Publications related to Laboratoire de systemes de transports urbains

Multivariate time series forecasting for freeway networks

This dissertation introduces traffic forecasting methods for different network configurations and data availability.Chapter 2 focuses on single freeway cases.Although its topology is simple, the non-linearity of traffic features makes this prediction still a challenging task.We propose the dynamic linear model (DLM) to approximate the non-linear traffic features. Unlike a static linear regression model, the DLM assumes that its parameters change over time.We design the DLM with time-dependent model parameters to describe the spatiotemporal characteristics of time-series traffic data. Based on our DLM and its model parameters analytically trained using historical data, we suggest the optimal linear predictor in the minimum mean square error (MMSE) sense.We compare our prediction accuracy by estimating expected travel time based on the traffic prediction for freeways in California (I210-E and I5-S) under highly congested traffic conditions with other baselines. We show significant improvements in accuracy, especially for short-term prediction.Chapter 3 aims to generalize the DLM to extensive freeway networks with more complex topologies.Most resources would be consumed to estimate unnecessary spatiotemporal correlations if the DLM was directly used for a large-scale network.Defining features on graphs relaxes such issues by cutting unnecessary connections in advance based on predefined topology information.Exploiting the graph signal processing, we represent traffic dynamics over freeway networks using multiple graph heat diffusion kernels and integrate the kernels into DLM with Bayes' rule. We optimize the model parameters using Bayesian inference to minimize the prediction errors.The proposed model demonstrates prediction accuracy comparable to state-of-the-art deep neural networks with lower computational effort. It notably achieves excellent performance for long-term prediction through the inheritance of periodicity modeling in DLM.Chapter 4 proposes a deep neural network model to predict traffic features on large-scale freeway networks.These days, deep learning methods have heavily tackled traffic forecasting problems of freeway networks because they are outstanding at learning highly complex correlations between variables both in time and space, which the linear models might be limited to.Adopting a graph convolutional network (GCN) becomes a standard to extract spatial correlations; therefore, most works have achieved great prediction accuracy by implanting it into their architecture.However, the conventional GCN has the drawback that receptive field size should be small, i.e., barely refers to traffic features of remote sensors, resulting in inaccurate long-term prediction.We suggest a forecasting model called two-level resolution deep neural network (TwoResNet) that overcomes the limitation.It consists of two resolution blocks:The low-resolution block predicts traffic on a macroscopic scale, such as regional traffic changes.On the other hand, the high-resolution block predicts traffic on a microscopic scale by using GCN to extract spatial correlations, referring to the regional changes produced by the low-resolution block.This process allows the GCN to refer to the traffic features from remote sensors.As a result, TwoResNet achieves competitive prediction accuracy compared to state-of-the-art methods, especially showing excellent performance for long-term predictions.

EPFL2022

The Forward Physics Facility at the High-Luminosity LHC

Martin Bauer, Enrico Bertuzzo, Alexey Boyarsky, Francesco Cerutti, Emanuele Copello, Patrick Foldenauer, Ji Hyun Kim, Karan Kumar, Jing Li, Yiming Li, Wei Liu, Zhen Liu, Saurabh Nangia, Lesya Shchutska, Zheng Wang, Wenjing Wu, Yi Zhang

High energy collisions at the High-Luminosity Large Hadron Collider (LHC) produce a large number of particles along the beam collision axis, outside of the acceptance of existing LHC experiments. The proposed Forward Physics Facility (FPF), to be located several hundred meters from the ATLAS interaction point and shielded by concrete and rock, will host a suite of experiments to probe Standard Model (SM) processes and search for physics beyond the Standard Model (BSM). In this report, we review the status of the civil engineering plans and the experiments to explore the diverse physics signals that can be uniquely probed in the forward region. FPF experiments will be sensitive to a broad range of BSM physics through searches for new particle scattering or decay signatures and deviations from SM expectations in high statistics analyses with TeV neutrinos in this low-background environment. High statistics neutrino detection will also provide valuable data for fundamental topics in perturbative and non-perturbative QCD and in weak interactions. Experiments at the FPF will enable synergies between forward particle production at the LHC and astroparticle physics to be exploited. We report here on these physics topics, on infrastructure, detector, and simulation studies, and on future directions to realize the FPF's physics potential. 429 pages, contribution to Snowmass 2021.

2022

Pyramid Architecture Search for Real-Time Image Deblurring

Wei Liu, Kaicheng Yu

Multi-scale and multi-patch deep models have been shown effective in removing blurs of dynamic scenes. However, these methods still suffer from one major obstacle: manually designing a lightweight and high-efficiency network is challenging and time-consuming. To tackle this obstacle, we propose a novel deblurring method, dubbed PyNAS (pyramid neural architecture search network), towards automatically designing hyper-parameters including the scales, patches, and standard cell operators. The proposed PyNAS adopts gradient-based search strategies and innovatively searches the hierarchy patch and scale scheme not limited to cell searching. Specifically, we introduce a hierarchical search strategy tailored to the multi-scale and multi-patch deblurring task. The strategy follows the principle that the first distinguishes between the top-level (pyramid-scales and pyramid-patches) and bottom-level variables (cell operators) and then searches multi-scale variables using the top-to-bottom principle. During the search stage, PyNAS employs an early stopping strategy to avoid the collapse and computational issues. Furthermore, we use a path-level binarization mechanism for multi-scale cell searching to save the memory consumption. Our primary contribution is a real-time deblurring algorithm (around 58 fps) for 720p images while achieves state-of-the-art deblurring performance on the GoPro and Video Deblurring datasets.

IEEE2021

Linear regression analysis of regional mean speed of Athens city network using drone data: A multi-modal approach

Emmanouil Barmpounakis, Nikolaos Geroliminis

The work proposes a multi-modal regional mean speed regression analysis for the city network of Athens, Greece. The dataset from pNUEMA experiment is used in the present context. Accumulations and mean speeds of different modes are estimated and compared to each other. Besides, the mean running speeds of all modes are also computed and compared. The mean speed of each mode is assumed to be a linear combination of accumulations of all modes and regression analysis is performed. Ordinary Least Squares (OLS) and Partial Least Squares (PLS) methods are used in the current work. It is noticed that multicollinearity between different modes is weak in the current dataset. The quality of multi-modal regression fits is compared to the uni-modal ones, where the mean speed of a given mode is assumed to be a function of the accumulation of that mode only. It is concluded that multi-modal regression fits outperform their uni-modal counterparts in terms of R2 and Root Mean Squared Relative Error (RMSRE).

2021

Impacts of Metering-Based Dynamic Priority Schemes

Nikolaos Geroliminis, Raphael Ali Francis Lamotte

Several works published over the last two decades have shown for a stylized set-up with homogeneous users that metering-based priority (MBP) schemes may generate Pareto improving departure time adjustments similar to those induced by congestion pricing, but without any financial transaction. We investigate whether MBP (i) still generates significant savings and (ii) remains Pareto-improving, with various sources of heterogeneity (in schedule flexibility, desired arrival time, and capacity usage). We consider two types of schemes: one where the priority status is allocated randomly (R-MBP) and another (HOV-MBP), which only prioritizes users with small capacity usage (e.g., carpoolers). We find that the relative total cost savings of R-MBP decrease with heterogeneity in flexibility, but may increase with heterogeneity in desired arrival time. It fails however to be Pareto-improving, as nonprioritized users are almost systematically worse-off. HOV-MBP circumvents this issue by generating an ordering effect and a modal shift, which both contribute to a better distribution of benefits among users. Under favorable circumstances, they may even restore a Pareto improvement. Overall, MBP appears as a realistic way to alleviate congestion, scoring well both in terms of efficiency and social acceptability.

INFORMS2021

, ,

Perimeter control based on Macroscopic Fundamental Diagram (MFD) aims to meter the number of transferring vehicles at the periphery of the protected urban region in order to obtain the desired number of vehicles in that region. The advantage of perimeter control is less computational effort, while its drawback is that it may create long queues and delays at the perimeter of the controlled area. For capturing boundary queue dynamics, an enhanced accumulation-based MFD model is proposed using colored Petri Nets by considering transfer flows, boundary queues and travel delays simultaneously. The gated intersections and related road segments on the border of a protected region are modeled as so-called boundary buffers. Based on the enhanced MFD model, anintegrated perimeter control framework is proposed with consideration of travel time and queuing time in buffers. In this framework, the controllers between peripheral and protected region are optimized using model predictive control theory. Then, internal flow controllers are adopted to homogenize traffic density among subregions, and route guidance is also used to balance the number of queuing vehicles among boundary buffers. Simulation results verify the effectiveness of the proposed integrated perimeter control. Furthermore, the impacts of buffer storage capacity on region heterogeneity and trip completion rates are also investigated in this paper.

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC2021

Empirical investigation of the emission-macroscopic fundamental diagram

Emmanouil Barmpounakis, Nikolaos Geroliminis, Eric Justin Gonzales, Martí Montesinos Ferrer

Vehicle emissions are a major contributor of air pollution in urban areas, with severe consequences for public health and the environment. This paper focuses on analyzing vehicle emissions in a multimodal urban context to better understand its determinants and investigate relations between emissions and congestion patterns. Thousands of naturalistic trajectories collected with a swarm of drones during the pNEUMA experiment are analyzed and EPA's microscopic emission model, project level MOVES, is implemented to estimate vehicular emissions. The method is developed and illustrated with a CO2 analysis. The results empirically show the aggregated relationships between congestion and vehicular emissions, while accounting for the individual characteristics of each trajectory. These systematic relationships on a network scale with the use of empirical data, allow us to extend the macroscopic fundamental diagram (MFD) to the emissions-MFD (e-MFD). The spatiotemporal distribution of emissions is also studied, highlighting areas with high concentration over the urban area.

PERGAMON-ELSEVIER SCIENCE LTD2021

,

This paper investigates the accuracy of mean density estimation from direct sensing at link and network levels. Different calculation methods are compared depending on sensor type, probe vehicles or loop detectors, and availability to quantify the magnitude of expected errors. Probe data are essential to reduce the error but accurate density estimation requires high penetration rates, which is hardly true in practice. We enhance the fishing rate method, i.e. using the ratio of probes detected at the loop locations over the loop flow, to estimate density. Accurate density estimation at the link level can only be obtained when probes and loop data are available in real-time. At the network level, accurate density estimations can be obtained when combining loop and probe observations, even if few links capture both data sources. It requires applying the proper analytical formulation to aggregate the local observations, i.e. carefully defining fishing rates at this scale.

TAYLOR & FRANCIS LTD2021

The Stability of Low-Density Parity-Check Codes and Some of its Consequences

Wei Liu, Rüdiger Urbanke

We study the stability of low-density parity-check codes under blockwise or bitwise maximum a posteriori decoding, where transmission takes place over a binary-input memoryless output-symmetric channel. Our study stems from the consideration of constructing universal capacity-achieving codes under low-complexity decoding algorithms, where universality refers to the fact that we are considering a family of channels with equal capacity. Consider, e.g., the right-regular sequence by Shokrollahi and the heavy-tail Poisson sequence by Luby et al. Both sequences are provably capacity-achieving under belief propagation decoding when transmission takes place over the binary erasure channel. In this paper we show that many existing capacity-achieving sequences of low-density parity-check codes are not universal under belief propagation decoding. We reveal that the key to showing this non-universality result is determined by the stability of the underlying codes. More concretely, for an ordered and complete channel family and a sequence of low-density parity-check code ensembles, we determine a stability threshold associated with them, which gives rise to a sufficient condition under which the sequence is not universal under belief propagation decoding. Moreover, we show that the same stability threshold applies to blockwise or bitwise maximum a posteriori decoding as well. We demonstrate how the stability threshold can determine an upper bound on the corresponding blockwise or bitwise maximum a posteriori threshold, revealing the operational significance of the stability threshold.

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC2021

On the Universality of Low-Density Parity-Check Block Codes

Wei Liu, Rüdiger Urbanke

An important class of modern channel codes is the capacity-achieving sequences of low-density parity-check block codes. Such sequences are usually designed for the binary erasure channel and are decoded by iterative message-passing algorithms. In this paper we investigate the universality of such sequences. We show that many existing capacity-achieving sequences designed for the binary erasure channel do not achieve the same capacity of many other binary-input memoryless output-symmetric channels under belief propagation decoding.

IEEE2020