The upcoming moon landing, scheduled for 2031, presents several challenges, one of which is automatic hazard detection during landing. While this task is relatively straightforward for humans, replicating it in the extreme safety-critical conditions of a moon landing is far from trivial for automated systems. Modern AI technologies often rely on large quantities of annotated datasets. While easy enough to create with the help of citizen scientists, the annotation process is labor-intensive, repetitive and tedious leading to early drop off rates by the volunteers involved.

While previous works did show promising results by utilizing gamification to boost motivation we hypothesize that integrating annotation tasks into a fully realized game, rather than relying on conventional gamification techniques, can significantly enhance user motivation and retention.

Over the past year we developed and launched a free game on Steam and Android that integrates the marking task into its game loop. This allowed us to compare the retention rate and annotation accuracy with traditional annotation tools.

Furthermore, we conducted a comparative analysis to evaluate the effectiveness of this game-based approach. This includes examining marking performance across different platforms (Windows and Android) and testing an optimized marking workflow tailored for mobile devices.

The presentation will feature an overview of the development process and our scientific results as well as some anecdotes about publishing an app on commercial platforms like Steam and the Google Play Store as an academic institution.