arXiv subjects

Haotian Shi

Publications and source records attributed to Haotian Shi.

38 records · Page 3Linked to original sources

arXiv2022

A Driving Risk Surrogate and Its Application in Car-Following Scenario at Expressway

Traffic safety is important in reducing death and building a harmonious society. In addition to studies of accident incidences, the perception of driving risk is significant in guiding the implementation of appropriate driving countermeasures. Risk assessment can be conducted in real-time for traffic safety due to the rapid development of communication technology and computing capabilities. This paper aims at the problems of difficult calibration and inconsistent thresholds in the existing risk assessment methods. It proposes a risk assessment model based on the potential field to quantify the driving risk of vehicles. Firstly, virtual energy is proposed as an attribute considering vehicle sizes and velocity. Secondly, the driving risk surrogate(DRS) is proposed based on potential field theory to describe the risk degree of vehicles. Risk factors are quantified by establishing submodels, including an interactive vehicle risk surrogate, a restrictions risk surrogate, and a speed risk surrogate. To unify the risk threshold, acceleration for implementation guidance is derived from the risk field strength. Finally, a naturalistic driving dataset in Nanjing, China, is selected, and 3063 pairs of following naturalistic trajectories are screened out. Based on that, the proposed model and other models use for comparisons are calibrated through the improved particle optimization algorithm. Simulations prove that the proposed model performs better than other algorithms in risk perception and response, car-following trajectory, and velocity estimation. In addition, the proposed model exhibits better car-following ability than existing car-following models.

math.NA↗

arXiv2020

Partially Connected Automated Vehicle Cooperative Control Strategy with a Deep Reinforcement Learning Approach

This paper proposes a cooperative strategy of connected and automated vehicles (CAVs) longitudinal control for partially connected and automated traffic environment based on deep reinforcement learning (DRL) algorithm, which enhances the string stability of mixed traffic, car following efficiency, and energy efficiency. Since the sequences of mixed traffic are combinatory, to reduce the training dimension and alleviate communication burdens, we decomposed mixed traffic into multiple subsystems where each subsystem is comprised of human-driven vehicles (HDV) followed by cooperative CAVs. Based on that, a cooperative CAV control strategy is developed based on a deep reinforcement learning algorithm, enabling CAVs to learn the leading HDV's characteristics and make longitudinal control decisions cooperatively to improve the performance of each subsystem locally and consequently enhance performance for the whole mixed traffic flow. For training, a distributed proximal policy optimization is applied to ensure the training convergence of the proposed DRL. To verify the effectiveness of the proposed method, simulated experiments are conducted, which shows the performance of our proposed model has a great generalization capability of dampening oscillations, fulfilling the car following and energy-saving tasks efficiently under different penetration rates and various leading HDVs behaviors. Keywords: partially connected automated traffic environment, cooperative control, deep reinforcement learning, traffic oscillation dampening, energy efficiency

eess.SY↗