Imputation of Ramp Flow Data for Freeway Traffic Simulation

Abstract

The Tools for Operations Planning (TOPL) employs the Link-Node Cell Transmission model for macroscopic freeway traffic simulations for specifying operational strategies like ramp metering, demand and incident management.Traffic flow and occupancy data from loop detectors is used for calibrating these models and specifying the inputs to the simulation. However, flow data from ramps are often to be found missing or incorrect. This paper elaborates an imputation procedure used to determine these ramp flows. This automated imputation procedure is based on an adaptive identification technique that tries to minimize the error between the simulated and the measured densities. The simulation results using the imputed flow data indicate good conformation with loop detector measurements.

Deep learning for short-term traffic flow prediction

“…Estimating boundary conditions is a challenging task, which even in systems that rely on loop detectors as traffic sensors are typically not installed on ramps. Missing data problems are usually addressed using data imputation Muralidharan and Horowitz [2009] or weak formulations of boundary conditions Strub and Bayen [2006]. Our results show that a data-driven approach can efficiently forecast flows without boundary measurements from ramps.…”

Grenoble Traffic Lab: An Experimental Platform for Advanced Traffic Monitoring and Forecasting [Applications of Control]

“…The raw macroscopic traffic data coming every 15 s from the Sensys magnetometers (see Level 1) are stored in a database and then passed through a suite of signal processing algorithms (the green box in Figure 3), which perform: » Imputation and diagnostics: If some data are lost or erroneous (for instance, as a result of communication problems or temporary sensor malfunction), suitable imputation algorithms [21], [22] are run for filling in the missing data with estimated values; see the section "Level 3: Results Display." In this respect, each magnetometer is evaluated not as a standalone but together with its neighbors and their past measurement history (see [23] and "Vehicle Counting" under "Data Analysis" in [24]).…”

Highway traffic model-based density estimation

“…A nice survey on the existing techniques for short-term traffic flow prediction can be found in [3]. One of the prerequisites for continuous prediction is an efficient real-time traffic conditions estimation using only a limited amount of data [9], [10]. The real-time density estimation is a challenging problem since the traffic is described by a system which is observable only when a segment situated between two vehicle detector stations (sensors) is entirely congested or entirely free.…”

“…Once the models validated they are used for open-loop stateestimation, forecasting and control ( [5], [8]). Other papers present imputation techniques to determine the missing onramp and off-ramp flows [10].…”