Exercising a Native Intelligence Metric on an Autonomous On-Road Driving System

Published

October 1, 2003

Author(s)

John A. Horst

Abstract

Intelligent artificial systems can be represented by strings of bits and the intelligence of such systems is well quantified by the amount of specified complexity inherent in the representation, provided we have tools to measure it. Some may generally agree with this claim, but argue that it is simply intractable to successfully and accurately measure the specified complexity of any system, no matter how it was represented. This important and substantive criticism may perhaps best be answered by a concise example problem. We have chosen autonomous on-road driving as the example problem for this research, a problem solved by known systems (one of many tasks performed by humans) that are known to be both complex and specified. We will begin with a concise statement of the scope of the problem and a summary description of our own system representation approach. The meat of this work is begin to apply a previously published Native Intelligence Metric (NIM) to measure the specification inherent in that representation and report these preliminary measurements.

Proceedings Title

Proceedings of the 2003 Performance Metrics for Intelligent Systems Workshop |||

Conference Dates

September 1, 2003

Conference Title

Performance Metrics for Intelligent Systems

Pub Type

Conferences

Keywords

4D/RCS, autonomous control, autonomous on-road driving, finite state automata, hierarchical control, intelligent systems, native intelligence metric, RCS, system intelligence

Citation

Horst, J. (2003), Exercising a Native Intelligence Metric on an Autonomous On-Road Driving System, Proceedings of the 2003 Performance Metrics for Intelligent Systems Workshop ||| (Accessed November 8, 2024)