Javier Sanchez-Medina

Traffic Signal Optimization in “La Almozara” District in Saragossa Under Congestion Conditions, Using Genetic Algorithms, Traffic Microsimulation, and Cluster Computing

Urban traffic congestion is a pandemic illness affecting many cities around the world. We have developed and tested a new model for traffic signal optimization based on the combination of three key techniques: 1) genetic algorithms (GAs) for the optimization task; 2) cellular-automata-based microsimulators for evaluating every possible solution for traffic-light programming times; and 3) a Beowulf Cluster, which is a multiple-instruction-multiple-data (MIMD) multicomputer of excellent price/performance ratio. This paper presents the results of applying this architecture to a large-scale real-world test case in a congestion situation, using four different variables as fitness function of the GA. We have simulated a set of congested scenarios for ¿La Almozara¿ in Saragossa, Spain. Our results in this extreme case are encouraging: As we increase the incoming volume of vehicles entering the traffic network - from 36 up to 3600 vehicles per hour - we get better performance from our architecture. Finally, we present new research directions in this area.

Suspended traffic lights detection and distance estimation using color features

Traffic Light Detection is a problem differently approached by many research groups around the world. Here we present a novel technique to detect suspended traffic lights, based on colors and features such as black area of traffic lights or area of lighting lamps. Additionally, the traffic light distance is estimated aiming at slowing down and stopping in the correct position, in case of red light. Some preliminary test results are presented to assess both the detection rate and the distance estimation.

Micro flip teaching An innovative model to promote the active involvement of students

The use of the flip teaching methodology, a new trend in educational innovation, has had a significant impact. It has led to the creation of spaces that provide online video for use in the classroom while also encouraging the active participation of students. However, the implementation of this methodology can be problematic in the classroom. A new micro flip teaching module has been designed and implemented to resolve these issues. The main objectives of this research were to measure the impact of learning, to determine the degree to which students are involved in the process through the creation of learning resources and to measure how participating students view their experience of micro flip teaching. This model incorporates answers to issues that currently pose a barrier, such as link activity or the major effort that would be required to change an entire course design. The results show that the micro flip teaching model has a direct impact on student learning. The study offers proof that the model is not in any way subject-dependent nor does it require a great effort for students to adapt. Student perception of the usefulness of the model is based more on the methodology itself than on either course content or the teachers participating in the experience.

How to Simulate Traffic with SUMO

In this paper we want to give a quick introduction about how to install and use SUMO to build a first simulation in a few steps.

Holistic Calibration of Microscopic Traffic Flow Models: Methodology and Real World Application Studies

This study proposes and applies a methodology to calibrate microscopic traffic flow simulation models. The proposed methodology has the capability to calibrate simultaneously all the calibration parameters as well as demand patterns for any type of network. Parameters considered include global and local as well as driver behaviour and vehicle performance parameters. Demand patterns, in terms of turning volumes, are included in the calibration framework. Multiple performance measures involving link counts and speeds are used to formulate and solve the proposed calibration problem. In addition, multiple time periods were considered. A Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm is used to search for the vector of the model’s parameters that minimizes the difference between actual and simulated network states. (Punzo V, Ciuffo B, Montanino M Transp Res Rec J Transp Res Board 2315(1):11–24 2012, Punzo et al. [1]) commented on the uncertainties present in many calibration methodologies. The motivation to consider simultaneously all model parameters is to reduce that uncertainties to a minimum, by leaving to the experience of the engineers as little parameter tuning as possible. The effects of changing the values of the parameters are taken into consideration to adjust them slightly and simultaneously. This results in a small number of evaluations of the objective function. Three networks were calibrated with excellent results. The first network was an arterial network with link counts and speeds used as performance measurements for calibration. The second network included a combination of freeway ramps and arterials, with link counts used as performance measurements. The third network was an arterial network, with time-dependent link counts and speed used as performance measurements. The experimental results illustrate the effectiveness and validity of this proposed methodology. The same set of calibration parameters was used in all experiments.

Assessment of Economic Impacts of Vehicle Miles Traveled Fee for Passenger Vehicles in Nevada

This study evaluated the effectiveness and equity of a fee for vehicle miles traveled (VMT) for passenger vehicles in Nevada. In the evaluation of the fees effectiveness, the collection capabilities, as well as the fees impact on the number of miles users drove, were considered. Equity was evaluated by considering the impact of the VMT fee on various population groups on the basis of socioeconomics, demographics, household type, location, and ownership of fuel-efficient vehicles. To estimate the impacts on various VMT fees, a linear regression model was developed with the use of data from the 2009 National Household Travel Survey, which provided a mechanism to estimate VMT in Nevada as a function of the cost to drive, among other characteristics. The effectiveness and the equity of two alternative VMT fees were compared with the existing fuel tax system. These fees were calculated on the basis of the average fuel efficiency of vehicles in Nevada and the historical revenue from the state fuel tax. In general, a VMT fee of 3.3 cents per mile seemed to be more effective than both the existing fuel tax and a VMT fee of 2.91 cents per mile. Although the 3.3 cents per mile fee had a slightly greater impact on various population groups, its equitable distribution of the tax burden among 71.1% of households created a small average cost increase of just 0.37% per household. Thus, a 3.3 cents per mile fee would provide the necessary revenue without significantly affecting Nevada households.

User equilibrium study of AETROS travel route optimization system

We have designed a new urban travel route optimization model in combination with a new microscopic simulation paradigm. The core of the system stands over the assumption that we can have a traffic network with an Advanced Traffic Information System (ATIS) installed that advises drivers at every intersection as to which direction to take, depending of their destination. We have implemented a parallel Genetic Algorithm, running over a Beowulf cluster that performs a real-time optimization yielding that set of advised routes. By these means we propose an implicitly adaptable, expandable and flexible solution for the traveler route assignment problem, in order to maximize the performance of an urban network. Although in Dynamic Traffic Assignment (DTA) a rigorous Wardrops User Equilibrium --- [6] --- cannot be fully carried out, we have developed a set of experiment aiming to see if there is a visible tendency to a User Equilibrium state for the proposed model. Results are encouraging.

Optimized Echo State Network with Intrinsic Plasticity for EEG-Based Emotion Recognition

Reservoir Computing (RC) is a paradigm for efficient training of Recurrent Neural Networks (RNNs). The Echo State Network (ESN), a type of RC paradigm, has been widely used for time series forecasting. Whereas, few works exist on classification with ESN. In this paper, we shed light on the use of ESN for pattern recognition problem, i.e. emotion recognition from Electroencephalogram (EEG). We show that the reservoir with its recurrence is able to perform the feature extraction step directly from the EEG raw. Such kind of recurrence rich of nonlinearities allows the projection of the input data into a high dimensional state space. It is well known that the ESN fails due to the poor choices of its initialization. Nevertheless, we show that pretraining the ESN with the Intrinsic Plasticity (IP) rule remedies the shortcoming of randomly initialization. To validate our approach, we tested our system on the benchmark DEAP containing EEG signals of 32 subjects and the results were promising.

Safe and Reliable Path Planning for the Autonomous Vehicle Verdino

This paper introduces a local planner which computes a set of commands, allowing an autonomous vehicle to follow a given trajectory. To do so, the platform relies on a localization system, a map and a cost map which represents the obstacles in the environment. The presented method computes a set of tentative trajectories, using a schema based on a Frenet frame obtained from the global planner. These trajectories are then scored using a linear combination of weighted cost functions. In the presented approach, new weights are introduced in order to satisfy the specificities of our autonomous platform, Verdino. A study on the influence of the defined weights in the final behavior of the vehicle is introduced. From these tests, several configurations have been chosen and ranked according to two different proposed behaviors. The method has been tested both in simulation and in real conditions.

Acceleration Signal Based Linear Formation Driving Model: Algorithmic Description and Simulation Results

Platoon vehicles coordination is an important topic at Intelligent Transportation Systems these days. For coordinating two or more vehicles, it is needed both a communication method between them, and an environment sensing strategy enabling the triangulation and correction, when needed, of each vehicle relative position within the desired platoon.