Subsumption Architecture
Updated 2026-08-10
INTRODUCTION
English translation pending.
CORE DEFINITION
A robot control architecture proposed by Rodney Brooks in which behavior is organized into parallel layers. Lower layers handle basic reactions such as obstacle avoidance, while higher layers pursue more complex goals and can suppress or subsume lower-layer output without deleting it. The key qualification is that the architecture abandons any centralized world model: coordination comes from layer interaction rather than from a central planner. This makes the system more robust, since the lower reactive layers continue to function if a higher layer fails.
SCAFFOLDING EFFECT
Reduce cognitive load
- Layer decomposition: split the capability into parallel behaviors ordered from basic to complex. - Incremental build: make each layer work on its own before adding the next one above it. - Suppression design: coordinate layers by allowing higher ones to override lower output temporarily.
Anchor fast decisions
A centralized controller must build and maintain a model of the world before it can act, and errors in that model corrupt every decision that depends on it. Layered reactive behaviors act directly on sensor input, so each layer remains valid regardless of the others. Higher layers achieve sophistication by suppressing rather than replacing lower ones, which preserves the reliable basics and confines the effect of any single layer's failure.
MINIMUM ACTION
In progress 0/1Practice this model in one real situation:
account_treeGenealogyexpand_more
menu_bookReferencesexpand_more
Source support: Explicit
- en.wikipedia.orghttps://en.wikipedia.org/wiki/Subsumption_architectureverified
PRIVATE NOTES · Only visible to you
SAVED Q&A
ENTRY Q&A · Private saving available
Ask with a clear boundary
thinkingmodels answers from published entry context only.
Your question is sent to thinkingmodels. The answer uses public entry context only.
RELATED MODELS