Skip to main content

Applicability in Transport and Traffic of Artificial Intelligence

Objective

The main goals of the project are:
- The identification of potential areas in traffic engineering to which AI may be beneficially applied;
- To give the traffic engineer a helpful guide which includes criteria and characteristics of AI methods and techniques in non-specialist language;
- To supply useful information to other DRIVE tasks dealing with the possible application of AI methods in road traffic engineering and also to maintain a close cooperation between the more application oriented projects within DRIVE such as 6205 and 6525.
The objectives of the project are the identification of potential areas in traffic engineering to which artificial intelligence (AI) may be beneficially applied and to give the traffic engineer a helpful guide which includes criteria and characteristics of AI methods and techniques in a common language. These methods will be assessed with respect to traffic engineering feasibility.
Current traffic engineering applications have to deal with incomplete, inconsistent uncertain and rapidly changing data. The still increasing traffic load aggravates the above mentioned situation and requires very complex traffic control systems.
Artificial intelligence research aims to a new way of representation and handling of fuzzy and symbolic knowledge which seems to be more convenient to humans. Up to now the information exchange between traffic engineers and AI researchers was insufficient due to the discrepancy of the description techniques of both domains. An open framework for interdisciplinary cooperation should push along the integrated application of AI methods and techniques in traffic engineering problem areas as follows: flexible control structures with the capability to adapt to rapidly changing circumstances, improvement of the reliability and consistency of traffic data, suitable man machine interfaces for better knowledge acquisition, better classification and prediction of relevant traffic events, correlations between traffic patterns and causes and behaviour oriented simulation of individual driver decisions.

Future systems in traffic control, traffic information management, transportation planning, fleet management and traffic safety will be more and more computer based to meet the requirements for speed, comfort and safety. The aim of this research was to evaluate the potential of artificial intelligence (AI) based software approaches to traffic engineering (TE) problems which can be solved only inadequately by conventional methods or which have had to be excluded previously from computerization. The research set out to analyse the requirements which should be present in considered traffic problem areas to ensure a beneficial application of artificial intelligence to enhance conventional techniques or to tackle completely new domains. A decision tree has been set up which leads to the identification of classes of AI applications which are similar or analogous to a given TE problem. This classification was derived from generic problems to which AI is typically applied and of which practical experienced from related areas is available. Together with this, a concise set of criteria and prerequisites concerning solution space, computational complexity, uncertainty of input parameters, presence of heuristics etc, was collected to support the traffic engineer in identifying problems which are suitable for applying AI. A major result is the proposal of several protoypical projects which exhibit the best chances in proving artificial intelligence as a powerful tool for solving current open questions or for enhancing inadequately realized existing systems.
Road traffic engineers will work closely together with AI experts to build up a common base of knowledge and understanding. This will be necessary for cooperation not only in this project but in all DRIVE tasks dealing with the application of Artificial Intelligence.
Main Deliverables:
Survey of potential use of AI techniques for traffic control.

Coordinator

Universität Fridericana Karlsruhe (Technische Hochschule)
Address
Haid-und-neu-straße 10-14
76131 Karlsruhe
Germany

Participants (4)

Automa Scrl
Italy
Centre d'Études Techniques de l'Équipement Méditerrannée (CETE)
France
Heusch-Boesefeldt GmbH
Germany
Address
Kleine Johannisstraße 9
20457 Hamburg
UNIVERSITY OF LEEDS
United Kingdom
Address
Woodhouse Lane
LS2 9JT Leeds