Big Geospatial Data Analytics, Connected Vehicle Technologies, and Visual Perception and Sensor Fusion Algorithms in Smart Transportation Networks.

• Author: Peters, Ellen

1. Introduction

   Spatial computing technology integrates augmented reality capabilities, predictive control algorithms, and image processing techniques. The purpose of my systematic review is to examine the recently published literature on smart transportation networks and integrate the insights it configures on big geospatial data analytics, connected vehicle technologies, and visual perception and sensor fusion algorithms. By analyzing the most recent (2019-2022) and significant (Web of Science, Scopus, and ProQuest) sources, my paper has attempted to prove that traffic management and analytics integrate autonomous vehicle perception sensor data, computer vision-based lane detection algorithms, mapping and navigation tools, and transportation analytics. The actuality and novelty of this study are articulated by addressing connected vehicle technologies, data fusion sensing and route planning algorithms, and traffic congestion monitoring systems (Bacalu, 2021; Konhausner et al, 2021), that is an emerging topic involving much interest. My research problem is whether self-driving safety and efficiency are developed on groundbreaking machine learning systems, decision-making software, and driving algorithms.

   In this review, prior findings have been cumulated indicating that deep reinforcement learning techniques and location tracking algorithms (Nica et al, 2022; Pelau et al, 2021; Vatamanescu et al, 2022) decrease traffic collisions and fatalities. The identified gaps advance deep learning-based sensing technologies, autonomous driving algorithms and behaviors, modeling and simulation tools, and real-time Internet of Things data. My main objective is to indicate that sensor data processing tools, self-driving car control algorithms, and virtual and augmented reality devices (Andronie et al, 2021a, b, c; Friedman et al., 2022) shape smart traffic analytics. This systematic review contributes to the literature on algorithm-driven sensing devices, monitoring and predictive capabilities, and mobility data processing tools (Balcerzak et al, 2022; Hackman and Reindl, 2022; Lazaroiu et al, 2017) by clarifying that cognitive artificial intelligence and collision avoidance algorithms, simulation and virtualization technologies, and geospatial mapping tools (Blake, 2022; Jenkins, 2022; Lazaroiu et al, 2020; Nica, 2017) further self-driving car acceptance and adoption.

2. Theoretical Overview of the Main Concepts

   Algorithm-driven sensing devices, crash avoidance optimization and mobile obstacle recognition tools, and intelligent traffic monitoring systems can shape autonomous vehicle adoption intention. Urban sensing technologies require deep learning-based computer vision algorithms, artificial intelligence-powered image recognition technology, and traffic flow prediction tools. The manuscript is organized as following: theoretical overview (section 2), methodology (section 3), deep learning-based computer vision algorithms, artificial intelligence-powered image recognition technology, and traffic flow prediction tools (section 4), sensor data processing tools, self-driving car control algorithms, and virtual and augmented reality devices (section 5), virtual navigation tools, spatio-temporal fusion and motion planning algorithms, and sensor and data processing technologies (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 March 2022, a quantitative literature review of the Web of Science, Scopus, and ProQuest databases was performed, with search terms including "smart transportation networks" + "big geospatial data analytics," "connected vehicle technologies," and "visual perception and sensor fusion algorithms." The search terms were determined as being the most employed words or phrases across the analyzed literature. As research published between 2019 and 2022 was inspected, only 95 articles satisfied the eligibility criteria. By taking out controversial or ambiguous findings (insufficient/irrelevant data), outcomes unsubstantiated by replication, too general material, or studies with nearly identical titles, I selected 14 mainly empirical sources (Tables 1 and 2). Extracting and inspecting publicly accessible files (scholarly sources) as evidence, before the research began no institutional ethics approval was required. Data visualization tools: Dimensions (bibliometric mapping) and VOSviewer (layout algorithms). Reporting quality assessment tool: PRISMA. Methodological quality assessment tools include: AMSTAR, Dedoose, Distiller SR, and SRDR (Figures 1-6).

4. Deep Learning-based Computer Vision Algorithms, Artificial Intelligence-powered Image Recognition Technology, and Traffic Flow Prediction Tools

   Connected vehicle technologies, data fusion sensing and route planning algorithms, and traffic congestion monitoring systems (Johnson and Nica, 2021; Li et al., 2021a) optimize navigation accuracy. Traffic management and analytics integrate autonomous vehicle perception sensor data, computer vision-based lane detection algorithms, mapping and navigation tools, and transportation analytics. Making expedient decisions at crossroads for driving safety, performance, and enjoyment can be optimized for connected vehicle technologies by use of deep reinforcement learning.

   Virtual simulation tools, spatial cognition algorithms, and monitoring and sensing technologies (Shannon et al., 2021; Zou et al., 2021) are pivotal in networked urban environments. Algorithm-driven sensing devices, crash avoidance optimization and mobile obstacle recognition tools, and intelligent traffic monitoring systems can shape autonomous vehicle adoption intention. The growing cyber connectivity of cars and between them and infrastructure will tremendously refashion mobility through deep neural networks. Sensor-based safety systems can reduce the main...