Community Research and Development Information Service - CORDIS

H2020

PROSPECT Report Summary

Project ID: 634149
Funded under: H2020-EU.3.4.

Periodic Reporting for period 1 - PROSPECT (PROactive Safety for PEdestrians and CyclisTs)

Reporting period: 2015-05-01 to 2016-10-31

Summary of the context and overall objectives of the project

Accidents involving Vulnerable Road Users are still a very significant issue for road safety. According to the World Health Organisation, pedestrian and cyclist deaths account for more than 25% of all road traffic deaths worldwide. Autonomous Emergency Braking Systems have the potential to improve safety for this group of VRUs.
PROSPECT aims to significantly improve the effectiveness of active VRU safety systems compared to those currently on the market by expanding the scope of scenarios addressed by the systems and improving the overall system performance.
PROSPECT focusses on active safety solutions, where vehicle-based sensors survey the vehicle surroundings and the system acts actively in case of a critical situation with a VRU.
Within the project all relevant VRU traffic scenarios are considered, with a special focus on urban environments, where the large majority of VRU accidents occur.
Knowledge and technologies to be developed in the project will be relevant for different vehicle categories but with main focus on passenger cars.
The findings within the PROSPECT project will contribute to the generation of state-of-the-art knowledge, advanced technologies, assessment methodologies and tools for advancing Advanced Driver Assistance Systems towards the protection of VRUs.
The introduction of a new generation safety system in the market will enhance VRU road safety in 2020-2025, contributing to the ‘Vision Zero’ objective of no fatalities or serious injuries in road traffic set out in the Transport White Paper. Furthermore, test methodologies and tools shall be considered for 2018 and 2020 Euro NCAP test programmes, supporting the European Commission goal of halving the road toll in the 2011–2020 timeframe.
To achieve the overall goal of VRU accident and injury reduction, the project follows an approach that integrates the following key objectives:
-OBJ1: Better understanding of relevant VRU accident scenarios.
-OBJ2 Improved VRU sensing and situational analysis.
-OBJ3: Advanced HMI and vehicle control strategies.
-OBJ4: Four vehicle demonstrators, one transportable mobile driving simulator and realistic VRU dummy specimen provided by leading EU industry partners for validation and effectiveness analysis.
-OBJ5: Testing in realistic traffic scenarios, system performance assessment of the novel VRU active safety functions and user acceptance study.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

Within the first period of the project, macro statistical and in-depth accident studies involving VRUs have been performed in Europe, focused mainly in pedestrians and cyclists. The most relevant accident scenarios have been obtained and clustered in use cases or target scenarios addressed by the project, which is the basis for the system specification.
Naturalistic urban observations have been carried out in order to provide additional information with respect to the defined Use Cases. Typical behavioral schemes of traffic participants (vehicle drivers, bicyclists or pedestrians) have been studied.
Focus Groups to study the characteristics of road users and HMI design have been performed in order to derive the user needs and provide additional characteristics of VRU through traffic.

In order to address the Use Cases defined in the project, the development work of the first sensor processing components have started. These intend to support a larger coverage of accident scenarios by means of an extended sensor field of view (e.g. frontal stereo vision coverage increased to about 90°, radar coverage increased up to 270° covering vehicle front and one side), high-resolution / sensitivity microwave radar sensors, and micro-Doppler effect.
From the video processing side, algorithms are under modelling in order to better detect VRUs.
VRU recognition performance will be improved by applying advanced Machine Learning techniques which model VRU appearance and motion.

Apart from dealing with VRU detection and localisation, work for the improvement of sensor and situational analysis extract has obtained cues that are indicative of VRU motion intent, enabling more sophisticated VRU modelling and situation analysis. Motion analysis and movement capabilities of a pedestrian are in a first step analysed and are currently extended to typical cyclist movement patterns. Based on recording sessions to gain insight in cyclist behaviour for advanced risk estimation. extended theoretical concepts for intention analysis of pedestrians and cyclist using Dynamic Bayesian Networks have been developed. The algorithms will greatly improve the path prediction.

Based on VRU detected and captured information over time, first models of determination of critical situation and collision risk estimation have been obtained.
Based on HMI studies within the project, a basic HMI is currently under study; conceptual plans have also been designed for regarding vehicle control strategies for braking, steering or combined braking / steering interventions for implementation into the demo vehicle.

According to the system specification, the integrated concept of the overall sensor set-up, HMI and actuators including all interfaces is being defined for the vehicle demonstrators.

Advanced articulated dummies are already under study to obtain higher degrees of freedom (head rotation, torso angle, pedaling, side leaning…) and improved movement patterns.

The driving simulator has already integrated the required characteristics in order to be able to implement a first subset of the PROSPECT use cases.

Finally, a collection of test cases that are representative for all accident scenarios has been defined and specified. These cases take into account relevant parameters and values in order to test and validate the new systems developed in the project. The first complete test protocol as a proposal for consumer has been obtained.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

- PROSPECT aims to effectively reduce car-to-pedestrian and car-to-cyclist accidents, which result in serious or fatal VRU injuries. This reduction is based on a higher scope of addressed scenarios and a better performance of the system developed in the project.
- Data from normal driving, incidents and accidents are the basis for road safety measures. The PROSPECT project will summarise the state-of-art and extend the knowledge about VRU accidents. Vehicle safety system development is dependent on the latest information from accident research and an up-to-date accident analysis focusing on VRUs will increase the relevance and effectiveness of the systems developed in PROSPECT as well as those in future projects.
- The different technologies and testing tools developed within PROSPECT will allow:
- To address Euro NCAP 2020 scenarios and fulfil the increasingly strict requirements.
- The entrance of new vehicles in the EU market.
- High safety standards that deliver a true benefit to consumers in Europe and to society as a whole.
- Information about the findings within PROSPECT will be forwarded as relevant input for Euro NCAP decision makers in parallel to the PROSPECT work.
- The PROSPECT partners are also interested in the participation in ISO Working Groups that aim at standardising VRU collision mitigation systems as well as test equipment and tools for active safety systems.
All stakeholders agree that this standardisation is needed for the future development of ADAS.

Related information

Record Number: 198165 / Last updated on: 2017-05-17
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top