METANET Validation of the Large-Scale Manchester Ring-Road Network Using Gradient-Based and Particle Swarm Optimization

Poole, Adam; Kotsialos, Apostolos

doi:10.1109/tits.2017.2724941

Cited by 10 publications

(7 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where y is the calculated output and y r is the real measurement of the output. Now, the decision to consider which traffic characteristic as output measurement is thoroughly investigated in References [19][20][21] and they all concluded that the F I G U R E 1 Hypothetical highway stretch measurements solely depend on the type of loop detectors installed in the highway. In this work, the authors considered the average speed of each section for simplicity and the fact that these measurements were provided.…”

Section: Calibration Proceduresmentioning

confidence: 99%

“…In this work, the authors considered the average speed of each section for simplicity and the fact that these measurements were provided. Another reason for this consideration is that authors in Reference [19][20][21] discuss that flow or average speed measurements are rather simpler than density measurements. In this work, to solve the optimization problem (9), the authors employed particle swarm optimization (PSO) based on the work presented in References [20,21].…”

Section: Calibration Proceduresmentioning

confidence: 99%

“…Expanding right side of Equation (19) gives, (20) where T = G + K. We can deduct from Equation 20that,…”

Section: Rm Based On Smcmentioning

confidence: 99%

“…The calibration problem can be reformulated into an optimization problem with the following formulation:

\begin{matrix} J (p) & = \min_{z} \sqrt{\frac{1}{N_{s}} \sum_{k} {false(y false(k false) - y^{r} false(k false) false)}^{2}} \\ s . t . x (k + 1) & = f (x (k), d (k), z, k) \\ z_{\min} \leq z \leq z_{\max}, \end{matrix}

where y is the calculated output and y r is the real measurement of the output. Now, the decision to consider which traffic characteristic as output measurement is thoroughly investigated in References [19–21] and they all concluded that the measurements solely depend on the type of loop detectors installed in the highway. In this work, the authors considered the average speed of each section for simplicity and the fact that these measurements were provided.…”

Section: Traffic Flow Control Problem Formulationmentioning

confidence: 99%

Section: Traffic Flow Control Problem Formulationmentioning

confidence: 99%

See 4 more Smart Citations

Feedback‐based cooperative ramp metering for highway traffic flow control: A model predictive sliding mode control approach

Aval

Eghbal

2020

Intl J Robust & Nonlinear

View full text Add to dashboard Cite

In this article, the authors proposed a model predictive based sliding mode control (MPSMC) based ramp metering (RM) strategy for highway traffic control. The strategy includes sliding mode control (SMC) which makes it immune to uncertainties and provides fast responses while the addition of model predictive Control (MPC) will provide optimality. Therefore, the RM strategy will simultaneously present real-time applicability, robustness, and optimality. To provide optimality, a new quadratic objective function homogeneous to total time spent (TTS) is proposed, and for both the SMC and MPSMC based RM the authors will prove the closed-loop asymptotic global stability of the strategy in the presence of disturbance and uncertainties with the help of the proposed cost function. This objective function is based on minimizing traffic dynamic deviations from desired values. The proposed RM control strategy was applied to M62 smart highway in England and results are compared with existing control measures in the highway, ALINEA and SMC-based RM. The simulations show that MPSMC will approach optimal behavior and thus, outperforms ALINEA and SMC regarding queue length and average speed. Moreover, the SMC and MPSMC based RMs are innately robust against disturbances, therefore, both will provide better performance regarding congestion reduction compared with ALINEA. Finally, it will be shown that MPSMC can improve TTS performance by 39% with respect to ALINEA while presenting global asymptotic stability.

show abstract

Section: Calibration Proceduresmentioning

confidence: 99%