Skip to main content

Sensors and system Architecture for VulnerablE road Users protection


Yearly more than 200 000 pedestrians are injured and 9 000 killed in the EU in traffic. Pedestrian accidents represent the second largest source of traffic-related injuries and fatalities. Initiatives to improve safety of unprotected road users (UPR) have been started both on national and European levels. SAVE-U is one of them.

The main objective of SAVE-U is to develop an integrated system for the active protection of UPR such as pedestrians and cyclists. The project deals with 3 different stages:
- Detection of the UPR at sufficient distance (up to 30 meters in urban environment) covering the largest range of scenarios;
- Definition and implementation of driver warning and vehicle control strategies in order to avoid a crash;
- Definition of strategies for active protection of UPR in case the crash cannot be avoided.

During the first part of the project, the novel sensor platform will be optimised for the detection of unprotected road users by means of low level and high level sensor fusion in order to reliably operate in most of the scenarios and to fulfil the strong requirements of an active safety system. The second part of the project deals with the development and implementation of the algorithms for data fusion, the integration of the entire sensor platform, the equipment of experimental cars as well as with the evaluation of the developments including the integrated system approach.

Work description:
Based on the analysis of relevant situations with unprotected road users (UPR) in urban environment (a database will be recorded in various European cities), the analysis of the reflectivity of the dressed human body for the sensing technologies considered here as well as on the basis of the advanced concepts for actuators to protect passengers and the know-how in terms of collision avoidance functions, SAVE-U will define suitable protection system concepts.

Accordingly, a novel SAVE-U sensor platform will be realised that will operate three different technologies and will fuse data for an optimised UPR detection:
- Several 24 GHz radar sensors will be used in parallel;
- Multi-sensor processing algorithms will be developed in order to operate the individual single beam sensors in the radar network;
- Merging the information from un-cooled IR and video cameras will offer the possibility for reliable and precise detection in all weather conditions;
- Dedicated image segmentation algorithms and the structure of a new IR camera specially designed for UPR detection will be developed;
- To achieve a real time operation, an embedded image processing platform (EIP) will be prototyped;
- Multi-cue detection algorithms for the classification of vulnerable traffic participants will be studied, e.g. combining stereo, motion, shape and texture in one system;
- Closed loop tracking strategies will be implemented in order to improve detection performance by means of prediction;
- SAVE-U will utilise sensor fusion both at low and at high level;
- High level data fusion alone is not sufficient to provide the required quality and reliability of the target data;
- SAVE-U will introduce a novel concept of low level data fusion;
- Sensor raw data will be exchanged between the image processing part and the radar processing part of the sensor platform.
At the end of the project, two experimental cars will be equipped with prototypes of the innovative sensor platform and with the novel driver warning and vehicle control strategies.

Performances evaluation will demonstrate the effective impact of the overall system on UPR safety in most of the traffic scenarios. Dissemination of results will take place mainly by means of a Web site and relevant congresses.

M1: Assessment of the realizability of the individual sensor requirements;
M2: Radar sensor, radar network and IR/vision sensor performance in accordance with specs;
M3: SAVE-U sensor platform validation: availability of first integrated sensor platform, performance of the entire platform in accordance with requirements, advantage from low and high level data fusion;
M4: Experimental cars with different warning/control strategies for the protection of UPR;
M5: System validation for protecting vulnerable traffic participant according to dedicated test methodology.

Funding Scheme

CSC - Cost-sharing contracts


2 Rue Hennape
92000 Nanterre

Participants (5)

31-33 Rue De La Federation
75752 Paris Cedex 15
Epplestrasse 225
70567 Stuttgart
United Kingdom
Watling Street
CV10 0TU Nuneaton Warwickshire
Wittelsbacherplatz 2
80333 Muenchen
Berliner Ring 2
38436 Wolfsburg