Ethical Artificial Intelligence in Smart Mobility Technologies: Autonomous Driving Algorithms, Geospatial Data Mining Tools, and Ambient Sound Recognition Software.

• Author: Balica, Raluca-Stefania

1. Introduction

   Sensor data processing algorithms, deep learning-based routing decisions, and computational object instantiation tools reduce traffic road congestion and car crashes. The purpose of our systematic review is to examine the recently published literature on ethical artificial intelligence in smart mobility technologies and integrate the insights it configures on autonomous driving algorithms, geospatial data mining tools, and ambient sound recognition software. By analyzing the most recent (2022) and significant (Web of Science, Scopus, and ProQuest) sources, our paper has attempted to prove that self-driving car control algorithms integrate traffic sensing and geospatial mapping technologies, urban mobility data analytics, and blockchain-enabled Internet of Things networks. The actuality and novelty of this study are articulated by addressing data modeling and mobility simulation tools, machine learning and spatial cognition algorithms (Kliestik et al., 2020; Nica et al., 2022; Vatamanescu et al., 2020), and vehicular communication technologies (Andronie et al., 2021; Lazaroiu et al., 2022; Podosek et al., 2022), that is an emerging topic involving much interest. Our research problem is whether Smart transportation networks deploy route planning and visual perception algorithms (Kliestik et al., 2022; Pelau et al., 2021; Zvarikova et al., 2022), predictive urban analytics, and cognitive data fusion techniques..

   In this review, prior findings have been cumulated indicating that self-driving cars necessitate monitoring and sensing technologies (Balica, 2022; Lyons and Lazaroiu, 2020; Poliak et al., 2021), deep reinforcement learning algorithms, and predictive simulation tools. The identified gaps advance object detection and autonomous driving algorithms, cloud computing and 3D virtual simulation technologies, and geospatial data mining tools. Our main objective is to indicate that signal processing and spatial simulation algorithms, data-driven planning technologies (Blake and Frajtova Michalikova, 2021; Nagy and Lazaroiu, 2022; Pop et al., 2022), and computational object recognition tools further sustainable urban monitoring and intelligent transportation systems. This systematic review contributes to the literature on data simulation and visualization tools (Frajtova Michalikova et al., 2022; Nica, 2021; Popescu et al., 2020), algorithm-driven sensing devices, and crash avoidance technologies by clarifying that cognitive wireless sensor networks, computer vision and predictive modeling algorithms, and data mining and behavior tracking tools (Gasparin and Schinckus, 2022; Nica et al., 2021; Popescu et al., 2021) shape data-driven smart sustainable urbanism.

2. Theoretical Overview of the Main Concepts

   Deep learning-based ambient sound processing, spatial data visualization tools, and connected vehicle technologies diminish highway congestion and collision probability. Computer vision algorithms, vehicle and pedestrian detection tools, and image recognition software decrease crashes and casualties. Geospatial data mining tools, sensor-based perception systems, and computer vision algorithms assist autonomous vehicle navigation and smart transportation networks. The manuscript is organized as following: theoretical overview (section 2), methodology (section 3), autonomous vehicle steering algorithms, multi-object detection and tracking tools, and real-time Internet of Things data (section 4), smart spatial planning tools, motion planning and sensor data processing algorithms, and multimodal transportation analytics (section 5), multi-sensor environment data fusion tools, cognitive computing systems, and object localization and autonomous driving perception algorithms (section 6), discussion (section 7), synopsis of the main research outcomes (section 8), conclusions (section 9), limitations, implications, and further directions of research (section 10).

3. Methodology

   Throughout June 2022, we performed a quantitative literature review of the Web of Science, Scopus, and ProQuest databases, with search terms including "ethical artificial intelligence" + "smart mobility technologies" + "autonomous driving algorithms," "geospatial data mining tools," and "ambient sound recognition software." As we inspected research published in 2022, only 179 articles satisfied the eligibility criteria. By eliminating controversial findings, outcomes unsubstantiated by replication, too imprecise material, or having similar titles, I decided upon 36, generally empirical, sources (Tables 1 and 2). Data visualization tools: Dimensions (bibliometric mapping) and VOSviewer (layout algorithms). Reporting quality assessment tool: PRISMA. Methodological quality assessment tools include: AXIS, Dedoose, ROBIS, and SRDR (Figures 1-6).

   Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines were used that ensure the literature review is comprehensive, transparent, and replicable. The flow diagram, produced by employing a Shiny app, presents the stream of evidence-based collected and processed data through the various steps of a systematic review, designing the amount of identified, included, and removed records, and the justifications for exclusions.

   To ensure compliance with PRISMA guidelines, a citation software was used, and at each stage the inclusion or exclusion of articles was tracked by use of custom spreadsheet. Justification for the removal of ineligible articles was specified during the full-text screening and final selection.

   Figure 6 Screening and quality assessment tools To ensure first-rate standard of evidence, a systematic search of relevant databases including peer-reviewed published journal articles was conducted using predefined search terms, covering a range of research methods and data sources. Reference lists of all relevant sources were manually reviewed for additional relevant citations. [down arrow] Titles of papers and abstracts were screened for suitability and selected full texts were retrieved to establish whether they satisfied the inclusion criteria. All records from each database were evaluated by using data extraction forms. Data covering research aims, participants, study design, and method of each paper were extracted. [down arrow] The inclusion criteria were: (i) articles included in the Web of Science, Scopus, and ProQuest databases, (ii) publication date (2022), (iii) written in English, (iv) being an original empirical research or review article, and (v) particular search terms covered; (i) conference proceedings, (ii) books, and (iii) editorial materials were eliminated from the analysis. [down arrow] Distiller SR screened and extracted the collected data. [down arrow] SRDR gathered, handled, and analyzed the data for the systematic review, being configured as an archive and tool harnessed in data extraction through transparent, efficient, and reliable quantitative techniques. Elaborate extraction forms can be set up, meeting the needs of research questions and study designs. [down arrow] AMSTAR evaluated the methodological quality of systematic reviews. [down arrow] ROBIS assessed the risk of bias in systematic reviews. [down arrow] The quality of academic articles was determined and risk of bias was measured by MMAT, that tested content validity and usability of selected studies in terms of screening questions, type of design, corresponding quality criteria, and overall quality score. [down arrow] AXIS evaluated the quality of cross-sectional studies. [down arrow] Dedoose analyzed qualitative and mixed methods research. Note: Table made from Figure. 4. Autonomous Vehicle Steering Algorithms, Multi-Object Detection and Tracking Tools, and Real-Time Internet of Things Data

   Data modeling and mobility simulation tools, machine learning and spatial cognition algorithms, and vehicular communication technologies (Aguiar et al., 2022; Lee et al., 2022; Tian et al., 2022; Wang et al., 2022) configure intelligent traffic monitoring systems. Urban big data analytics leverages real-time data monitoring and visual recognition tools, multisensor fusion algorithms, and deep learning-based autonomous driving technologies. Object recognition algorithms, environment and vehicle sensors, and vehicular communication technologies are pivotal in smart mobility planning and transportation systems across urban network infrastructures.

   ...