Search

This is the era of big-data: high-volume, high-velocity and high-variety information assets are being collected, demanding cost-effective information processing. Analytic techniques primarily based on statistical methods are showing astonishing results, but exhibit also limited reasoning capabilities. On the other end of the spectrum the era of big- reasoning is emerging with next-generation cognitive and autonomous end-To-end solvers. A problem description in terms of text and diagrams is given: problem solvers should automatically understand the problem, identify its components, devise a model, identify a solving technique and find a solution with no human intervention. We propose a challenge: To design and implement an end-To-end solver for mathematical puzzles able to compete with primary school students. Mathematical puzzles require mathematics to solve them, but also logic, intuition and imagination are essential ingredients, thus calling for an unprecedented integration of many different AI techniques.

The aim of this document is to indicate how the results of COLOMBO can be exploited. The document does not only focus on the duration of the project, but also assesses the possibilities beyond the end at 31st of October 2015. A good reference for the exploitability of the COLOMBO results is the Technology Readiness Level (TRL), which is used in the document to indicate the maturity and marketability of the results. COLOMBO detection algorithms were developed for several situations. Data fusion between green phase duration and travel times from for instance navigation systems, can give useful estimates of traffic volumes. Full cooperative vehicles can give both queue estimates and traffic flow estimations and a work is being carried out on a separate traffic anomalies detection algorithm. Apart from the detection and traffic control algorithms, COLOMBO also developed and improved simulation tooling to facilitate the work in the project. The document also reviewed exploitation from a stakeholder perspective. Road operators and governments will profit from the availability of a lower cost alternative to current traffic control solutions both in initial investment and maintenance. Road users benefit from more efficient traffic control, resulting in lower travel time and fuel savings. Having a cooperative system in the vehicle also guarantees the queue the road user currently is in, has been detected by the controller. Therefore, cooperative vehicles are expected to have a lower average travel time. Traffic engineers, consultants, industry and researchers benefit from the (open source) tooling released by COLOMBO for their work.

The aim of D2.3 is to show how the traffic control system proposed by COLOMBO in the previous D2.2 has been extended and how it reacts without full knowledge of the vehicles approaching the controlled intersection. At first, the developed extensions of the traffic light algorithms to deal with incomplete knowledge are presented. Two methods to estimate the real values of a traffic measure from observations of only a fraction of vehicles are given. The first estimates the amount of vehicles on the controlled roads using vehicle lengths and positions, the second attempts to determine the length of queues in front of the controlled intersections. A third method to cope with incomplete knowledge puts additional constraints to the selection of a traffic light phase, prohibiting to select the same one too often. The document then presents how the parameters for the traffic light were obtained off-line, this means using configuration process performed before evaluating the traffic light. The results obtained from subsequent evaluations are then presented. They show that the developed traffic light systems outperform conventional, fixed-cycle traffic lights for equipment rates above 10%.

The Deliverable 3.2 presents the technology behind the off-line configuration process used to find proper parameters for the traffic light algorithms presented in WP2’s Deliverable D2.3. The document gives an introduction into the topic of automatic configuration. It describes the main automatic algorithm configurators that have been considered for the COLOMBO tuning tasks, compares, their performance on available benchmark configuration scenarios, analyses in more detail the performance of the most promising configurator, the irace software package, and proposes various improvements to it. The deliverable then describes how to apply irace to automatic configuration tasks for the off-line configuration of the traffic light control software developed in WP2 and gives an in-depth experimental analysis of the performance we achieved by the automatic configuration on some traffic control scenarios that so far have been considered in the COLOMBO project.

International audience; Book Front Matter of LNCS 9249

In chapter 2, a literature research points out a lot of differences regarding to light signals between Austria, Belgium, France, Germany, Italy, Netherlands, and the United Kingdom. Normally the street user can handle the different light signals in a safe way, but when planning light signals certain differences could be recognised. In this chapter, a common and unambiguous glossary is defined, highlighting the concepts that will be used in the rest of the document. In chapter 3, a general overview of signal control methods and a classification after Fixed Time (time dependent) Signal Control, Actuated (traffic dependent) Signal Control and Adaptive Signal Control is presented. In chapter 4, time-dependent, fixed time signal control methods are presented, and a thorough explanation of the prerequisites for designing this kind of control methods is shown In chapter 5, the coordination of arterials and networks is properly explained. While this is a special case of time-dependent control, it deserved a full chapter because coordination is very important when controlling arterial streets, mostly in urban areas. In chapter 6, traffic dependent signal control methods are presented, with a peculiar emphasis on fully traffic dependent, actuated signal control. Chapter 7 introduces adaptive signal control systems, providing also some historical background on their development and adoption in Europe. The different approaches and results of several methods used all over the world are presented. Chapter 8 presents the traffic control method designed for the COLOMBO project, how it is split into a macroscopic level adaptive control system and into several microscopic level policies. Preliminary results on the usage of these policies compared to other policies already implemented into the simulation environment are reported.