Benchmarking the influence of information-processing architectures on intelligent systems

Abstract

The design of the information-processing architecture used to develop an intelligent robot plays a significant role in the behaviour of the final system. In this paper we discuss the possibilities for benchmarking the influence of architecture designs on intelligent robots. We separate this problem into two sub-problems: benchmarking the architecture design and benchmarking the implementation of the design. For each of these sub-problems we list some design- and run-time properties that could be investigated. To further demonstrate these ideas we present some early efforts to benchmark the run-time properties of a previously developed architecture schema.