Use of Ecological Gestures in Soccer Games Running on Mobile Devices

Mobile soccer gaming often suffers from the limitations of traditional touchscreen controls, which frequently lack the tactile feedback and immersion required for authentic sports simulation. To address this, wearable insole sensors facilitate ecological interactions, allowing players to mirror real-world physical movements such as kicking and passing within a virtual environment. This approach bypasses the latency and environmental constraints inherent in computer vision-based systems, providing a more intuitive and responsive interface for mobile hardware.

The core of this interaction model relies on a lightweight, feature-based gesture recognition algorithm designed specifically for the computational constraints of mobile devices. By prioritizing real-time performance, the system achieves an impressive processing speed of 1.5kHz, significantly outperforming traditional Dynamic Time Warping and FastDTW methods. While the algorithm currently demonstrates a 57.6% recognition rate for intended soccer gestures—a figure lower than the 89.8% accuracy achieved by more resource-intensive DTW models—it exhibits exceptional reliability in filtering out noise. Specifically, the system maintains a 100% success rate in rejecting unlisted or extraneous movements, ensuring that only intentional physical actions trigger in-game responses.

This research establishes a viable framework for integrating physical activity into mobile gaming, emphasizing the necessity of balancing recognition precision with hardware efficiency. Future development will focus on refining the algorithm’s accuracy through the application of genetic algorithms and the expansion of training datasets. By bridging the gap between physical exertion and digital input, this technology offers a scalable path toward more immersive and physically engaging mobile sports experiences.