Data
Trace sensor, location, time, weather, cultivar and annotation.
System anatomy · 06
Perception and agronomy
Separate sensing of crops, weeds, fruit, disease and soil from the agronomic rule that turns an estimate into action.
Evidence statusField evidence framework · crop, season, site and operating-envelope validation requiredLast reviewed: 16 August 2026
06
FUURAA thesis
The same detection error can have very different cost at different growth stages and treatment types.
Engineering map
Turn the title into engineering objects that can be observed, measured and reviewed.
Each layer states the boundary to establish and the evidence needed for the next decision.
Uncertainty
Calibrate confidence and define abstention or review.
Agronomy
Link thresholds to timing, dose, crop tolerance and follow-up.
Verification questions
Write the questions first, then decide whether a demo, test, pilot or operating record can answer them.
Each question needs an object, conditions, denominator, threshold and accountable decision owner.
- 01
Does the dataset cover the deployment population?
- 02
How is calibration drift detected?
- 03
Which errors cause crop damage or missed control?
- 04
Can a farmer inspect and override the recommendation?
Evidence to preserve
Enable the next reader to reconstruct conditions, results, failures and the decision.
A conclusion alone loses reviewability; raw records, configuration and exclusions matter too.
- 01
Evidence package 1
Dataset provenance and coverage table
- 02
Evidence package 2
Calibrated error matrix by crop stage
- 03
Evidence package 3
Agronomic review and override record
Scope boundary
State what this evidence still cannot be generalised to.
- 01
Image accuracy does not establish field efficacy when sampling, prevalence, treatment thresholds or crop response differ.
- 02
This is an engineering reading and decision framework; it does not replace application-specific risk assessment, conformity work, procurement acceptance or professional advice.
Sources and evidence status
Read standards scope, measurement evidence and application conclusions separately.
Source dates and review status remain visible; external sources open in a new tab.
01
USDA Agricultural Research Service · Programme page verified 2026-08-16
↗02Digital Agriculture Systems Hub (DASH)
Connects standardised metadata, automated deployment, stakeholder design and field research tools such as autonomous crop-imaging platforms.
Food and Agriculture Organization of the United Nations · Published 2022-11-02
↗03The State of Food and Agriculture 2022 — Leveraging automation
Assesses productivity, resilience, sustainability, labour, access and inequality across automation technologies and farming contexts.
China Ministry of Agriculture and Rural Affairs · Published 2026-04-20
↗Call for representative AI + Agriculture application scenarios
Prioritises narrow, deeply applied, replicable scenarios across agricultural research, production, operations and public services.