Traffic state prediction has always been challenge for both urban planners and traffic engineers as the enhancement of infrastructures is at stake by them. With the wide technological innovations and advancements happening in today's field of autonomous vehicles, these concerns are bound to rise soon (Walker and Johnson, 2016). The General Motors and Radio Corporation of America Sarnoff Laboratory first proposed the idea of autonomous vehicles in the 1950s (Pendleton et al.,2017).
Hence, predictive techniques are required by infrastructure operators for further modeling. The real-time prediction of traffic flow on a road segment enables transportation authorities to take actions to control and relieve traffic congestion.
During the past few decades, technological advances have enormously influenced transportation infrastructures, strategies, and policies starting to evolve dramatically, especially within developed countries. One of the most topical issues throughout modern transportation is automated driving. It has also been said that the introduction of automated vehicles (AVs) will cause a mobility revolution (Wang et al., 2019).
To provide the transportation system with the best benefits of AVs, it is necessary to consider some futuristic strategies consist of designing appropriate infrastructures, scheduling public transportation services, employing system integration solutions, and systematic pricing and subsidies. According to (Van den Berg et al., 2016), AVs can safely drive in a much less headway than cars driven by humans. Hence, they possibly increase road capacity. Also, by allowing users to perform other activities in the vehicle, they can reduce the value of travel time (VOT) losses. On the other hand, (Van den Berg et al., 2016) indicated that CAVs can influence Congestion in a harmful way by imposing increased congestion externalities, vehicle miles traveled and car-trip frequencies. All in all, according to these potential services and facilities in the condition of the availability of CAVs in addition to the threats that automated driving will pose to the network congestion, it is significant to predict the performance of automated services throughout different urban areas.
FHWA Vehicle Classification Scheme F
In a very recent study (Litman, 2020) casts doubts on another critical issue which is the degree that some potential goals can be accomplished when only a portion of vehicles in the network are autonomous. Some benefits, such as improved mobility for CAV users, may achieve even if they do not find much popularity among the public due to their high price, but many other benefits, such as reduced congestion, emission rate, traffic signals and lane widths, require exclusive lanes for such vehicle to operate optimally. It should also be noted that the prediction of parameters such as traffic flow, average speed and also travel demand would play a key role in studying such issues. To address the problem of generating data for such studies, most studies have adopted simulation methods. (e.g. Basu et al., 2018).
(Ranwa Al Mallah et al., 2019) have studied the cause of traffic congestion using cooperation between CVs in five different scenarios. They inspected that a quite unsurmountable proportion of CVs is essential so as to achieve desirable performance in inclement weather. Also, the performance turned out to be highly dependent on the assumed parameters of the car-following model for the simulation of urban mobility in the inclement weather scenario. In the recent publication of the same authors (Ranwa Al Mallah et al., 2020), they aimed for short term traffic flow prediction with Deep Neural Network trained in simulated scenarios using a realistic data set on a certain road segment via CVs. They indicated that under high-density circumstances, the number of sent messages will increase and the communication network becomes congested. Thus, if a vehicle misses a part of traveling vehicles' messages on its current road segment, its evaluation of traffic flow will not be of high accuracy anymore.
As a matter of fact, it is crystal clear that connected and autonomous vehicles (CAVs) potentially increase vehicle travel by reducing travel and parking costs and by providing improved mobility to both the youth and the elderly. CAV technologies are raising questions about how people deal with personal cars. Potential benefits in the studies are merely the hypothesis of scholars, designing for the best-case scenario with no understanding of the complexities associated with such benefits. However, the main contributing factors associating these predictions should be taken into account in order to have a better understanding of the network planning. These risks can affect the user (Internal impacts) in ways such as increased vehicle costs, additional user risks (as additional crashes) and reduced security and privacy. However, External impacts include increased traffic problems, reduced security, social equity concerns, reduced employment, increased infrastructure costs and reduced support for other solutions. It is also mentioned in studies that the appropriate time to plan the required infrastructure for smooth adoption of AVs and solve the congestion due to the gap between current traffic flow and roads ‘capacities is already passing (Mahdavian et al., 2019).
Recent studies suggest that fully automated vehicles are so likely to have significant implications for smart mobility and planning policies. The pace of development and subsequent implications depends on technological evolution, policies and users’ attitude to a great extent. These driving forces could be handled differently from one country or an expert to the other, based on their potentiality and impacts. Such assumptions can provide input to subsequent models of these impacts and their probable interactions. Sensitivity analysis in these modeling exercises could solve uncertainty issues about the proposed impacts.
Significant data can be used to better evaluate the implications of CAVs on fuel consumption and traffic flow. The simulation results have shown that fuel consumption can be significantly reduced under all traffic conditions, while the highest benefits are achieved at medium traffic flows. An identical pattern was found for the travel time. One important finding in studies is that full penetration of CAVs can bring more stable traffic patterns even when the traffic density is high. Further investigation of the implications for different penetration levels of optimally coordinated CAVs is the current challenge in this area. Future works are expected to assess the impacts of communication-related instabilities in the performance of the optimal coordination framework.
References:
Al Mallah, R., Quintero, A. and Farooq, B., 2019. Cooperative evaluation of the cause of urban traffic congestion via connected vehicles. IEEE Transactions on Intelligent Transportation Systems, 21(1), pp.59-67.
Al-Mallah, R., Quintero, A. and Farooq, B., 2020. Prediction of Traffic Flow via Connected Vehicles. IEEE Transactions on Mobile Computing.
Basu, R., Araldo, A., Akkinepally, A.P., Nahmias Biran, B.H., Basak, K., Seshadri, R., Deshmukh, N., Kumar, N., Azevedo, C.L. and Ben-Akiva, M., 2018. Automated Mobility-on-Demand vs. Mass Transit: A Multi-modal Activity-driven Agent-based Simulation Approach. Transportation Research Record, 2672(8), pp.608-618.
Litman, T., 2020. Autonomous vehicle implementation predictions: Implications for transport planning.
Mahdavian, A., Shojaei, A. and Oloufa, A., 2019, November. Assessing the long-and mid-term effects of connected and automated vehicles on highways’ traffic flow and capacity. In International Conference on Sustainable Infrastructure 2019: Leading Resilient Communities through the 21st Century (pp. 263-273). Reston, VA: American Society of Civil Engineers.
Pendleton, S.D., Andersen, H., Du, X., Shen, X., Meghjani, M., Eng, Y.H., Rus, D. and Ang, M.H., 2017. Perception, planning, control, and coordination for autonomous vehicles. Machines, 5(1), p.6.
Van den Berg, V.A. and Verhoef, E.T., 2016. Autonomous cars and dynamic bottleneck congestion: The effects on capacity, value of time and preference heterogeneity. Transportation Research Part B: Methodological, 94, pp.43-60.
Walker, J. and Johnson, C., 2016. Peak car ownership: the market opportunity of electric automated mobility services. Rocky Mountain Institute. content/uploads/2017/03/Mobility_PeakCarOwnership_Report2017.pdf.
Wang, S., Correia, G.H.D.A. and Lin, H.X., 2019. Exploring the Performance of Different On-demand Transit Services Provided by a Fleet of Shared Automated Vehicles: An Agent-based Model. Journal of Advanced Transportation, 2019.
Comments