Anassa breaks down spatial ai framework into executable architecture

ANASSA: From High-Level Framework to Low-Level Architecture for Spatial Agentic AI

Artificial IntelligenceMultiagent SystemsSoftware Engineering

Summary

Spatial AI systems need to do more than just think on their own—they must also manage tasks like checking locations, handling errors, and keeping humans in control. The authors had developed a big-picture plan for such systems but hadn't explained how to build them exactly. In this paper, they turned that plan into a detailed and practical design with clear roles and processes to make these systems work smoothly and safely. This design works independently of specific AI tools or software, so it can be adapted widely.

What this means in practice

  • For geospatial system architects: Design and implement complex spatial AI workflows with clear roles, validation, and governance using the detailed architecture derived from the ANASSA framework.
  • For enterprise ai integration teams: Build spatial AI solutions that need robust failure handling and human oversight without tying to specific AI models or platforms.

Authors

Constantinos Papantoniou, Brian Hilton

Abstract

Agentic AI systems operating over geospatial data require more than autonomous reasoning and tool use; they must also support explicit orchestration, spatial validation, governance, provenance, failure handling, and human decision authority. Prior work introduced ANASSA, an agentic AI orchestration framework for spatial intelligence that established high-level components and responsibility boundaries for governed agentic execution (arXiv:2609.14824). However, the framework did not specify how those high-level responsibilities should be systematically decomposed into an implementation-facing low-level architecture. This paper addresses that gap by applying Design Science Research (DSR) to refine the ANASSA framework from high-level architectural (HLA) responsibilities into a low-level architecture (LLA). The resulting design preserves the responsibility boundaries of the foundational framework while translating them into executable architectural structures, including 47 logical roles, versioned task graphs, capability-based agent selection, typed contracts, explicit runtime states, spatial validation mechanisms, governance controls, failure and recovery paths, and provenance records. The architecture is demonstrated through an end-to-end execution trace and analytically evaluated for requirements traceability, internal consistency, responsibility preservation, and technology independence. The resulting high-level-to-low-level design provides an implementation-facing bridge between the ANASSA framework and future system implementations, defining how governed spatial agentic workflows can be made inspectable, reconstructable, and testable without binding the architecture to a specific AI model, geospatial platform, communication protocol, or deployment environment.