Community Research and Development Information Service - CORDIS


ITERATE Report Summary

Project ID: 218496
Funded under: FP7-TRANSPORT
Country: Sweden

Periodic Report Summary 2 - ITERATE (IT for error remediation and trapping emergencies)

Project context and objectives:

The ITERATE project started at 1 January 2009 and ended at 31 December 2011. Its principal objective was to develop, test and validate a Unified model of driver (UMD) behaviour and specifically the model should consider drivers' interaction with support systems. The model should be applicable to all the surface transport modes (road, rail and water). ITERATE was based on the assumption that the underlying factors influencing human behaviour such as age, experience, gender, culture etc. are constant between transport modes. A UMD behaviour will be of great use when designing innovative technologies since it will allow for assessment and tuning of the systems in a safe and controllable environment without actually putting them to use in real traffic. At a concept stage, the model could guide designers in identifying potential problem areas whilst at the prototype stage, the model could inform on the scenarios to be used in system evaluation. In this way the systems will be better adapted to drivers' needs and abilities before being available on the market and provide better support to the driver even in critical situations. Along the same lines, the model could be of use for authorities to assess and approve innovative technologies without performing extensive simulator experiments or large scale field trials. The work was outlined as an iterative process theory ? experiments numerical simulation of the UMD validation experiments - refinement of the numerical UMD simulation. Thus, the ambition was to test the UMD model in several experiments and also use it as a foundation to build a numerical simulation of the model. Validation experiments conducted as a means to further test and develop the UMD model and the numerical simulation. The ITERATE work was built on previous work done in e.g. the AIDE project.

Finally, an important objective was to build a database with data from the various experiments carried out in the project. This database will be utilised by the consortium even after the project ends. The database will be made public to the research community two years after the project ended (i.e. 2014) as described in a memorandum of understanding (MoU) signed by the project consortium. Thus, the objectives were in short to develop and test a UMD behaviour, build a numerical simulation based on the model, conduct experiments as a means for an iterative development and finally, build a database to be used beyond the project lifetime.

Project results:

The project consortium consisted of seven partners from five countries. The project was split into nine Work packages (WPs) including management and dissemination. Management was done in WP0. WP1 (Critical state of the art and UMD) contained an extensive and critical literature review on driver behaviour models and theories (D1.1). The most suitable type of models was found to be motivational models and specifically a model developed in the AIDE project was selected for further development. The model was revised and extended to cater for factors like culture, personality, driver state, experience and workload. Furthermore, it was based on the assumption that factors mentioned would influence driver behaviour on a similar way independent of transport mode (road, rail and water). The model developed was called UMD behaviour and a description can be found in D1.2. The second WP (Selection of system) carried out a survey of driver support systems available for the three modes of transport addressed (D2.1) and made a selection of systems for experiments (D2.2). The third WP (Experimental design and scenario specification) was based on input from the two preceding WPs and conducted necessary preparations for the experiments. Thus, it included the development of scenarios, scales and measures to be used as well as selection criteria for participants (D3.1).

Furthermore, two identical portable simulators (car and train) were built. The portable simulators were quite simple with two Liquid crystal display (LCD) screens, sound system and gaming type of controls for the drivers. They were circulated among partners in order to ensure that all experiments were performed with the same hard- and software setup. One of the simulators was demonstrated at the Human modelling in assisted transportation (HMAT) workshop in June 2010. This was also the start of WP4 (Experiments). In total 183 car driver and 110 train drivers from five different countries participated in the experiments. Two full-scale simulators (car & train) using the same software as was used in the portables were used to validate the results from the portable simulators. A detailed description of the experiments can be found in D4.1. The results from the experiments were analysed in WP5 (Analysis), partly in parallel to the execution of the experiments but mainly after completion of WP4. The main results of the analysis, published in D5.1, were that country and gender were found to be important factors for both car and train drivers. Thus, it was concluded that gender should be included in the model. Significant effects of experience, personality and workload were also established in many of the events analysed. However, driver fatigue was not found to affect driver behaviour which was attributed to a too low level of fatigue. A comparison between the portable and the full scale simulators showed that portable simulators are suitable tools to study driver behaviour even if there is a scaling factor to consider. WP6 (Model numerical development and tuning) was running in parallel with WP5. The UMD model was used as a framework when developing the simulation software which consists of two parts: Simulator Daemon (SimUD) and driver vehicle and environment simulator (DVESim). SimUD (simulation of the UMD) is the actual implementation in a numerical set of expressions and software implementation tool of the UMD theoretical representation. DVESim is the software platform that enables the representation of the interactions between a SimUD and the rest of the actors involved in a scenario development. The UMD factors (personality, experience, state and workload) were given discrete values depending on the driver's character. Culture was considered separately. A comparison was made between the experimental results from WP4 and the results achieved with the numerical simulation and it was found to provide comparable results, specifically for speed control. The first results identified some areas for improvements and further tuning. The results of WP6 work is primarily described in three deliverables (D6.13) but also in WP7 deliverables (final version of the software). WP7 aimed at validating previous findings and to further develop the numerical simulation. Three experiments were carried out new transport mode (ship), new simulator platform and system (train) and new support system and task (car). The results show that model is still valid but primarily for car and train divers but less for ?drivers? of ships. The results are published in D7.1 and 7.2. Deliverable 7.3 (Final report) is still pending and will contain a summary of the project results and achievements. Dissemination was carried out in WP8 and consisted of a project webpage, newsletters, HMAT workshop, a final public webinar, several presentations at conferences and publishing in journals. The WP8 work is described in D8.1 4.

Potential impact:

The final results of ITERATE were in short:
- a theoretical model of driver behaviour applicable for different transportation modes UMD;
- a database built on the simulator experiments carried out with close to 400 entries to be exploited beyond the project's lifetime;
- a numerical model of UMD implemented as a simulation tool;
- a book from the HMAT workshop 'Human modelling of assisted transportation' published by Springer;
- more than 20 publications (journals and conference proceedings) based on project work.

Project deliverables

The actual socioeconomic impact of the project is hard to estimate as the nature of the project is mostly theoretical and not so close to production and implementation. However, a simulation tool like the one developed in ITERATE has a potential to improve the development of advanced driver support systems and thus contribute to both safety and mobility aspects of driving behaviour.

ITERATE results could also have an impact on traffic safety. Even if the road fatalities in the European Union (EU) tend to decrease it seems like the target for 2020 (halving the overall number of road deaths starting from 2010) will be hard to reach. Substantial efforts are made not the least in terms of introducing driver support systems with the aim to improve safety. However, the use of driver support system can have unexpected adverse effects depending variations in driver behaviour which can be guided by human factors considered the UMD model. Thus, the ITERATE work is expected to contribute to the development of test methods and tools for safe guarding driver support systems to emerge on the market.

There is a demand from industry for tools and methods to test new systems before they put on the market. Furthermore, traffic safety authorities could benefit from the use of a UMD behaviour for assessing and approving innovative technologies without performing extensive simulator experiments or large-scale field trials.

In this way we believe that ITERATE can contribute to the EU road safety target for 2020 or beyond.

List of websites:

Related information


Björn PETERS, (Research Leader, M. Sc, PhD)
Tel.: +46-132-04070
Fax: +46-131-41436
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top