Future directions in agent programming

Agent programming is a subfield of Artificial Intelligence concerned with the development of intelligent autonomous systems that combine multiple capabilities, e.g., sensing, deliberation, problem-solving and action, in a single system. There has been considerable progress in both the theory and practice of agent programming since Georgeff & Rao’s seminal work on the Belief-Desire-Intention paradigm. However, despite increasing interest in the development of autonomous systems, applications of agent programming are currently confined to a small number of niche areas, and adoption of agent programming languages (APLs) in mainstream software development remains limited. In this paper, I argue that increased adoption of agent programming is contingent on being able to solve a larger class of AI problems with significantly less developer effort than is currently the case, and briefly sketch one possible approach to expanding the set of AI problems that can be addressed by APLs. Critically, the approach I propose requires minimal developer effort and expertise, and relies instead on expanding the basic capabilities of the language