SIMUSAFE (SIMUlator of Behavioural Aspects for SAFEr Transport)

Simusafe - Simulation of behavioural aspects for safer transportSimuSafe, Simulator of behavioural aspects for safer transport


File: 723386

The goal of SIMUSAFE (SIMUlator of Behavioural Aspects for SAFEr Transport) is to make use of state-of-art Simulation, Artificial Intelligence, Virtual Reality and Data Science methodologies to retrieve accurate actor and behavioural models in a transit environment, reproduce the same into controllable settings of Traffic Simulators and be able to determine cause, consequences on incidents of interest, to understand the underling behaviour and motivations of the involved actors.

Impacting factors causing an event (crash, near-collision, infractions) from the environment and road users will be identified and quantified. This will enable the evaluation of scenarios which are not possible with naturalistic driving. The same principle will be employed to comprehend the impact of Altered Driving Conditions (ADC) caused by substance usage, distractions and psychological factors as well changes in driving behaviour and decision making due safety devices and autonomous assistance driving systems. Moreover, modelling and simulating realistic human behaviour can be a useful tool for research and improvement of autonomous driving systems.

Such knowledge will be the base for the development of more effective and pro-active measures for the mitigation of such factors, with subsequent impact in the safety devices market, regulations and driver education.

SIMUSAFE project will have its efforts concentrated in the following topics:

  1. Model Development and Data Collection – develop an Actor Model for representation and measurement of risk-taking and risk potential at individual state from actor data (biometric, vehicular, and environmental) and underlying infrastructure for data filtering and metric computation.
  2. Accurate Road User Simulation and integration with Naturalistic Driving – develop a multi-driver Driving Simulator and Multi-Agent Simulator (MAS) providing realistic interactions by a Distributed Artificial Intelligence (DAI) system able to incorporate and reproduce the Actor Model behaviour.
  3. Social Impact – Use of the aforementioned tools for the creation of effective interventions and identification of behaviour and motivations for unsafe driving and ADCs.

The project partners:

Instituto Tecnológico de Castilla y León, Institut Francais des Sciences et Technologies des Transports, De L’Amenagement et des Reseaux, Brainsigns SRLEuropean Driving Schools AssociationMälardalen University, Associazione Italiana Professionisti Sicurezza StradaleHök Instrument ABPrometeo Innovations, Progres 123 s. r. o., Twente Medical Systems International B.V.Coventry UniversityUniversity of PortoLink Innova EngineeringDelphi;  Estados Unidos con la Universidad de Iowa e Israel con IBM.