Cognition · Software component
Uniform-Cost Search Planner
Software componentCognitionCognition & Memoryarc:UniformCostSearchPlanner
A graph search planner that expands nodes by accumulated cost (or depth) without heuristic guidance, yielding shortest paths from the start to all reachable nodes.
Responsibility. Computes shortest paths from one start state to every reachable destination in a single search.
Also known as: Dijkstra planner, Breadth-first search planner, Uninformed search planner
Variant of Graph Search Planner abstract
When to choose. Choose when many destinations share one origin (20+), when the graph is tiny (~50 nodes), when repeated queries on a static graph amortise preprocessing, or when no meaningful goal-distance heuristic exists.
Relationships
alternative to variability
Design guidance
- SHOULD be preferred over repeated A* queries when paths to many destinations are needed from one origin.
- MAY be preferred for very small graphs or where no informative heuristic exists.
Quantitative guidance
As stated by the sources; verify before use.
- 20 destinations: one Dijkstra run ~15,000 nodes vs 20 A* runs x 200 nodes; separate A* wins for 3-4 destinations, Dijkstra for 20+ (Ch5.6).
Classification
- Patterns
- Dijkstra's algorithmBreadth-first searchAll-pairs shortest-path preprocessing
- Quality attributes
- Maintainability (ISO/IEC 25010)Functional suitability: correctness and validity (ISO/IEC 25010 | NIST AI RMF: valid)
Sources
- Ch5.6: T. Nguyen, "A* Search and Replaning," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.6. ISBN: 9798244538229.