Efficient Octree-based 3 Dimensional Pathfinding
Bibliographic record
Abstract
Three-dimensional virtual environments are commonplace in modern games.Nonplayer character (NPC) character movement planning, however, is still largely 2D, with off-plane or vertical movement modeled through limited, and frequently custom connections between 2D surfaces.Naive extension of 2D pathfinding methods to 3D by building a full 3D grid (voxelization) is possible but easily becomes too expensive for non-trivial 3D spaces.In this thesis we develop an efficient solution to 3D pathfinding by building a reduced, hierarchical grid representation within which we can extend traditional 2D navigation mesh (navmesh) pathing.Our design begins with an octree representation, which already provides more flexibility than a more traditional voxelization.By merging adjacent cells while preserving their convexity, we obtain a coarser representation that greatly reduces path computation costs.We then build a navigation graph from this octree within which we can search for paths using the popular A* search algorithm.To increase the quality of the paths we obtain we implemented two forms of path refinement: a visibility-based path pruning heuristic, and a novel implementation of the classic "funnel" algorithm that computes minimal homotopic paths, in our case extended to our 3D environment.We further extend our work to handle dynamic environments with local and efficient updates to the octree and the movement graph.Experiments on a variety of scenarios show that our approach remains fast and efficient even for very large 3D maps and could be used for real-time pathfinding in video i games.We implemented our work in Unity, one of the most popular game engines, as an effort to make pathfinding in 3D environments accessible to game developers.First and foremost I want to thank my supervisor, Clark Verbrugge, for his invaluable help and support throughout these last two years.This thesis would not exist without him.I also thank Thomas Nobes for providing his results and code on the 3D JPS algorithm and his openness to discuss his work.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.000 | 0.001 |
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 teacher head, 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".