Automated Guard Deployment and Squad Coordination: Enhancing Game Level Design
Bibliographic record
Abstract
Enemy guards play a pivotal role in video games, particularly in stealth and action genres, where their behavior directly impacts the player experience. Designing guards presents several challenges, including ensuring realistic behavior, balancing coverage and blind spots, optimizing patrol paths, and enabling dynamic responses. Moreover, squad coordination (teams of guards working together) adds further complexity by requiring strategic collaboration. Manually designing guard placements and patrol paths is labor-intensive and requires extensive playtesting. While procedural generation has been applied in level design, few automated tools specifically address guard placement and squad coordination. Existing approaches often struggle to balance feasibility, challenge, and adaptability. This paper presents two key contributions to assist level designers. First, we introduce an algorithmic model for automated guard placement and patrol paths. Using the Fisk algorithm (originally proposed to solve the art gallery problem), the Bowyer-Watson algorithm for path triangulation, and A* for optimal route computation, we provide a plugin for Unreal Engine 5 that allows level designers to automate guard placement and adjust game difficulty. Second, we propose a Squad Manager architecture to automate squad-based enemy behavior, ensuring strategic coordination among guards. Our approach has been validated through simulations and case studies, demonstrating its effectiveness in optimizing level design while maintaining engaging gameplay. By integrating our tools, designers can focus on high-level strategies while benefiting from automation. Both of our contributions are freely available online and can be used directly in Unreal Engine 5.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".