CNLS '89 Proceedings of the ninth annual international conference of the Center for Nonlinear Studies on Self-organizing, Collective, and Cooperative Phenomena in Natural and Artificial Computing Networks on Emergent computation
Anchoring Symbols to Sensor Data: Preliminary Report
Proceedings of the Seventeenth National Conference on Artificial Intelligence and Twelfth Conference on Innovative Applications of Artificial Intelligence
Symbol grounding and its implications for artificial intelligence
ACSC '03 Proceedings of the 26th Australasian computer science conference - Volume 16
Agile and Iterative Development: A Manager's Guide
Agile and Iterative Development: A Manager's Guide
Intelligence by design: principles of modularity and coordination for engineering complex adaptive agents
The behavior-oriented design of modular agent intelligence
NODe'02 Proceedings of the NODe 2002 agent-related conference on Agent technologies, infrastructures, tools, and applications for E-services
Proceedings of the 49th SIGMIS annual conference on Computer personnel research
Hi-index | 0.00 |
Our focus is the development of software for controlling autonomous robots. In particular, we are concerned with the problem of grounding representations - creating representations which endow the agent with a meaningful understanding of the problem at hand. For most tasks, the development of an appropriate representation (e.g. world models, ontologies, schemas for decision-making and action) is a laborious process in which the developer handcrafts the representation for the particular control problem. The quality of the resultant representation is too dependent upon the designer's experience. Such agents can lack the capability to cope with, or adapt to, circumstances that were not anticipated by the developer. Despite the large labor costs incurred through handcrafting representation, and the cost of redesign when unanticipated circumstances force changes to agent software, there exist few methods for guiding the developer through this process. We present a preliminary methodology for designing and grounding representations.