{"resourceId":"uscg-crada-lookout-ai-target-detection-search-and-rescue-evaluation-20261026","versions":[{"version":"external-c2a0010109059c02df7ec9b44e691166c461f689069e811472f3f04f8ff50b5a","resource":{"id":"uscg-crada-lookout-ai-target-detection-search-and-rescue-evaluation-20261026","title":"U.S. Coast Guard proposes a CRADA with LOOKOUT to evaluate AI target detection on its sensor feeds for search and rescue, aiming for a measurable evaluation tool","organization":"U.S. Coast Guard Research and Development Center (DHS); LOOKOUT","sector":"Maritime search and rescue","geography":"United States (federal)","publishedAt":"2026-10-26","publicationDate":"2026-10-26","eventDate":"2026-10-26","sourceName":"Federal Register via GovInfo","sourceLabel":"Federal Register notice 2026-21723: CRADA for Super Intelligence situational awareness tools for SAR","sourceUrl":"https://www.govinfo.gov/content/pkg/FR-2026-10-26/html/2026-21723.htm","evidenceClass":"government-evaluation","outcomeClass":"emerging","topics":["governance-procurement","infrastructure","data-security"],"finding":"In a Federal Register notice of intent (docket USCG-2026-0374, comments due November 25, 2026), the Coast Guard R&D Center plans a cooperative research and development agreement with LOOKOUT to test AI situational-awareness tools that could automatically detect, classify and track targets in low light, fog, precipitation and rough seas. The assessment aims to benchmark performance, evaluate interoperability, propose data-handling practices, establish a reliable coverage-factor value for data-driven lateral range curves, and characterise system confidence to build a measurable evaluation tool. Tools must be commercially available, high readiness, portable, edge-capable and sensor-agnostic. The notice stresses CRADAs are not procurement contracts and the Coast Guard provides no funding; other participants may propose.","sledRelevance":"A federal SAR agency is defining how to measure AI detection before operational adoption; state and local SAR, fire and emergency management drone programs can borrow the coverage-factor and confidence-measurement approach.","evidence":"Primary government notice of a planned evaluation; no results yet. It describes objectives and requirements, not performance.","architectureImplications":"Edge-computing, self-contained tools working across video-feed-agnostic sensors, integrated with existing SAR workflows and data feeds.","governanceImplications":"Uses a CRADA technology assessment with public comment and a stated goal of measurable evaluation before integration, separating testing from procurement.","securityPrivacyImplications":"Data-handling best practices are an explicit objective; Coast Guard sensor feeds shared with a private partner need controls.","caveats":"Federal maritime context, not state or local; notice of intent only, with outcomes unknown. Published Oct 26, two days before this edition; included because it was not archived and comments remain open. Federalregister.gov page blocked automated access, so the GovInfo copy was read.","streamIds":["emergency-services"],"roles":{"sales":"For sales, the framing point is that the Coast Guard wants a measurable evaluation tool before it integrates AI detection into search and rescue. State and local SAR and drone teams face the same question. A bounded engagement is an evaluation design that measures detection confidence and coverage for an agency's own sensors. This notice cannot support claims that AI detection improves SAR outcomes; testing has not started.","engineering":"For pre-sales engineering, the requirements are edge-capable, self-contained tools that work across different video sensors. Prerequisites are access to representative sensor feeds and ground truth for targets. Constraints include low light, fog and sea state, and data-handling limits on shared feeds. Proof of value should produce detection-confidence and lateral-range or sweep-width style measures compared with human observers.","delivery":"For delivery, the work is a structured technology assessment: installation, test scenarios, data handling and a joint summary report. Ownership sits with the R&D or innovation unit, with operators involved. Acceptance should rest on documented detection performance under defined conditions. Risks include unrepresentative tests, data-sharing exposure and treating a research agreement as a procurement commitment."},"roleDetails":{"sales":{"customerProblem":"SAR teams need evidence that AI detection works on their sensors and conditions before relying on it.","stakeholders":["SAR program managers","Drone or aviation unit leads","Innovation or R&D office","Legal and data governance"],"discoveryQuestions":["How do you measure detection performance today?","Which sensors and conditions matter most?","Could a no-funding research agreement fit your rules?"],"valueHypothesis":"A measured evaluation could show where AI detection helps searchers before money is committed (interpretation).","potentialEngagement":"Design of a detection-performance evaluation using an agency's own sensor feeds and scenarios.","unsupportedClaims":["Improved SAR outcomes","LOOKOUT system performance","Readiness for operational use"]},"engineering":{"fit":"Fits SAR programs with video sensors on aircraft, vessels or drones.","architectureIntegration":"Edge AI processing on sensor-agnostic video feeds feeding existing SAR workflows.","prerequisites":["Representative sensor feeds","Ground-truth target data","Installation access"],"constraints":["Low light, fog, precipitation and sea state","Data-handling limits on shared feeds","Federal maritime context differs from local SAR"],"security":"Control partner access to sensor feeds and define retention in the agreement.","proofOfValue":["Detection probability by range and condition","Confidence calibration","False alarms per hour against human observers"]},"delivery":{"work":"Run a structured assessment with test scenarios and a joint summary report.","dependencies":["Signed research agreement","Data feeds","Partner technical staff"],"ownership":"Agency R&D or innovation unit with SAR operators.","skillsAdoption":"Operators learn to interpret AI detections and confidence levels.","governanceCheckpoints":["Public notice and comment","Data-handling plan approval","Results review before any procurement"],"acceptanceCriteria":["Proposed: published detection metrics by condition","Proposed: data-handling plan audited"],"risks":["Unrepresentative test conditions","Data exposure","Research agreement mistaken for adoption"]}},"retrievedAt":"2026-10-28T22:00:32-05:00","enrichedAt":"2026-10-28T22:04:35-05:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"AI is framed as augmenting SAR watchstanders' decision-making, not replacing searchers.","procurementImplications":"Shows test-before-buy via no-funding research agreements with public notice, an option for agencies wary of early procurement.","operatingModelImplications":"AI detection would augment existing multi-sensor SAR workflows, with performance data informing future adoption.","sourceVerification":{"responseId":"curl-frgov-2026-10-28","openedUrl":"https://www.govinfo.gov/content/pkg/FR-2026-10-26/html/2026-21723.htm","referenceExcerpt":"may allow the CG to automatically detect, classify, and track targets","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}