Virtual Reality Based Independent Travel Training System for Children with Intellectual Disability

Travel independently is one of the core skills required for children with intellectual disability to lead an independent life. Virtual reality based travel training system with natural interaction could help them to practice their travel skill in a safe environment without supervision. Developing real-time travel training system is a challenging work due to inherent complexity of 3D traffic simulation. This paper focuses on developing a simplified travel training system but keeping all the significant features of simulated traffic entities related to travel training. Multi-agent system (MAS) with centralized data manager is introduced to simplify the interaction among all traffic entities and virtual traffic environment. The travel training system concentrates on more realistic interaction among virtual avatar, vehicles and pedestrians to help the trainee get more immersive travel experience. Experiment results show the different behaviors of simulated pedestrians and vehicles in vehicle-pedestrian interaction.