Search

New approaches to control stochastic non-linear time-variant processes include the application of machine learning techniques. One of the problems with learning- based controllers is their reliability in a wide area of process parameters as the controller is trained using a limited set of representative scenarios, either chosen by the designer or taken from historic records. Thus, reliable controller behavior can be guaranteed only in scenarios applied during controller training. Due to the very larger number of random variables and possible scenarios, not all variations can be applied in the controller training process using simulators to guarantee good controller behavior when applied in a real system. One case is traffic control (signal programs, variable speed limit, ramp metering) having large travel patterns variety. The concept of Structured Simulations Framework (SSF) can cover most probable learning scenarios. Thus, applying SSF enables a systematic controller training approach by complementing existing scenarios with synthesized ones that evoke or replicate substantial aspects of real traffic. Such training is necessary to ensure reliable learning-based controllers. This paper discusses the concept of applying SSF to ensure the reliability of learning-based controllers and proposes the application in traffic control for the case of variable speed limits on motorways.

Intelligent Transportation Systems (ITS) offer a solution for alleviating congestion in urban areas based on its traffic control service. It is part of the build + ITS approach where new infrastructure with different ITS services is applied to improve traffic throughput, reduce vehicle emissions, increase the Level of Service (LoS), and improve the usage of existing vehicles with the Mobility as a Service (MaaS) concept. The latter enables the cities to offer mobility users concepts of car-sharing, ride-sharing, and tailored multi-modal travel patterns. The city or urban road infrastructure can be divided into urban motorways and networks of signalized intersections. Traffic control approaches related to Variable Speed Limit (VSL) control, ramp metering, lane change recommendation, and Adaptive Traffic Signal Control (ATSC) are applied to manage the urban road infrastructure.

To cope with today’s problems with congestions in urban motorways and the inability to expand their capacity in an urban environment, new solutions in the form of advanced control methods on the existing road infrastructure are applied. Such solutions are part of Intelligent Transportation Systems (ITS). ITS essentially integrates information and communication technologies in order to resolve mentioned congestion problems. Often used ITS based traffic control methods on urban motorways are ramp metering (RM) and variable speed limit control (VSLC). A dedicated algorithm for RM or VSLC uses sensed data form urban motorway to compute actions that will make a positive impact on traffic flows. This study will place focus on cooperative approach between RM and VSLC, and integration of several different RM algorithms into the one algorithm – INTEGRA. Mentioned algorithm is created by using adaptive neuro- fuzzy inference system (ANFIS) as the one of machine learning techniques. INTGERA is furthermore expanded in other to integrate its original functionality with recurrent neural network for traffic demand prediction. As the final step, evaluation of different criteria function setups for learning data set design based on which ANFIS neural network of INTEGRA is learned, is presented and compared with previously mentioned approaches. Results of all mentioned approaches will be compared and discussed in relation with other commonly used urban motorway control methods.

An Adaptive Neuro-Fuzzy Inference System (ANFIS) is a neural network-based fuzzy inference system that includes combination of two soft-computing methods: Artificial Neural Networks (ANN) and fuzzy logic Fuzzy logic has the ability to transform the qualitative aspects of human knowledge and insights into the process of precise quantitative analysis. However, it is very problematic to transform the human thought into a rule based Fuzzy Inference System (FIS), and adequately adjust Membership Functions (MFs) of the mentioned FIS. ANFIS uses ANN’s ability of self- adaptation to the environment through the machine learning process in order to automatically adjust the MFs, and reduce the rate of errors in the determination of rules in FIS. Fuzzy logic based approaches such as FIS are often used for ramp metering. Ramp metering as one of the control methods for urban motorways is formulated as the regulation of the on-ramp flow access rate into the motorway mainstream according to the several inputs. Most ramp metering algorithms based on fuzzy logic require a robust and comprehensive approach for adjusting of the FIS rule base and MFs in a complex non-linear environments such as the urban motorway traffic system.

Uviđaji prometnih nesreća su prometni i tehničko- tehnološki procesi koji u svojoj osnovi sadržavaju dva međusobno suprotstavljena zahtijeva. Prvi zahtjev odnosi se na potrebu vremenski što kraćeg raščišćavanja mjesta prometnih nesreća, kako bi se spriječili sekundarni incidenti, a prometnica vratila u puni operativni kapacitet za odvijanje prometa i kako bi se prometni tok u potpunosti normalizirao. Drugi zahtjev odnosi se na provedbu očevida kao istražne radnje čija je svrha utvrđivanje postojanja ili nepostojanja elemenata kaznenog djela izazivanja prometne nesreća ili prometnog prekršaja. Provedba očevida danas, vremenski je zavisan proces koji zahtjeva relativno duži vremenski period kako bi se prikupili dokazi, obavila potrebna mjerenja, opisali tragovi i sl. U ovom radu prikazati će se mogućnosti digitalizacije i izrade trodimenzionalnih 3D modela mjesta prometne nesreće. Opisati će se inovativne tehnologije, kao što je izrada trodimenzionalnih (3D) računalnih modela stvarnog mjesta prometne nesreće uz pomoć́ bespilotne letjelice i fotogrametrijskih postupaka.

U radu je opisan značaj i mogućnost razvoja različitih usluga iz područja inteligentnih transportnih sustava (ITS), a u svrhu unapređenja kvalitete usluge e-pokretljivosti, a prije svega prema krajnjim korisnicima električnih vozila. Ukazano je na značaj ovog pristupa u odnosu na budući razvoj različitih infrastruktura uključenih u e-mobilnost. Dat je prikaz modela i odgovarajuće arhitekture za punjenje električnih vozila i njima nadograđene usluge prema krajnjim korisnicima – vozačima.

This poster presents the application of Variable speed limit control (VSLC) as the urban motorway control method. Mentioned control method presents an efficient way to increase the Level of Service (LoS) and safety of urban motorways. Main impacts of VSLC on traffic flow are reduction of the mean speed under critical densities („shock-wave suppression”), homogenization of vehicle speeds, and increased safety on motorway. Intelligent speed adaptation (ISA) system represents augmentation of described VSLC. ISA monitors the location and speed of the vehicle, compares it to a defined posted speed, and takes corrective action such as advising the driver and/or governing the top speed of the vehicle. This poster proposes innovative VSLC infrastructure-to- vehicle communication concept along with the new approach which enable modeling of ISA penetration rate within ACTM based macro-simulation environment. According to the initial simulation results, it is possible to conclude that increase of ISA penetration rate improves the overall results. Furthermore, it is possible to notice that that greatest difference between travel time and delay without use of ISA and with penetration rate of ISA up to 25 percent. All simulation results are conducted under the neutral prognosis of speed limit obedience among drivers. This poster is presented at The BMW Summer School in Bad Wörishofen, Bavaria, Germany, July 09-14, 2017.

Urban motorways serve the traffic demand from adjacent dense populated urban regions and transit traffic. Combinations of these two traffic demand types generate periodically traffic congestions during the peak hours. The capacity drop effect is the most common cause of congestions. It occurs when the current traffic flow reached its maximal capacity of the observed urban motorway section. In order to alleviate the capacity drop effect and enable better level of service on motorways new traffic control methods are introduced. One of the used traffic control methods is variable speed limit control (VSLC). VSLC conducts homogenization and reduction of vehicle speeds what consequently decreases time needed to create a critical shockwave, which will induce the capacity drop. This paper presents a short overview of the latest VSLC algorithms originally designed for mitigation of traffic congestions and their effects on the motorway mainline. Firstly, simple reactive VSLC algorithms are introduced. Then approaches based on model based optimization are described. Possible coordination between VSLC and other motorway management strategies is also described in this paper.

Prezentacija rada djelatnika Fakulteta prometnih znanosti u programskim alatima PTV Visum i PTV Vissim.

The number of traffic participants has significantly increased in recent decades. Such a trend is also evident in road traffic. The result is that today’s urban highways are under influence of the now increased traffic demand and cannot fulfil desired level of service anymore. Classical solution to this problem is to build new road infrastructure (urban and intercity highways, bypass roads and new segments of urban roads). That only increases the traffic demand following the saying: “If you built it, they will come!” Additional problem is that in most cases there is no more space available for infrastructure build up. Daily significant congestions appear, mostly in dense populated urban areas. To cope with the heavy congestions, new traffic control approaches are used. These approaches are solutions (services) from the domain of Intelligent Transport Systems, such as ramp metering, variable speed limit control, adaptive control of traffic lights, optimal control for consecutive intersections, etc. In this paper the control of on-ramp traffic flow on urban highways known as ramp metering is examined. To cope with significant daily changes of traffic demand various approaches with autonomic properties like self-learning are applied for ramp metering. One of the approaches with this property is reinforced learning. In this paper the Q-learning algorithm is applied to learn a local ramp metering control law in a simulation environment implemented in the microscopic simulator VISSIM. Proposed approach is tested in simulations with emphasis on a real world situation (typical part of a working day) containing significant changes in traffic demand.

The scope of the poster is to present an interface between the structured simulation framework and the VISSIM traffic simulator. Mentioned interface is defined mainly through discussion with the host prof. René Schumann, PhD from HES-SO University, and with his associates. He has received a grant for a project named "Structured Simulation: A Framework for the Automated Analysis of Adaptive Systems". The project's aim is to create a generic structured simulation framework that can be used for a different kind of the simulations such as traffic simulation, smart grid simulation and social simulation. From the traffic engineering perspective, the structured simulation framework can provide most noticeable contribution in evaluation of self-adaptive traffic controllers, implemented in a complex traffic network. Highway system will be initially controlled by the Variable Speed Limit Control (VSLC).

Urban highways have undergone a transformation in the recent years due to increased traffic demand. Optimal usage of the existing road transport infrastructure has become a priority. To ensure efficient usage of the existing traffic system capacity, intelligence needs to be added to its control systems. Such control solutions are from the domain of intelligent transportation systems (ITS). Mostly used solutions are ramp metering and variable speed limit control (VSLC). This paper presents a new learning based cooperative ramp metering strategy in which several well-known ramp metering strategies (ALIENA, SWARM, HELPER) are used to create a learning set and then create an ANFIS based cooperative ramp metering controller. Proposed urban highway control approach is tested in simulations using an augmented version of the CTMSIM simulator, traffic data for a typical working day and city of Zagreb bypass as case study. Obtained results are compared to standard ramp metering approaches and VSLC using travel time, Delay, and maximum vehicle queue length as quality measures.

In recent decades a significant increase in traffic demand has occurred. This trend is specially present in dense populated areas where daily traffic congestions during rush hours occur regularly. Congestions are significant in road traffic where they can simultaneously reduce public transportation level of service (LoS) also. As consequence, even more people are using their car additionally increasing the congestion problem. Classic solution for solving the road traffic congestion problem is infrastructure build up. Today’s dense urban areas mostly do not allow this approach because of the lack of available building space. More advanced road traffic control solutions from the domain of intelligent transportation systems (ITS) are being more and more applied to optimally use the existing infrastructure (Papageorgiou et al., 2003.). Such solutions include coordination between several consecutive crossroads, dynamic traffic assignment, driver informing systems, etc. One of the ITS application areas is related to urban highways which present a class of highways used as a city bypass or are just passing a dense urban are. Their main characteristic is that they have a larger number of on- and off-ramps often placed at small distances. Due to the small distance, mutual on-ramp influence can occur enlarging the problems of daily congestions and associated decrease of highway LoS. In order to prevent the appearance of traffic standstill or to reduce its duration control approaches as ramp metering and variable speed limit control (VSLC) are being applied (Hegyi et al., 2010.). In recent years, new cooperative concepts between several on-ramps and VSLC are used as a combined urban highway control system (Ghods et al., 2007.). This paper presents a new learning based cooperative ramp metering strategy in which several well-known ramp metering strategies (ALIENA, SWARM, HERO) are used to create a learning set for an ANFIS (Gregurić et al., 2013.) based control structure. Optimal ramp metering values are obtained for a wide range of traffic demand on the urban highway and belonging on-ramps. Optimal ramp metering values for specific traffic demand characteristics obtained from most suitable ramp metering strategies are integrated into only one control strategy. Thus, the need of applying several ramp metering strategies and switching between them is avoided. Additionally, cooperation between VSLC and vehicle control by an on-board unit is described and a discussion about possible implementation is given. Proposed cooperative urban highway management approach is tested in simulations using the city of Zagreb bypass as case study. For simulation, the macroscopic highway traffic simulator CTMsim (Kurzhanskiy et al., 2008.) is used. Used CTMsim simulator augmented to enable simulation of VSLC and cooperative ramp metering approaches.

Intelligent traffic control systems tackle today’s most significant problems in road transport, i.e. periodical road congestions. To ensure minimal losses related to congestion and optimal use of existing road infrastructure new approaches from the domain of intelligent transport systems (ITS) are applied. Today’s road traffic problems emphasis the need of cooperation between more standalone systems to achieve desired optimal use of the existing infrastructure. Development of such intelligent cooperative control systems presents an open challenge today. In order to implement such approaches, reliable measurements of traffic parameters and appropriate control algorithms are required. Trend in today’s traffic control systems is an broad increased use of cameras as the main sensor. From the video footage of traffic surveillance cameras several traffic parameters can be estimated like flow, vehicle gap and type, traffic speed, origin-destination matrices, etc. In this paper new approaches to road traffic control based on intelligent methods and computer vision proposed by the authors are described and newest acquired results are presented. For testing of the system real road traffic video footage was used from Croatian highways. Use case for testing of the proposed ramp metering framework was bypass of the city of Zagreb, Croatia. Section between problematic nodes with many on- and off-ramps was simulated in an augmented version of the macroscopic simulator CTMSIM.

Upravljanje vršnim opterećenjima u gradskoj cestovnoj mreži moguće je vršiti na njenim čvorištima odnosno, raskrižjima. Dinamičkim promjenama signalnog plana raskrižja ovisno o prometnim opterećenjima na prilazima moguće je postići bolju protočnost raskrižja. Dizajnirani simulacijski model upravljanja izoliranim raskrižjem uključuje nekoliko međusobno povezanih modula. Na temelju dobivenih podatka od odgovarajućih modula simulacije, baze znanja i mehanizama neizrazitog zaključivanja sustav neizrazitog zaključivanja određuje potrebno produljenje zelenog svijetla u trenutnoj fazi. Rezultati simulacije su pokazali zadovoljavajuću propusnost raskrižja, pri scenariju većeg opterećenja na jednom prilazu i umjerenog opterećenja na preostalim prilazima. Takav scenarij karakterističan je prilikom dnevnih vršnih opterećenja čiji negativan utjecaj na prometnu mrežu želimo smanjit.

Upravljanje širim prostorom prometnog sustava autoceste u užem urbanom području sastoji se od dvije sastavnice. Prva sastavnica je upravljanje tokom same autoceste, a druga je upravljanje tokovima arterijskih regionalnih i lokalnih cesta direktno ili indirektno vezanih za autocestu. Za te potrebe razvijena je posebna tehnika poznata pod nazivom na engleskom jeziku Ramp metering, koja se odnosi na upravljanje interakcijom priljevnih tokova arterijskih cesta s glavnim tokom autoceste. Za hrvatski naziv ove tehnike koristi se naziv: upravljanje priljevnim tokovima. Općenito se može zaključiti kako upravljanje priljevnim tokovima predstavlja korištenje odgovarajućih algoritama u određivanju signalnog plana upravljačkog sustava (mjerenja značajki prometnih tokova, obrada signala i upravljanje, posebni prometni semafori) koji upravljaju stupnjem propuštanja priljevnih tokova u glavni prometni tok. Osnovni cilj upravljanja priljevnim tokovima je zadržavanje optimalne operativne protočnosti autoceste, kontroliranjem prometne potražnje koja se generira na prilazima autoceste. Održavanje optimalne operacijske propusnosti trase autoceste sprječava nastajanje zagušenja i time stvara uštede u vremenu putovanja trasom. U diplomskom radu potrebno je razviti algoritam za zagrebačku obilaznicu te simulacijom na računalu provjeriti njegove značajke. Ključni dijelovi diplomskog rada su: UVOD, INTELIGENTNI TRANSPORTNI SUSTAVI I FUNKCIONALNA PODRUČJA, TEMELJNE ZNAČAJKE SUSTAVA UPRAVLJANJA PRILJEVNIM TOKOVIMA, PRIMJENA ADAPTIVNOG NEURO- FUZZY UPRAVLJANJA, MODELIRANJE I SIMULACIJA PROMETA NA ZAGREBAČKOJ OBILAZNICI, OPIS REZULTATA, ZAKLJUČAK

U radu je dan prikaz mogućnosti primjene upravljanja priljevnim tokovima auto-ceste (ramp metering)zasnovan na inteligentnom upravljanju, gdje su korištene umjetne neuronske mreže i algoritmi neizrazitog upravljanja. Opisane su temeljne značajke upravljanja priljevnim tokovima, kao usporedna analiza postojećih algoritama. Razmotrene su mogućnosti njene primjene na zagrebačkoj obilaznici.

U radu se daje pregled dosadašnjeg razvoja inteligentnih transportnih sustava u Republici Hrvatskoj. Posebno je istaknut razvoj u sustavu visokog obrazovanja i gospodarstva. Prikazani su primjeri radova studenata i poslijediplomanata u ovom području.