In congested locations, a game-theory-based priority control strategy for designing successful crossings for autonomous automobiles (Yu et al., 2020). Ordinary and autonomous vehicles pass by roadside equipment, which share data and make decisions for the management of their passing traffic (RSU). The priority control algorithm ensures that automobiles do not collide at junctions. The VISSIM simulation tool was used to evaluate the algorithms and compare the results of each model to the regular traffic control at the intersection. The simulation looked at three distinct traffic volume and speed combinations (Zhu & Ukkusuri, 2015). The study's primary goal was to eliminate delays in these situations. When combined with maximum pressure control, the autonomous intersection management approaches known as AIM-ped will produce the total throughput (Shrestha et al., 2020). As a result, in order to estimate the calculations for the best vehicle trajectory, the simulation of this technique will provide current trajectory optimization and maximum pressure control. The intersection efficiency is evaluated using a simulation approach based on pedestrian demand. Because pedestrian and vehicle delays are negatively associated, this algorithm will alter the conflict trajectory between pedestrian demand and vehicle mobility (Martinez et al., 2010). When demand for both automobiles and pedestrians grows, the algorithm can adjust to meet the demand by allowing cars to pass through the crossroads on opposing paths. Autonomous vehicles have revolutionized and improved the transportation system, according to (Li & Liu, 2020). Ordinary automobiles will be phased out in favour of advanced self-driving vehicles that can make decisions while driving. These automobiles have sensors fitted so that they can sense their surroundings utilizing 5G technology, which is one of the vehicle's innovations. By ignoring a set of criteria such as safety, traffic management, comfort, energy, and speed, vehicles have been steered to more intelligent behaviours (Wu et al., 2015). An appropriate option is to combine a number of severe and correlative sensors that work together to overcome their individual deficiencies. This article provides a thorough examination of the best-in-class strategies for exhibiting AV frameworks in short-reach or close automobile situations. It focuses on late investigations that use sensor combination computations based on profound learning for insight, limitation, and planning (Hampshire et al., 2020). Self-driving automobile innovation has been steadily increasing in owing largely to breakthroughs in deep learning and artificial intelligence (AI). The goal of this research is to give a high-level summary of recent deep learning advancements in autonomous driving. We'll start with AI-based self-driving models, convolutional and recurrent neural networks, and the deep support learning perspective. The algorithms researched in the fields of driving scene perception, route planning, behaviour arbitration, and movement control are based on these methodologies (Horowitz & Varaiya, 2000). The measured perception planning action pipeline, in which each module is developed using deep learning algorithms, and End2End frameworks, which directly transfer sensory information to control commands, are both examined in this study. Furthermore, we address current issues in AI structure planning for these smart autonomous driving vehicles. For these autonomous, the correlation offered an overview that aids in gaining awareness of the qualities and limitations of deep learning and AI technologies (Grigorescu et al., 2020).
Before autonomous vehicles can be widely implemented, vehicle makers must solve a key moral challenge: algorithmic ethical quality (Dai et al., 2016). Autonomous cars, like other unavoidable disasters, are presented with basic dynamic scenarios in which at least two or more lives are in danger. In such cases, moral dilemmas arise, such as whether to sacrifice one's life in order to reduce the number of deaths. An algorithm is utilized to mimic this programming philosophy for self-driving cars. Minheap sorting employs a priority queue (Gao et al., 2020). Algorithms are used to categorize people's lives according to specific criteria. This method prioritizes between nodes with different attributes by using deletion, insertion, and other operations. A priority queue is utilized for BST as a critical line. The phrase "probability of endurance" is also used in this computation to describe the chances of an individual surviving. Clients and car manufacturers can design moral choices based on their own values thanks to the introduction of an adaptive and fragile algorithm (Fayyad et al., 2020).
The recent increase of both provincial and metropolitan street traffic streams has resulted in numerous transportation mistakes, such as traffic congestion, mishaps, and excessive levels of pollution (Ji & Hong, 2019). When standard traffic lights fail or when there are persistent traffic problems at street crossings, elective traffic light methods are necessary. The purpose of this study is to determine how successful and efficient artificial intelligence (AI) solutions such as the artificial neural network (ANN) are at removing or reducing traffic volume generated by non-independent vehicles in a blended South African rush hour jam stream. Electronic traffic data for 126 vehicles was provided by Micros Traffic Monitoring (MTM), a South African subsidiary of the Syntell Group of Companies (Hancock et al., 2019). The traffic data was collected using MTM's traffic innovations, which are basically sensors installed on street surfaces to filter and control cars that pass by the traffic counter on a daily basis (Bengler et al., 2014). By using the description of vehicle class and associated speed as information variables, the dataset acquired by MTM was constructed, tested, and verified using an artificial neural network model at signalized street crossing sites under heterogeneous conditions (Hancock et al., 2019). After a series of tests, the results show that the ANN model produced the best traffic congestion results in a heterogeneous rush hour gridlock situation. In our ever-growing urban regions, a functional traffic signal and control system is a basic and vital test (Cui et al., 2019). The major goals of today's intelligent traffic light foundations are to raise street limit utilization, manage traffic streams, reduce emanation, improve traffic security, and supply drivers with the best start to finish transportation experience possible. These models rely largely on the creation of traffic signal or traffic light regulator frameworks, which adjust and carefully plan individual traffic lights in order to reach city-wide traffic task destinations (Rath, 2018). Coordination of such frameworks with vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) correspondence strategies (collectively referred to as V2X) correspondence strategies (considering dynamic constant data exchange between all players in the traffic controlling environment and cultivating useful metropolitan information) could undoubtedly be the next stage of this development. In this paper, we provide a design for V2X-coordinated traffic signal regulator frameworks, as well as a testbed with a proof-of-concept implementation of the proposed method (Islam & Rashid, 2018). Crossings and crossing point situations are characterized at the office layer by the administrations of Intersection Geometry and Topology (MAP) and Signal Phase and Timing (SPAT). MAP contains information on sophisticated crossing point geometry (used with the SPAT message), fast bend diagrams (used in bend wellbeing alarms), and street sections (utilized in platooning applications) (Tangade & Manvi, 2016).
We used machine learning to discover an ideal VANET granularity by combining least configurations of choice measures, which has been known as a good asset in independent direction (Contreras-Castillo et al., 2018). When compared to the restricted and fundamental rules, the ML prediction performance produced more palatable results with less preparation data. By combining the ML reasoned grouping granularity with the NMDP-APC message succession and method, we were able to plan an NMDP-APC message succession and method. The current 3GPP C-V2X requirements have been changed in terms of strategy and succession. The proposed approach can be easily deployed with forthcoming 5G cellular systems thanks to the PC5 interface. To deal with the adaptation of nomadic vehicles' mobility, this system, in particular, employs a circulating control mechanism (Tientrakool et al., 2011). Two critical bounds C and were adopted to tune the prediction execution in order to better develop ML expectation execution. According to the simulation results, a precise border specification resulted in a superior presentation. We observed a presentation with decreased access inertness in a robust PLMN system through simulation, demonstrating that grouping ability increases aggregation effects and so provides a stable framework in any situation when a large number of objects interact with a PLMN (Zhang et al., 2016). Table 1 explain detail Summary of modern literature.
Table 1
Summary of the existing literature
Author and Year
|
Paper Title
|
Techniques
|
Strength
|
Weakness
|
Reported Performance
|
(Tientrakool et al., 2011)
|
Highway Capacity Benefits from Using Vehicle-to-Vehicle Communication and Sensors for Collision Avoidance
|
Ordered probity model,
heterogeneity to identify autonomous vehicles
|
AV adequately tackled the bad weather
|
Technical dependability
|
The findings demonstrate that AVs need to be enhanced in the following areas: equipment and safety, legal liability, system and security, responsiveness to the driving environment, and performance in inclement weather.
|
(Obaid & Torok, 2021)
|
Macroscopic Traffic Simulation of Autonomous Vehicle Effects
|
Deep Reinforcement learning (drl)
|
Using a drl-based signal control system, traffic lights are dynamically adjusted to reflect the level of congestion at crossings.
|
Due to excessive traffic, all vehicle routes must be changed.
|
We dismantle data silos and combine all the data from sensors, detectors, cars, and roads to generate long-lasting outcomes in order to reap the true benefits of the suggested strategy. For our simulations, we employed the sumo micro-simulator. The outcomes demonstrate the importance of our suggested strategy.
|
(Lu & Kim, 2017)
|
A Genetic Algorithm Approach for Expedited Crossing of Emergency Vehicles in Connected and Autonomous Intersection Traffic
|
Dynamic traffic control,
Vehicle-to-infrastructure
|
Using wireless technologies for bidirectional communication, an efficient and secure autonomous vehicle may be created.
|
Not suitable for V2X communication
|
Utilizing several metrics relating to operating time, data flow, and resource management, the study's V2I was evaluated for performance.
|
(Zhu & Ukkusuri, 2015)
|
A linear programming formulation for autonomous intersection control within a dynamic traffic assignment and connected vehicle environment
|
Mixed integer linear programming,
|
The developed lane-based methods can be formulated as mixed integer linear programming (milp) problems, which can be solved using the cplex solver.
|
Not good for mixed integer linear programming for emergency autonomous vehicle in v2x communication
|
The suggested method's potential for use in real-time is demonstrated by the further use of a heuristic to quickly solve the milp model. To evaluate the performance and efficacy of the suggested approaches and heuristics, numerical analysis is done. We discovered that lane/route optimization is frequently more important than entry time.
|
(Hancock et al., 2019)
|
On the future of transportation in an era of automated and autonomous vehicles
|
Public transport, vaimo
|
Raise awareness about the future of public transport map by comprising different driverless vehicles
|
Every driverless vehicle has its advantages and disadvantages so not sport able on all environment
|
This essay intends to spread knowledge regarding the future of all forms of transportation, including public transportation as well as products delivery, tour buses, aircraft, and more.
|
(Shrestha et al., 2020)
|
Evolution of V2X Communication and Integration of Blockchain for Security Enhancements
|
Vehicle-to-every thigs communication
|
Light weight message verification scheme for vehicle to everything (V2X) communication
|
The proposed scheme is efficient for vanet and its security
|
Comparing the suggested protocols findings to those of other pertinent communication techniques, it is found that it consumes less computing power and also secure communication than other vehicular communication schemes.
|
(Li & Liu, 2020)
|
Intersection management for autonomous vehicles with vehicle-to-infrastructure communication
|
Vehicle-to-infrastructure communication, traffic light
|
Reduce the average time delay
|
Just use one network on the base of static conflict matrix technique
|
The proposed strategy could significantly reduce the average time delay caused by the intersection and the corresponding variance, which shows the efficiency and fairness of the proposed strategy in intersection management.
|
(Rath, 2018)
|
Smart Traffic Management System for Traffic Control using Automated Mechanical and Electronic Devices
|
Vehicular ad hoc network
|
in smart sensors of connected vehicles are handle by blockchain framework
|
It frames work is not suitable for some condition
|
The suggested architecture dramatically decreased the number of fictitious requests from users, iot device compromises, and changes to the ratings of previously saved users. Comparing the suggested framework against the current technique against the specified parameters, the simulation results against various parameters revealed a 79% success rate in the proposed framework.
|
(Cui et al., 2019)
|
safety failures, security attacks, and available countermeasures for autonomous vehicles
|
Conventional vehicles and connected autonomous vehicles, car on traffic safety under various penetration rates
|
Traffic safety is greatly improved with the increase in the car penetration rate.
|
Not good for emergency vehicles
|
The percentage of smooth driving has grown together with the penetration rate of cars. Vehicle speed differences are reduced, and traffic flow is substantially slowed. Traffic jams will be considerably reduced.
|