An Efficiency Enhancing Methodology for Multiple Autonomous Vehicles in an Urban Network Adopting Deep Reinforcement Learning

Abstract

To reduce the impact of congestion, it is necessary to improve our overall understanding of the influence of the autonomous vehicle. Recently, deep reinforcement learning has become an effective means of solving complex control tasks. Accordingly, we show an advanced deep reinforcement learning that investigates how the leading autonomous vehicles affect the urban network under a mixed-traffic environment. We also suggest a set of hyperparameters for achieving better performance. Firstly, we feed a set of hyperparameters into our deep reinforcement learning agents. Secondly, we investigate the leading autonomous vehicle experiment in the urban network with different autonomous vehicle penetration rates. Thirdly, the advantage of leading autonomous vehicles is evaluated using entire manual vehicle and leading manual vehicle experiments. Finally, the proximal policy optimization with a clipped objective is compared to the proximal policy optimization with an adaptive Kullback–Leibler penalty to verify the superiority of the proposed hyperparameter. We demonstrate that full automation traffic increased the average speed 1.27 times greater compared with the entire manual vehicle experiment. Our proposed method becomes significantly more effective at a higher autonomous vehicle penetration rate. Furthermore, the leading autonomous vehicles could help to mitigate traffic congestion.

A Phase‐Change Memristive Reinforcement Learning for Rapidly Outperforming Champion Street‐Fighter Players

“…Here, λ was defined as the GAE-lambda hyperparameter, which was a measure of the degree of the dependence on the current-value estimate for calculating updated-value estimates. [18,89] The high and low values of the λ were associated with the strong and weak degrees of the dependence on current-value estimates (high bias, low variance)/actual rewards from the environment (low bias, high variance), respectively. Moreover, a difficulty emerged from the trade-off between the bias and variance, and finding an optimal value of the λ facilitated the training process.…”

A study on multi-agent reinforcement learning for autonomous distribution vehicles

“…Tran and Bae (2021) present an advanced deep RL technique in [ 3 ] that investigates the impact of leading autonomous vehicles in a mixed-traffic scenario on the urban network. They propose a set of hyperparameters to improve the performance of deep RL agents.…”

“…Research has demonstrated the effectiveness of using a multi-agent approach for vehicle routing in studies [ 3 , 12 ]. Furthermore, studies [ 7 , 8 , 10 ] have shown the effectiveness of using RL for learning in complex systems like cooperative driving.…”

“…With the growth of the smart city agenda, automated vehicles are currently a hot topic. A potentially disruptive breakthrough in the context of urban mobility is autonomous vehicle delivery [ 3 ].…”

A Multi-Objective Optimal Control Method for Navigating Connected and Automated Vehicles at Signalized Intersections Based on Reinforcement Learning

“…A framework using convolutional neural networks for prediction of time consumption at intersections was proposed, enabling optimal passing order and continuous control for connected vehicles [27]. The impact of leading autonomous vehicles on urban networks was investigated, showcasing potential congestion mitigation benefits [28]. A DRL-based reference speed-planning strategy was introduced for hybrid electric vehicles, with the goal of optimizing fuel economy and enhancing driving safety [29].…”