Evolving MoonBoard Bouldering Routes

The challenge of utilizing artificial intelligence to generate indoor rock climbing routes with a specific grade remains an interesting and unsolved problem due to its complexity and subjectivity. We use MAP-Elites, an evolutionary, quality-diversity algorithm, to produce a set of disjoint climbing routes that sufficiently challenge a climber without exceeding their physical and technical limitations. Following this, we attempt human judgement of the routes through climbing as well as visual observation.