New Approach to Modeling Mixed Traffic Containing Motorcycles in Urban Areas

Lee, Tzu-Chang; Polak, John; Bell, M.G.

doi:10.3141/2140-22

Cited by 68 publications

(40 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In recent years, MCMC algorithm has played an important role in travel behaviour modelling problems in transportation, with successful applications in agent-based simulations [22], hybrid choice models [23,24], and population synthesis [25,26,27]. However, in order to match the asymptotic efficiency of maximum likelihood, MCMC draws must grow at a rate faster than the square root of the number of agents [3,28].…”

Section: Literature Reviewmentioning

confidence: 99%

A bi-partite generative model framework for analyzing and simulating large scale multiple discrete-continuous travel behaviour data

Wong

Farooq

2020

Transportation Research Part C: Emerging Technologies

View full text Add to dashboard Cite

The emergence of data-driven demand analysis have led to the increased use of generative modelling to learn the probabilistic dependencies between random variables. Although their apparent use has largely been limited to image recognition and classification in recent years, generative machine learning algorithms can be a powerful tool for travel behaviour research by replicating travel behaviour by the underlying properties of data structures. In this paper, we examine the use of generative machine learning approach for analyzing multiple discrete-continuous (MDC) travel behaviour data. We provide a plausible perspective of how we can exploit the use of machine learning techniques to interpret the underlying heterogeneities in the data. We show that generative models are conceptually similar to choice selection behaviour process through information entropy and variational Bayesian inference. Without loss of generality, we consider a restricted Boltzmann machine (RBM) based algorithm with multiple discrete-continuous layer, formulated as a variational Bayesian inference optimization problem. We systematically describe the proposed machine learning algorithm and develop a process of analyzing travel behaviour data from a generative learning perspective. We show parameter stability from model analysis and simulation tests on an open dataset with multiple discrete-continuous dimensions from a data size of 293,330 observations. For interpretability, we derive the conditional probabilities, elasticities and perform statistical analysis on the latent variables. We show that our model can generate statistically similar data distributions for travel forecasting and prediction and performs better than purely discriminative methods in validation. Our results indicate that latent constructs in generative models can accurately represent the joint distribution consistently on MDC data.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

A bi-partite generative model framework for analyzing and simulating large scale multiple discrete-continuous travel behaviour data

Wong

Farooq

2020

Transportation Research Part C: Emerging Technologies

View full text Add to dashboard Cite

show abstract

“…Chandra, Agarwal, and Rajamma (2009) developed exponential models to estimate service delay to different categories of vehicles for various priority movements and for varying proportions of heavy and light vehicles in the conflicting traffic under mixed traffic conditions. Lee, Polakand, and Bell (2009) developed a new approach to model mixed traffic focusing on depicting the movements of motorcycles. Patil, Patare, and Sangole (2011) developed gap acceptance probabilistic models for two-wheelers at three-legged unsignalized intersections.…”

Section: Leadmentioning

confidence: 99%

Study of unique merging behavior under mixed traffic conditions

Kanagaraj

Srinivasan

Sivanandan

et al. 2015

Transportation Research Part F: Traffic Psychology and Behaviou

View full text Add to dashboard Cite

“…Specifically, λs y 0 indicates the lateral desired speed induced by a longitudinal acceleration shear of 1 m/s 2 per meter, and λ itself the maximum ratio between the desired lateral speed and the longitudinal interaction deceleration induced by vehicle j. However, expression (14) still lacks a dependence on the relative lateral speed w j − w i which clearly is relevant in order to avoid collisions or prevent unnecessary steering actions. Since such a term cannot be derived from MOBIL, we augment (14) in the simplest possible way that satisfies the plausibility condition "no lateral force if there is no longitudinal force" by adding to (14) a linear relative speed dependence −λ ∆w sign(∆y ij )(w j − w i ) multiplied by (14) itself.…”

Section: Lateral Dynamicsmentioning

confidence: 99%

“…The core behavioral models such as car-following models [2,13,1,21] and lanechanging models [3,6,12,15,20,17] are designed for lane-based traffic,only. There exist a few models describing staggered car following [14,9,4], or continuous lane changing based on a constant lateral speed [16]. To our knowledge, there are no models describing the full dynamics of non-lane based traffic.…”

Section: Introductionmentioning

confidence: 99%

Self-driven particle model for mixed traffic and other disordered flows

Kanagaraj

Treiber

2018

Physica A: Statistical Mechanics and its Applications

View full text Add to dashboard Cite

Vehicles in developing countries have widely varying dimensions and speeds, and drivers tend to not follow lane discipline. In this flow state called "mixed traffic", the interactions between drivers and the resulting maneuvers resemble more that of general disordered self-driven particle systems than that of the orderly lane-based traffic flow of industrialized countries. We propose a general multi particle model for such self-driven "high-speed particles" and show that it reproduces the observed characteristics of mixed traffic. The main idea is to generalize a conventional acceleration-based car-following model to a two-dimensional force field. For in-line following, the model reverts to the underlying car-following model, for very slow speeds, it reverts to an anisotropic social-force model for pedestrians. With additional floor fields at the position of lane markings, the model reverts to an integrated car-following and lane-changing model with continuous lateral dynamics including cooperative aspects such as zip merging. With an adaptive cruise control (ACC) system as underlying car-following model, it becomes a controller for the acceleration and steering of autonomous vehicles in mixed or lane-based traffic.

show abstract

New Approach to Modeling Mixed Traffic Containing Motorcycles in Urban Areas

Cited by 68 publications

References 10 publications

A bi-partite generative model framework for analyzing and simulating large scale multiple discrete-continuous travel behaviour data

A bi-partite generative model framework for analyzing and simulating large scale multiple discrete-continuous travel behaviour data

Study of unique merging behavior under mixed traffic conditions

Self-driven particle model for mixed traffic and other disordered flows

Contact Info

Product

Resources

About