Review Article

Simulation-Based Connected and Automated Vehicle Models on Highway Sections: A Literature Review

Table 1

Simulation-based intelligent-vehicle studies: objectives, models, and scenarios.

Ref #ObjectivesBase model(s)Scenarios

[32]Develop the ACC and CACC car-following models and estimate their impact.An error-based control law for the ACC and CACC.
The lane change is under human control.
A 100% market-penetration rate of each vehicle type.

[29]Examine the ACC vehicles’ lane-changing effects compared to manual vehicles.Manual vehicle: Pipes model [47].
ACC model from [48].
Comprehensive Modal Emissions Model (CMEM).
Position of ACC vehicles (2, 4, 6, 8th in the string of 10 vehicles).
Market-penetration rate of ACC (5%, 10%, 15% and 30%).

[31]Propose the ACC-based traffic-assistance system intended to improve traffic flow and road capacity.IDMMarket-penetration rate of ACC (0%, 5%, 15% and 25%).

[14]Propose the ACC-based traffic assistance system aimed at improving the traffic flow and road capacity.IDMMarket-penetration rate of ACC (0%, 5%, 15% and 25%).

[30]Propose the new ACC car-following model with its impact analysisIDM with constant-acceleration
heuristic (CAH).
Market-penetration rate of ACC (10%, 20%, 30%, 40%, and 50%).

[18]Propose an analytical framework to estimate the AVs’ impacts on highway sections.Car-following model for manual vehicles in [49, 50].
First order control law for AVs.
Different combinations of manual vehicles, AVs, and CAVs (0-100 % by 10% gap).

[19]Develop an improved cellular automaton as an AV modeling platform.Cellular AutomatonThe lane-changing rules in the same and opposite direction.
Market-penetration rate of ACC (0%, 50%, and 100%).

[46]Develop a cooperative IDM (CIDM) to examine the system performance under different proportions of the AVs.The Full Velocity Difference
Model (FVDM) and IDM.
Market-penetration rate of the AVs (0%, 5%, 15%, and 25%).

[45]Propose an acceleration framework to address the limitations of micro-simulation models in capturing the changes in driver behavior in a mixed environment.MIXIC model for the AV modeling.
IDM for the CAV modeling.
Market-penetration rate of the CAVs and AVs (0%, 20%, 40%, 60%, 80%, and 100%).

[44]Develop a micro-simulation framework for CAVs to analyze the impact on fuel consumption and travel time.Optimal control for CAVs.
Gipps model for manual vehicles [51].
Two single-lane merging roadways where CAVs communicate to each other.

[15]Propose a hardware-in-the-loop (HIL) testing system for the CAV applications.Hardware-in-the loop (HIL) testing.Type I: String leader’s smooth acceleration and deceleration between 20-30mph.
Type II: Sharp brakes from 30mph to 10mph and quick recovery to 30mph.
Type A: Perfect communication/radar.
Type B: Compromised communication/radar (radar delay 100ms; radar noise = 0.05; DSRC Latency = 100ms and DSRC Packet Loss =10%).

Examine the impact of the CACC vehicles on traffic flow characteristics of a multilane highway.IDMArrival rate scenarios: 7,000v/h (moderate), 8,000v/h (saturated),
9,000v/h (oversaturated), 10,000v/h (oversaturated).
Penetration rates of CACC varied in multiples of 20% (truck is fixed in 10%).

[52]Develop a simulation framework to facilitate the heavy-duty vehicle (HDV) platooning and establish the related concept and operations.Carbon dioxide emission model [53].
The HDM platoon model with the ACC/CACC car-following model.
Average density, average travel time, and average travel speed.

[17]Investigate AVs’ impact on traffic performance.Calibration on car following model (Wiedemann 99).
Lane changing behavior based
on a research project [54].
Each vehicle type of a 100% market-penetration rate.

[37]Extend the CACC modeling framework to incorporate new algorithms describing the interactions between the CACC and manual vehicles in mixed traffic.The CACC model reported in [55].
The anticipatory lane change (ALC) for lane changing.
Market-penetration rate of the CACC (0%, 20%, 40%, 60%, 80% and 100%).

[36]Investigate the impact of the CACC vehicle string operation on the capacity of multilane highway with merging bottlenecks.The ACC and CACC car-following models developed [33].Market-penetration rate of the CACC (0%, 20%, 40%, 60%, 80% and 100%).

[56]Propose a new algorithm for the CACC systems for collaborative driving based on the use of agent technology and information sharing.Effective CACC (ECACC) algorithm consists of speed and distance control algorithms.Market-penetration rate of the CACC (0%, 20%, 40%, 60%, 80% and 100%).

[27]Estimate the effect on highway capacity of varying market-penetrations of vehicles with the ACC and the CACC.The manual vehicle: NGSIM oversaturated freeway flow model [57].
ACCs: Proprietary to Nissan.
CACCs: Car-following behavior was described [33].
The ACC and CACC vehicles 10 % increase proportion.

[21]Investigate the impact of the CACC on traffic-flow characteristic.MIXIC modelMarket-penetration rate of the CACC (0%, 20%, 40%, 60%, 80% and 100%).

[58]Develop the models of both ACC and CACC control systems based on real experimental data.IDMTen consecutive CACC and five consecutive ACC vehicles.
A mixed case, where the two first followers are ACC-equipped and the next seven are CACC-equipped.

[59]Estimate the emissions and energy use (i.e., fuel consumption) associated with an Automated Highway System (AHS) using advanced simulation modeling tools.Smart AHS framework developed at PATH program.Congestion levels (LOS A - F).

[60]Analyze roundabout safety level in the circumstances where different numbers of the AVs are mixed with manual vehicles.Safety impact: Surrogate Safety Assessment Model (SSAM).
Manual vehicles: Wiedemann 74.
AVs: VISSIM parameter adjustment.
Market-penetration rate of the AVs (0%, 10%, 25%, and 50%).

[61]Develop the decision-making CAV control algorithm in the VISSIM for safety evaluations.Safety impact: SSAM.
CAV: External driver model API written in C++.
Manual vehicles: Wiedemann 99.
Market-penetration rate of the CAVs (0%, 25%, 50%, 75%, and 100%).
Daily based estimation, Monday to Friday.