenges, and opportunities

AbstractThe advancements and rapid adoption of service oriented architecture, utility computing, virtualization, etc., have emerged in a new computing paradigm called cloud computing, in the last few years. Cloud computing enables on‐demand delivery of data, software, and hardware in the form of services over the Internet. But the growing scale of cloud‐users and type‐of‐services urge for an autonomous environment to automate the highly complex cloud environment in a suitable way to improve the overall management of the cloud systems. To counter these automation related issues, autonomic computing (AC) emerges as a potential solution to optimize the performance of cloud computing. The merge of these two prominent technologies gives rise to a new technological domain called autonomic cloud computing (ACC). In this survey article, the existing AC assisted solutions for cloud computing is broadly categorized into two groups, namely features based and parameters based. Features based solutions are further classified into management (of services, workload, and resources) and security aware solutions. Similarly, parameters based solutions are grouped into quality‐of‐service aware and performance management solutions. Finally, the comparative analysis of state‐of‐the‐art ACC approaches with respect to the approach referred, performance metrics, outcome, and the derived inference is also presented to have an in‐depth understanding of the existing literature. Overall, the work excites the research community to design the effective autonomous solutions for ACC resources and services. Additionally, several research gaps and challenges are identified and discussed in this article based on the study with a focus on the future research directions.

Data Management System for a Semiautonomous Shuttle Car for Underground Room and Pillar Coal Mines

In recent years, autonomous solutions in the multidisciplinary field of mining engineering have been an extremely popular applied research topic. This is a result of the increasing demands of society on mineral resources along with the accelerating exploitation of the currently economically viable resources, which lead the mining sector to turn to deeper, more-difficult-to-mine orebodies. An appropriate data management system comprises a crucial aspect of the designing and the engineering of a system that involves autonomous or semiautonomous vehicles. The vast volume of data collected from onboard sensors, as well as from a potential IoT network dispersed around a smart mine, necessitates the development of a reliable data management strategy. Ideally, this strategy will allow for fast and asynchronous access to the data for real-time processing and decision-making purposes as well as for visualization through a corresponding human–machine interface. The proposed system has been developed for autonomous navigation of a coalmine shuttle car and has been implemented on a 1/6th scale shuttle car in a mock mine. It comprises three separate nodes, namely, a data collection node, a data management node, and a data processing and visualization node. This approach was dictated by the large amount of collected data and the need to ensure uninterrupted and fast data management and flow. The implementation of an SQL database server allows for asynchronous, real-time, and reliable data management, including data storage and retrieval. On the other hand, this approach introduces latencies between the data management node and the other two nodes. In general, these latencies include sensor latencies, network latencies, and processing latencies. However, the data processing and visualization module is able to retrieve and process the latest data and make a decision about the next optimal movement of the shuttle car prototype in less than 900 ms. This allows the prototype to navigate efficiently around the pillars without interruptions.

Accuracy Assessment of Atlas L-band GNSS Global Correction Service and Autonomous Positioning at Perlis, Malaysia

Differential Global Navigation Satellite System (DGNSS) is the most common positioning method used for navigation in the hydrography field. During the loss of the correction signal, the differential solution becomes an autonomous solution that may affect the accuracy of the position during that time. However, the availability of the Atlas L-band global correction service that adapts the Real-Time Precise Point Positioning (RT-PPP) technique has broadened the choice of solutions that can be used for navigation in the maritime industry and may solve the problem of signal loss. This research compares the positioning between autonomous solution GNSS and Atlas L-band correction solution using the static method to assess the accuracy of positioning between both methods. Data acquisition of the autonomous positioning and Atlas L-band service was conducted by using Hemisphere receiver VS330 and antenna A43. The statistical T-test reveals that the accuracy of analysis Atlas-L band and autonomous solution GNSS using static positioning was significant, as the p < 0.05 with 95% confidence interval. Besides, the result also shows that the position given by the Atlas L-band is more accurate and precise than Autonomous Solution GNSS, with an average position of 0.479 meters and 2.281 meters, respectively. Ultimately, the continuity of positioning data given by the Atlas L-band in the northern part of Malaysia is good, and positioning using Atlas L-band can be classified as Special Order based on the classification table by the International Hydrographic Organisation (IHO).
 Keywords: Atlas L-Band, GNSS, RT-PPP, Autonomous Positioning

Voltage-Based Droop Control of Electric Vehicles in Distribution Grids under Different Charging Power Levels

If left uncontrolled, electric vehicle charging poses severe challenges to distribution grid operation. Resulting issues are expected to be mitigated by charging control. In particular, voltage-based charging control, by relying only on the local measurements of voltage at the point of connection, provides an autonomous communication-free solution. The controller, attached to the charging equipment, compares the measured voltage to a reference voltage and adapts the charging power using a droop control characteristic. We present a systematic study of the voltage-based droop control method for electric vehicles to establish the usability of the method for all the currently available residential electric vehicle charging possibilities considering a wide range of electric vehicle penetrations. Voltage limits are evaluated according to the international standard EN50160, using long-term load flow simulations based on a real distribution grid topology and real load profiles. The results achieved show that the voltage-based droop controller is able to mitigate the under voltage problems completely in distribution grids in cases either deploying low charging power levels or exhibiting low penetration rates. For high charging rates and high penetrations, the control mechanism improves the overall voltage profile, but it does not remedy the under voltage problems completely. The evaluation also shows the controller’s ability to reduce the peak power at the transformer and indicates the impact it has on users due to the reduction in the average charging rates. The outcomes of the paper provide the distribution grid operators an insight on the voltage-based droop control mechanism for the future grid planning and investments.

A Deep Segmentation Network of Stent Structs Based on IoT for Interventional Cardiovascular Diagnosis

The Internet of Things (IoT) technology has been widely introduced to the existing medical system. An eHealth system based on IoT devices has gained widespread popularity. In this article, we propose an IoT eHealth framework to provide an autonomous solution for patients with interventional cardiovascular diseases. In this framework, wearable sensors are used to collect a patient's health data, which is daily monitored by a remote doctor. When the monitoring data is abnormal, the remote doctor will ask for image acquisition of the patient's cardiovascular internal conditions. We leverage edge computing to classify these training images by the local base classifier; thereafter, pseudo-labels are generated according to its output. Moreover, a deep segmentation network is leveraged for the segmentation of stent structs in intravascular optical coherence tomography and intravenous ultrasound images of patients. The experimental results demonstrate that remote and local doctors perform real-time visual communication to complete telesurgery. In the experiments, we adopt the U-net backbone with a pretrained SeResNet34 as the encoder to segment the stent structs. Meanwhile, a series of comparative experiments have been conducted to demonstrate the effectiveness of our method based on accuracy, sensitivity, Jaccard, and dice.

Progress Toward Zero Entry Mining: Automation Enabling Safer, More Efficient Mining

Around the world, there are some complex orebodies that necessitate truly manless operations to facilitate the safe economic extraction of ore in certain mining zones. This is called zero entry mining, because no human need enter the mining zone. Zero entry mining has the potential to significantly enhance the value of marginal or unrecoverable ore reserves and turn significant quantities of mineral resources into ore reserves. Australia is currently leading the charge in developing and deploying autonomous mining solutions and is now seeing benefits in productivity and safety. Currently, the autonomous solutions are applied to parts of the value chain (e.g., trucks and drills) and initially used original equipment manufacturer (OEM) proprietary approaches.