University POLITEHNICA of Bucharest (ROMANIA)
About this paper:
Appears in: EDULEARN17 Proceedings
Publication year: 2017
Pages: 1776-1781
ISBN: 978-84-697-3777-4
ISSN: 2340-1117
doi: 10.21125/edulearn.2017.1378
Conference name: 9th International Conference on Education and New Learning Technologies
Dates: 3-5 July, 2017
Location: Barcelona, Spain
Many adults find that they are less active and mobile as they get older, fact that impedes them from performing daily activities or living independently. Regular exercise can improve mobility in many old adults. For instance, stretching is an important part of elderly and seniors flexibility that can help offset the effects of normal decline in the flexibility of the joints and help remain active and independent. Stretching is useful for reducing pain, improving posture and ameliorating arthritis. Competition and teamwork can inspire elderly people to work harder and to keep up their commitment to perform physical exercises, especially if it is done in the comfort of their homes.
This paper presents a new concept in the context of serious games for physical activity stimulation, offering the user the possibility of playing the game via a network. Taking into account the fact that the system based on a trainer character and a player character animated by the Microsoft Kinect device was successful, passing efficiency tests, in this version of the game we have an extended architecture.

The previous game analyses the user’s movements and compares them to those of a trainer character, taking into account performance and efficiency parameters. The exercises are automatically selected according to the user’s profile, medical condition and acquired progress and consequently the application generates a series of reports and statistics within a personal user account. The system’s architecture is composed of six modules: motion capture, movement analysis and feedback, activity control, animation control, user interface and progress management. The scene has a trainer character who presents the physical exercises. It can be animated by medical staff or the user can choose the automated mode in which the system executes the exercises with the speed corresponding to the user’s physical condition.

As teamwork, in most cases, improves the overall efficiency of an activity, the current version of the game allows introducing in the scene a maximum number of four users who can perform the exercises simultaneously and communicate via a network. There are training sessions at the end of which each user receives a detailed personal report with individual performance. Moreover, in order to stimulate teamwork, there is a general score assigned according to the overall team performance. Anytime the users can communicate and visualize their personal movements and the movements of their peers. The participants will be immersed in the virtual environment in order to create the illusion of a real gym. The possibility of real-time user interaction during the exercises induces an increase of performance in the context of teamwork support, an important psychological element. Furthermore, together with the rewarding elements, it contributes to maintaining the players in the training virtual environment for a longer period of time.

The main architecture consists of a desktop client application developed using the Unity game engine which will gather data from the Microsoft Kinect V2 sensor to animate the user character. The physical activities together with the users profiles are stored into a Microsoft Azure database. Moreover, the network capability is ensured by a C# server hosted by a virtual machine in Microsoft Azure Cloud.