Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
6.8
3.9
Journals
Sensors
Special Issues
Trustworthy AI for Vehicle-to-Everything (V2X): Opportunities and Challenges
sensors-logo
Submit to Sensors Review for Sensors Propose a Special Issue

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Trustworthy AI for Vehicle-to-Everything (V2X): Opportunities and Challenges

A special issue of Sensors (ISSN 1424-8220). This special issue belongs to the section "Vehicular Sensing".

Deadline for manuscript submissions: closed (30 September 2022) | Viewed by 7280

Special Issue Editors

Dr. Ali Safaa Sadiq

E-Mail Website
Guest Editor
School of Mathematics and Computer Science, University of Wolverhampton, Wulfruna Street, Wolverhampton WV1 1LY, UK
Interests: wireless communication; AI applications in networking and security; IoT and IoV
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Amar Aggoun

E-Mail Website
Guest Editor
Department of Mathematics and Computer Science, University of Wolverhampton, Wolverhampton WV1 1LY, UK
Interests: 3D imaging; computer vision; optics
Prof. Dr. Prashant Pillai

E-Mail Website
Guest Editor
Wolverhampton Cyber Research Institute (WCRI), School of Mathematics and Computer Science, University of Wolverhampton, Wolverhampton WV1 1LF, UK
Interests: security and privacy for mobile/satellite networks; protocol development in heterogeneous networks; intelligent algorithms for wireless networks / mobile communications / IoT

Special Issue Information

Dear Colleagues,

Autonomous vehicles (AVs), or self-driving cars, have been the focus of many researchers and industries, with the goal of achieving a real-time decision-making solution for determining the right decision in the right time and place; though this does present a crucial question: could AI determine the ‘right’ decision on the road? In order to answer such a question, the autonomous intelligent systems in self-driving cars should be designed and tested in such a way that allows their decisions to be interpreted or explained. By doing so, humans/experts can be held accountable for their use by developing such trustworthy AI models for real-time decision making on roads for V2X to maintain a safe AV experience.

In a recent study (Petrović et al. 2020), 64.2% rear-end-type accidents were reported to have been witnessed to occur with AV, in comparison to 28.3% with conventional cars. Additionally, when proceeding in a straight line under the manoeuvres of the autonomous driver at an unsafe speed due to an error, AVs obtained 65.2% and 43.5%, respectively. Therefore, there is a rising demand for the development of a trustworthy AI model capable of handling such crucial scenarios.

According to a recent survey by the global analytics firm FICO and Corinium (San Jose, Calif, 2021), 65% of companies cannot explain how artificial intelligence (AI) model decisions/predictions are determined, and poor data have generated, on average, GBP 11.8 million/year of financial waste. Additionally, according to VentureBeat (2021), while over the last few years many companies have invested in AI applications, there has not been a corresponding investment of resources into AI governance.

Most companies, especially small and medium businesses, choose AI-as-a-Service to save costs, presenting the big challenge of how much trust should be given to an output produced by an AI-as-a-Service (AIaaS) platform? For instance, several studies have shown that currently used standard AI algorithms are not suitable for regulated financial services due to their lack of security, transparency, reliability and explainability. Assuring responsible AI models from the design/development phase is crucial for AI developers and data scientists, as having AI-based applications implemented responsibly is perceived differently by different businesses, some businesses considering AI to be responsible when it implies ethics, transparency and accountability, while other businesses, such as financial firms, hold different values towards responsible AI, such as laws, regulations and other customers' and organisational values.

Therefore, the vision of this Special Issue is mainly to target all research works related to the development of secure and trustworthy AI models/approaches/frameworks suitable for all AI developers and data scientists to help them check the quality and trustworthiness levels of their datasets and AI models. This Special Issue also looks for research/survey articles discussing different issues related to finding a method for regulating responsible, explainable and trustworthy AI modelling for decision-making applications.

Dr. Ali Safaa Sadiq
Prof. Dr. Amar Aggoun
Prof. Dr. Prashant Pillai
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Sensors is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • trustworthy AI
  • ethical AI
  • explainable AI
  • responsible AI
  • secure AI
  • trustworthy optimisation
  • quality of data for AI modelling
  • trusted AI for decision-making applications
  • safety and robustness in AI
  • non-discrimination, fairness, privacy, auditability and accountability in AI
  • environmental well-being assurance in AI

Published Papers (3 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

31 pages, 7984 KiB  
Article
V2ReID: Vision-Outlooker-Based Vehicle Re-Identification
by Yan Qian, Johan Barthelemy, Umair Iqbal and Pascal Perez
Sensors 2022, 22(22), 8651; https://doi.org/10.3390/s22228651 - 9 Nov 2022
Cited by 2 | Viewed by 2079
Abstract
With the increase of large camera networks around us, it is becoming more difficult to manually identify vehicles. Computer vision enables us to automate this task. More specifically, vehicle re-identification (ReID) aims to identify cars in a camera network with non-overlapping views. Images [...] Read more.
With the increase of large camera networks around us, it is becoming more difficult to manually identify vehicles. Computer vision enables us to automate this task. More specifically, vehicle re-identification (ReID) aims to identify cars in a camera network with non-overlapping views. Images captured of vehicles can undergo intense variations of appearance due to illumination, pose, or viewpoint. Furthermore, due to small inter-class similarities and large intra-class differences, feature learning is often enhanced with non-visual cues, such as the topology of camera networks and temporal information. These are, however, not always available or can be resource intensive for the model. Following the success of Transformer baselines in ReID, we propose for the first time an outlook-attention-based vehicle ReID framework using the Vision Outlooker as its backbone, which is able to encode finer-level features. We show that, without embedding any additional side information and using only the visual cues, we can achieve an 80.31% mAP and 97.13% R-1 on the VeRi-776 dataset. Besides documenting our research, this paper also aims to provide a comprehensive walkthrough of vehicle ReID. We aim to provide a starting point for individuals and organisations, as it is difficult to navigate through the myriad of complex research in this field. Full article
(This article belongs to the Special Issue Trustworthy AI for Vehicle-to-Everything (V2X): Opportunities and Challenges)
►▼ Show Figures

Figure 1

27 pages, 9636 KiB  
Article
Dynamic Learning Framework for Smooth-Aided Machine-Learning-Based Backbone Traffic Forecasts
by Mohamed Khalafalla Hassan, Sharifah Hafizah Syed Ariffin, N. Effiyana Ghazali, Mutaz Hamad, Mosab Hamdan, Monia Hamdi, Habib Hamam and Suleman Khan
Sensors 2022, 22(9), 3592; https://doi.org/10.3390/s22093592 - 9 May 2022
Cited by 5 | Viewed by 2307
Abstract
Recently, there has been an increasing need for new applications and services such as big data, blockchains, vehicle-to-everything (V2X), the Internet of things, 5G, and beyond. Therefore, to maintain quality of service (QoS), accurate network resource planning and forecasting are essential steps for [...] Read more.
Recently, there has been an increasing need for new applications and services such as big data, blockchains, vehicle-to-everything (V2X), the Internet of things, 5G, and beyond. Therefore, to maintain quality of service (QoS), accurate network resource planning and forecasting are essential steps for resource allocation. This study proposes a reliable hybrid dynamic bandwidth slice forecasting framework that combines the long short-term memory (LSTM) neural network and local smoothing methods to improve the network forecasting model. Moreover, the proposed framework can dynamically react to all the changes occurring in the data series. Backbone traffic was used to validate the proposed method. As a result, the forecasting accuracy improved significantly with the proposed framework and with minimal data loss from the smoothing process. The results showed that the hybrid moving average LSTM (MLSTM) achieved the most remarkable improvement in the training and testing forecasts, with 28% and 24% for long-term evolution (LTE) time series and with 35% and 32% for the multiprotocol label switching (MPLS) time series, respectively, while robust locally weighted scatter plot smoothing and LSTM (RLWLSTM) achieved the most significant improvement for upstream traffic with 45%; moreover, the dynamic learning framework achieved improvement percentages that can reach up to 100%. Full article
(This article belongs to the Special Issue Trustworthy AI for Vehicle-to-Everything (V2X): Opportunities and Challenges)
►▼ Show Figures

Figure 1

23 pages, 3761 KiB  
Article
A Hybrid Dragonfly Algorithm for Efficiency Optimization of Induction Motors
by Niraj Kumar Shukla, Rajeev Srivastava and Seyedali Mirjalili
Sensors 2022, 22(7), 2594; https://doi.org/10.3390/s22072594 - 28 Mar 2022
Cited by 7 | Viewed by 1989
Abstract
Induction motors tend to have better efficiency on rated conditions, but at partial load conditions, when these motors operate on rated flux, they exhibit lower efficiency. In such conditions, when these motors operate for a long duration, a lot of electricity gets consumed [...] Read more.
Induction motors tend to have better efficiency on rated conditions, but at partial load conditions, when these motors operate on rated flux, they exhibit lower efficiency. In such conditions, when these motors operate for a long duration, a lot of electricity gets consumed by the motors, due to which the computational cost as well as the total running cost of industrial plant increases. Squirrel-cage induction motors are widely used in industries due to their low cost, robustness, easy maintenance, and good power/mass relation all through their life cycle. A significant amount of electrical energy is consumed due to the large count of operational units worldwide; hence, even an enhancement in minute efficiency can direct considerable contributions within revenue saving, global electricity consumption, and other environmental facts. In order to improve the efficiency of induction motors, this research paper presents a novel contribution to maximizing the efficiency of induction motors. As such, a model of induction motor drive is taken, in which the proportional integral (PI) controller is tuned. The optimal tuning of gains of a PI controller such as proportional gain and integral gain is conducted. The tuning procedure in the controller is performed in such a condition that the efficiency of the induction motor should be maximum. Moreover, the optimization concept relies on the development of a new hybrid algorithm, the so-called Scrounger Strikes Levy-based dragonfly algorithm (SL-DA), that hybridizes the concept of dragonfly algorithm (DA) and group search optimization (GSO). The proposed algorithm is compared with particle swarm optimization (PSO) for verification. The analysis of efficiency, speed, torque, energy savings, and output power is validated, which confirms the superior performance of the suggested method over the comparative algorithms employed. Full article
(This article belongs to the Special Issue Trustworthy AI for Vehicle-to-Everything (V2X): Opportunities and Challenges)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-3
Back to TopTop