Dynamic Prediction of Vehicle Cluster Distribution in Mixed Traffic: A Statistical Mechanics-Inspired Method

Jerath, Kshitij; Ray, Asok; Brennan, Sean; Gayah, Vikash V.

doi:10.1109/tits.2015.2409798

Cited by 14 publications

(11 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Jerath et al modeled human-driven vehicles and self-driving vehicles [8], and studied the effects of vehicle density and adjacent nodes on vehicle motion. Based on this, the distribution of vehicles' position is predicted in a short period of time.…”

Section: B Prediction Based On Statistical Modelsmentioning

confidence: 99%

Location Prediction Model Based on the Internet of Vehicles for Assistance to Medical Vehicles

et al. 2020

View full text Add to dashboard Cite

Along with the Internet of Vehicles, some intelligent systems can help the medical vehicles transport medical supplies and patients. In terms of emergency issues like catastrophic natural disasters or serious accidents, safe and timely transportation for medical vehicles is particularly important. For assistance to medical vehicles on the road, models with position prediction can provide accurate position information of ambient vehicles in the next seconds. However, with the increasing number of vehicles on the road and the changing road environment, it is an important challenge to predict the location of vehicles in the road correctly. Current location prediction models for vehicles usually use the previous trajectory of vehicles, lacking the consideration of vehicle's state and real-time traffic information, which leads to relatively low accuracy and safety. Based on the deep belief nets (DBN) and long short-term memory (LSTM), this paper presents a location prediction model for assistance to medical vehicles (LPMVs), which fully considers vehicle's attributes, road information and driving environment as well as the relationship between the factors that influence vehicle driving behaviors and vehicle positions. By experiment, we prove that LPMVs can predict more accurately than current location prediction models and be a good choice for assistance to medical vehicles. INDEX TERMS Internet of Vehicles, medical vehicles, neural network, prediction model.

show abstract

Section: B Prediction Based On Statistical Modelsmentioning

confidence: 99%

Location Prediction Model Based on the Internet of Vehicles for Assistance to Medical Vehicles

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Recently, these approaches have witnessed a resurgence. For instance, (Jerath et al 2015) recently studied the effect of driver algorithms on traffic flow. (Suzuki, Imura, and Aihara 2013) applied the Ising model to study chaotic dynamics and serve as a starting point for considering statistical mechanics of traffic signals.…”

Section: Related Workmentioning

confidence: 99%

Coping with Large Traffic Volumes in Schedule-Driven Traffic Signal Control

Hu¹,

Smith²

2019

Preprint

View full text Add to dashboard Cite

Recent work in decentralized, schedule-driven traffic control has demonstrated the ability to significantly improve traffic flow efficiency in complex urban road networks. However, in situations where vehicle volumes increase to the point that the physical capacity of a road network reaches or exceeds saturation, it has been observed that the effectiveness of a schedule-driven approach begins to degrade, leading to progressively higher network congestion. In essence, the traffic control problem becomes less of a scheduling problem and more of a queue management problem in this circumstance. In this paper we propose a composite approach to real-time traffic control that uses sensed information on queue lengths to influence scheduling decisions and gracefully shift the signal control strategy to queue management in high volume/high congestion settings. Specifically, queue-length information is used to establish weights for the sensed vehicle clusters that must be scheduled through a given intersection at any point, and hence bias the wait time minimization calculation. To compute these weights, we develop a model in which successive movement phases are viewed as different states of an Ising model, and parameters quantify strength of interactions. To ensure scalability, queue information is only exchanged between direct neighbors and the asynchronous nature of local intersection scheduling is preserved. We demonstrate the potential of the approach through microscopic traffic simulation of a real-world road network, showing a 60% reduction in average wait times over the baseline scheduledriven approach in heavy traffic scenarios. We also report initial field test results, which show the ability to reduce queues during heavy traffic periods.

show abstract

Section: Related Workmentioning

confidence: 99%

Coping with Large Traffic Volumes in Schedule-Driven Traffic Signal Control

Smith

2017

ICAPS

View full text Add to dashboard Cite

Recent work in decentralized, schedule-driven traffic control has demonstrated the ability to significantly improve traffic flow efficiency in complex urban road networks. However, in situations where vehicle volumes increase to the point that the physical capacity of a road network reaches or exceeds saturation, it has been observed that the effectiveness of a schedule-driven approach begins to degrade, leading to progressively higher network congestion. In essence, the traffic control problem becomes less of a scheduling problem and more of a queue management problem in this circumstance. In this paper we propose a composite approach to real-time traffic control that uses sensed information on queue lengths to influence scheduling decisions and gracefully shift the signal control strategy to queue management in high volume/high congestion settings. Specifically, queue-length information is used to establish weights for the sensed vehicle clusters that must be scheduled through a given intersection at any point, and hence bias the wait time minimization calculation. To compute these weights, we develop a model in which successive movement phases are viewed as different states of an Ising model, and parameters quantify strength of interactions. To ensure scalability, queue information is only exchanged between direct neighbors and the asynchronous nature of local intersection scheduling is preserved. We demonstrate the potential of the approach through microscopic traffic simulation of a real-world road network, showing a 60% reduction in average wait times over the baseline schedule-driven approach in heavy traffic scenarios. We also report initial field test results, which show the ability to reduce queues during heavy traffic periods.

show abstract

Dynamic Prediction of Vehicle Cluster Distribution in Mixed Traffic: A Statistical Mechanics-Inspired Method

Cited by 14 publications

References 21 publications

Location Prediction Model Based on the Internet of Vehicles for Assistance to Medical Vehicles

Location Prediction Model Based on the Internet of Vehicles for Assistance to Medical Vehicles

Coping with Large Traffic Volumes in Schedule-Driven Traffic Signal Control

Coping with Large Traffic Volumes in Schedule-Driven Traffic Signal Control

Contact Info

Product

Resources

About