Testing Super Mario Using a Behavior Model Autonomously (Part 1)
Summary
Autonomous testing of Super Mario uses a mutation based input generator and a population based search to explore the game's state space deterministically. The article maps this approach to a genetic algorithm, discusses inputs, path selection, backtracking, pruning, and performance instrumentation, and presents results across four levels while noting limitations in correctness validation and the plan to integrate a behavior model in Part 2.