在生成遊戲內容的相關領域中，許多文獻提出使用程序化內容生成（Procedural Content Generation）的方法，其優點為降低開發成本並增添遊戲內容的隨機性和多樣性，而大部分的方法皆在生成結果前訂定一些由數學公式或其他參數組成的限制條件，使非相關背景的關卡設計師需要冒著誤解限制條件的風險及控制它們的迷茫。

本論文提出一套混合主導設計（Mixed-Initiative Design）系統，並參考動作遊戲中的遊玩特徵（Gameplay Patterns）歸納出關卡設計師能理解的抽象概念－基本戰術類型，借此有效輔助關卡設計師設計欲生成的戰術內容。我們將生成戰術內容的過程分為兩階段－生成空間（Space）與生成遊戲物件（Game Object）結構，而兩種結構的生成方法皆採用基因演算法（Genetic Algorithms）。

根據本論文的實驗結果，關卡設計師憑藉我們提出的抽象概念有效地理解與控制欲生成的戰術內容結果。不僅如此，在生成戰術內容的過程分為兩段式處理亦對生成結果具有控制性，使生成的戰術內容結果更符合關卡設計師欲生成的目標戰術類型。另外，借助混合主導設計系統中電腦演算法的影響，使生成的戰術內容結果在符合關卡設計師的控制下有不一樣的細節內容，具備多樣性並增添關卡設計師的創造力。

In the field of generating game content, there are many research proposed the method of procedural content generation (PCG). The advantage of using PCG is that reduce the costs of designing level and increase the randomness and diversity of game content. Most of methods set the constraints that based on mathematical formulas and parameters before generating the results. This causes the level designers without relevant knowledge to have the risk of misunderstanding the constraints and have the confusion.

In this paper we present a mixed-initiative design system. We reference the gameplay patterns of action game for define the abstract concepts that the level designers can understand. We called those abstract concepts basic tactics and used them to help level designers design the game content that they want to generate. We divided the process of generating game content into two stages－generating Space and generating Game Object, and both method of those two stages is genetic algorithms.

Our results show that the level designers understand and effectively control the result of game content that they want to generate by using the basic tactics. Also, the two stages of generating game content are able to control the result of game content that they want to generate. With the effect of computer algorithms in the mixed-initiative design system, the detail in same basic tactics but different results are different. Therefore, the result of game content is diverse and making the level designer has more creativity.

[1] Baldwin, A., Dahlskog, S., Font, J. M., and Holmberg, J. Mixed-initiative procedural generation of dungeons using game design patterns. In Computational Intelligence and Games (CIG), 2017 IEEE Conference on (2017), IEEE.

[2] Baldwin, A., Dahlskog, S., Font, J. M., and Holmberg, J. Towards pattern-based mixed-initiative dungeon generation. In Proceedings of the 12th International Conference on the Foundations of Digital Games (2017), ACM.

[3] Dormans, J. Level design as model transformation: a strategy for automated content generation. In Proceedings of the 2nd International Workshop on Procedural Content Generation in Games (2011), ACM.

[4] Dormans, J. Engineering emergence: applied theory for game design. Universiteit van Amsterdam [Host], 2012.

[5] Dormans, J. Cyclic dungeon generation. https://www.youtube.com/watch?v=mA6PacEZX9M, 2016. Accessed: 2018-06-26.

[6] Dormans, J. A handcrafted feel: ‘unexplored’explores cyclic dungeon generation. http://ctrl500.com/tech/handcrafted-feel-dungeon-generationunexplored - explores - cyclic - dungeon - generation/, 2016. Accessed: 2018-06-26.

[7] FromSoftware. Dark souls, 2011.

[8] Harisa, A. B. Pacing-based procedural dungeon level generation - a study on a level design tool for satisfying game designer’s expectation, 2018.

[9] Hart, P. E., Nilsson, N. J., and Raphael, B. A formal basis for the heuristic determination of minimum cost paths. IEEE transactions on Systems Science and Cybernetics 4, 2 (1968), 100–107.

[10] Holland, J. H. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control and Artificial Intelligence. MIT Press, 1992.

[11] Karavolos, D., Bouwer, A., and Bidarra, R. Mixed-initiative design of game levels: Integrating mission and space into level generation. In FDG (2015).

[12] Kazemi, D. Spelunky generator lessons. http://tinysubversions.com/spelunkyGen/, 2013. Accessed: 2018-06-26.

[13] Kimbrough, S. O., Koehler, G. J., Lu, M., and Wood, D. H. On a feasible–infeasible two-population (fi-2pop) genetic algorithm for constrained optimization: Distance tracing and no free lunch. European Journal of Operational Research 190, 2 (2008), 310–327.

[14] Kora, P., and Yadlapalli, P. Crossover operators in genetic algorithms: A review. International Journal of Computer Applications 162, 10 (2017), 34–36.

[15] Lipowski, A., and Lipowska, D. Roulette-wheel selection via stochastic acceptance. Physica A: Statistical Mechanics and its Applications 391, 6 (2012), 2193–2196.

[16] Nintendo. The legend of zelda, 1991.

[17] Smelik, R. M., Tutenel, T., de Kraker, K. J., and Bidarra, R. A declarative approach to procedural modeling of virtual worlds. Computers & Graphics 35, 2 (2011), 352–363.

[18] Togelius, J., Yannakakis, G. N., Stanley, K. O., and Browne, C. Search-based procedural content generation: A taxonomy and survey. IEEE Transactions on Computational Intelligence and AI in Games 3, 3 (2011), 172–186.

[19] Van der Linden, R., Lopes, R., and Bidarra, R. Designing procedurally generated levels. In Proceedings of the the second workshop on Artificial Intelligence in the Game Design Process (2013).

[20] van der Linden, R., Lopes, R., and Bidarra, R. Procedural generation of dungeons. IEEE Transactions on Computational Intelligence and AI in Games 6, 1 (2014), 78–89.

[21] Wang, Z. H. Game design goal oriented approach for procedural content generation, 2017.