SafeChild: An Intelligent Virtual Reality Environment for Training Pedestrian Safety Skills

Yecheng Gu, Sergey Sosnovsky, Carsten Ullrich

In: Design for Teaching and Learning in a Networked World. European Conference on Technology Enhanced Learning (EC-TEL-15) 10th September 15-18 Toledo Spain Seiten 141-154 Springer 2015.


Training children safe behavior in traffic situations is both important and challenging. One of the problems is children’s limited perceptual-motor abil‐ ities and associated difficulties with important cognitive skills required to be safe pedestrians. Existing traffic education programs focus more on theoretical knowl‐ edge, while training practical skills in the real world is dangerous, expensive and hard to organize. This paper presents a promising alternative – an intelligent virtual reality training environment that allows children to practice their pedes‐ trian skills. It describes the interface and architecture of the system, as well as the skill model of the pedestrian safety domain. The results of the conducted pilot study show that children of the target age group rarely have problems with applying (and acquiring) “basic” pedestrian skills in the developed virtual envi‐ ronment. However, when applying and learning “advanced” skills, they require additional support.


Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence