SentinelAI: A Multi-Agent Framework for Structuring and Linking NG9-1-1 Emergency Incident Data

Kliment Ho; Ilya Zaslavsky

cs.AIMar 25, 2026

SentinelAI: A Multi-Agent Framework for Structuring and Linking NG9-1-1 Emergency Incident Data

Authors: Kliment Ho, Ilya Zaslavsky

Organizations: San Diego Supercomputer Center UC San Diego La Jolla, CA

Abstract

Emergency response systems generate data from many agencies and systems. In practice, correlating and updating this information across sources in a way that aligns with Next Generation 9-1-1 data standards remains challenging. Ideally, this data should be treated as a continuous stream of operational updates, where new facts are integrated immediately to provide a timely and unified view of an evolving incident. This paper presents SentinelAI, a data integration and standardization framework for transforming emergency communications into standardized, machine-readable datasets that support integration, composite incident construction, and cross-source reasoning. SentinelAI implements a scalable processing pipeline composed of specialized agents. The EIDO Agent ingests raw communications and produces NENA-compliant Emergency Incident Data Object JSON.

Figures & tables

Figure 1: SentinelAI reference architecture showing the three-agent workflow. The EIDO Agent structures incoming reports into EIDO-JSON, the IDX Agent links related EIDOs into composite incidents, and the Geocoding Agent enriches location information. Agents interact through well-defined interfaces and support analytical and integration services.

Legacy CAD Field	EIDO-JSON Target	EIDO Agent Structuring Task
Incident Type	incidentComponent. incidentTypeCommonRegistryText	Normalize proprietary codes (for example, ‘211A’) to IANA registry value (‘ROBBERY-ARMED’).
Initial Problem Description	notesComponent. notesActionComments	Summarize and extract core event narrative from multi-paragraph call-taker notes.
Response Hour, Day, Month	callComponent. callStartTimestamp	Synthesize disparate temporal fields into compliant ISO 8601 timestamp.
Priority Level	incidentComponent. incidentCommonPriorityNumber	Map agency-specific codes (‘P1’, ‘Code 3’) to standardized numeric scale (1-5).
Sector / Beat	locationComponent. locationDescriptionText	Contextualize jurisdictional areas, appending them to location description.
Call Disposition	incidentComponent. incidentDispositionText	Map disposition codes (‘ADV’) to descriptive text (‘Advised’).

Table 1: Representative transformations performed by the EIDO Agent

Figure 2: Decision logic employed by the IDX Agent to determine if an incoming report represents a new incident or an update to an existing one.

Figure 3: Integration with FME showing EIDOReader and EIDOWriter components facilitating bidirectional data exchange.

Figure 4: Visual transformation of the NWS Flood Warning text (left) into a structured EIDO-JSON document (right) by the EIDO Agent, highlighting the extracted event type and polygon location data.

Figure 5: The SentinelAI Dashboard visualization showing the IDX Agent’s linking of the NWS Warning and the News Report into a single composite incident timeline.

Disaster Response

Disaster response uses data, models, and planning tools to assess hazards and coordinate evacuation, relief, and recovery.

See all 59 papers

Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations
May 12, 2026Junjue Wang, Weihao Xuan, Heli Qi +7LLM Agent Evaluation Geospatial Reasoning
The University of Tokyo · RIKEN AIP · Waseda University +1
Interpretable Crisis Behavior Analysis Using Mobility and Social Media Data
Jun 8, 2026Muhammad Hamza Arshad Majeed, Sidahmed Benabderrahmane, Talal RahwanSocial Media Analysis Association Rule Mining
New York University (NYUAD), Division of Science, Computer Science Department
Verifiable Disaster Storylines and Causal Knowledge Graphs: A Citation-Grounded Pipeline from Heterogeneous Humanitarian Sources
Sep 1, 2026Ivan Decostanzi, Michele Ronco, Sergio Consoli +8KG Construction Grounded Text Generation
ISI Foundation · Turin, Italy · European Commission, Joint Research Centre (JRC) +4

May 12, 2026J/K move · Enter open · S save

Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations

Junjue Wang, Weihao Xuan, Heli Qi, Pengyu Dai+6

The University of Tokyo · RIKEN AIP · Waseda University+1

LLM Agent Evaluation Geospatial Reasoning AI Agent Benchmarks Disaster Response Tool-Using Agents Multimodal Agents+1

Open paper AskPDF arXiv

Operational disaster response goes beyond damage assessment, requiring responders to integrate multi-sensor signals, reason over road networks, populations and key facilities, plan evacuations, and produce actionable reports. However, prior work largely isolates remote-sensing perception or evaluates generic tool use, leaving the end-to-end workflows of emergency operations underexplored. In this paper, we introduce Disaster Operational Response Agent benchmark (DORA), the first agentic benchmark for end-to-end disaster response: 515 expert-authored tasks across 45 real-world disaster events spanning 10 types, paired with expert-verified, replayable gold trajectories totaling 3,500 tool-call steps. Tasks span five dimensions that cover the operational disaster-response pipeline: disaster perception, spatial relational analysis, rescue and evacuation planning, temporal evolution reasoning, and multi-modal report synthesis. Agents compose calls from a 108-tool MCP library over heterogeneous geospatial data: optical, SAR, and multi-spectral imagery across single-, bi-, and multi-temporal sequences (0.015-10m GSD), complemented by elevation and social vector layers. We comprehensively evaluate 13 frontier LLMs on our benchmark, revealing three persistent challenges: 1) disaster-domain grounding exposes unique failure modes (damage-semantic grounding, sensor-modality mismatch, and disaster-pipeline composition); 2) agents are doubly bottlenecked by tool selection and argument grounding, where gold tool-order hints improve accuracy by only 1.08-4.40%, and alternative scaffolds yield at most a 3.24% gain; 3) compositional fragility scales with trajectory length, the agent-to-gold gap widening from 7% to 56% on long pipelines. DORA establishes a rigorous testbed for operationally reliable disaster-response agents. Code and data are available at https://github.com/Junjue-Wang/DORA.

Abstract

Disaster Response

Explore similar work

Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations

Interpretable Crisis Behavior Analysis Using Mobility and Social Media Data

Verifiable Disaster Storylines and Causal Knowledge Graphs: A Citation-Grounded Pipeline from Heterogeneous Humanitarian Sources