Adapting Playgrounds for Children's Play using Ambient Playware

Abstract

This paper presents an approach on how to adapt playgrounds using artificial neural networks (ANN). The playground consists of small tiles each capable of outputting coloured light, sensing force applied on them and communicating with neighbouring tiles. It is shown that a child's behaviour on the playground can be classified within eight categories by an ANN inputting the child's perception of the playground and the corresponding action performed; thus, enabling the playground to adapt real-time according to the classification.The playground has been developed, implemented and tested on a group of children. An ANN was trained using a subgroup of the children whose behaviour corresponds to the eight categories. Validating the ANN against the rest of the children's behaviours it was found that 96% were correctly classified. The trained ANN has been utilized in adapting the playground according to several playing strategies and validated by means of a statistical analysis of physiological signals (e.g. the heart rates of the children).

Distributed Robotics Education

“…Feedback and multimodal feedback are introduced below. We have shown in more then one case [29,30] that using modular interactive tiles enabled us to detect user characteristics and adapt software execution to those. In further tests, we found that by capturing a user's conditional attitude and adapting software execution to this that, in some cases, we could eventually modify user behavior itself [30].…”

(Steering) Interactive Play Behavior

“…Digital and mixed reality games allow for interventions in the game state, game difficulty, or game behavior. Personalized adaptation of the interventions allows people with different skills to play together, increasing fairness and engagement in a fun and challenging experience [30,64].…”

From traditional to interactive playspaces: automatic analysis of player behavior in the interactive tag playground

“…Derakhshan et al presented an interac-tive playground that consisted of tiles that children could step on and interact with through force sensors and LEDs [70]. They used neural networks to learn and model different types of game styles, such as fast, slow, or continuous.…”

“…This can be achieved by developing models that reflect the player's behavior or preferences. Derakhshan et al used a platform that could recognize several player styles based on tempo, age and continuity of play [70]. The tempo of the game was adapted, along with other characteristics, to match what the player felt comfortable with to successfully promote physical activity.…”

“…This can be reduced by using multiple projectors to fill in the image. LEDs are other sources of light, and their limited size and low energy demands makes them suitable in building blocks or buttons to signal their presence and affordance [70,173]. Grids of LEDs have been developed for outdoor use and as interactive floors.…”