Intelligent Feedback for Simulation Based Training

This paper describes the architecture and the process of system designed to provide intelligent feedback in Virtual Reality-based training simulators. This system will give multimodal feedback to the student as a response to actions in real-time. We propose a two step process. The fist step is designed to select the most relevant messages about student's activity. The second step selects the feedback that will be communicated to the student. The process takes consideration some factors such as the number and nature of the committed mistakes, the Student's level, the intrusiveness each kind of feedback, the time elapsed since the last feedback and so on. The instructor can customize the behavior of feedback system by choosing the relevance of the used in the feedback selection process. In this way, the feedback system can be helpful for the instructor, as he/she can use a Suitable behavior in different types of training sessions (exams. further training, training for beginners and so on). Meanwhile, the instructor can focus on other instructional tasks.