Scalability of the Subsumption Architecture

Brooks considers the problem of scalability in his article "Intelligence without Representation". Brooks suggests there that there may be limits on both the number of layers the subsumption architecture may support as well as the complexity of the resulting behavior. It has not yet been demonstrated that these limitations do, in fact, exist. Mataric points out that the distributed nature of the architecture allows it to scale well. Brooks adds that the addition of hormones increases the size of the system linearly but there is a limit on the intersections between hormones, effectively limiting that system's overall size.

Return to the Issues Index for this Architecture.

Return to the top of this architecture.

Go to a Discussion of this issue for multiple architectures.

Current Location: Subsumption-Issues-Scalability

Go to NEXT page.