Models and systems
Research models, evaluation frameworks and physical platforms are developed together, with explicit status, access and limitation records.
No record on this site should be interpreted as a public release or commercial deployment unless that status is stated explicitly.
Research models
Each record identifies finalized and proposed specifications.
Calibrated two-model maize symptom screening system
A public development candidate combining categorical and specialist FP16 TensorFlow Lite models with locked fusion calibration and explicit abstention for offline, human-reviewed maize-leaf symptom screening.
Two-stage offline maize screening system
A research-preview, two-stage LiteRT and TensorFlow Lite vision system evaluated on documented development datasets and limited field samples for offline, human-reviewed maize-leaf screening.
Multimodal field foundation model
FieldMind-1 is a family of multimodal models being developed to understand agricultural environments across images, video, metadata, location and time.
Embodied agricultural policy model
FieldAction-1 investigates how agricultural machines can learn useful field behaviors from teleoperation, demonstrations, interventions, failures and experience across platforms.
Low-resource speech and voice model family
TwiField Speech investigates speech recognition and generation for Asante Twi, Akuapem Twi, Ghanaian English and agricultural code-switching.
Real-world physical-intelligence evaluation framework
FieldBench is being designed to evaluate autonomous and multimodal systems under real agricultural operating conditions.
Experimental platforms
Experimental field-intelligence platform
Farmba Scout is a modular agricultural rover prototype used to study field data collection, perception, teleoperation-to-autonomy and offline physical-intelligence workflows.
Structured field-observation workflow
FieldMind Collector is the working name for an internal workflow intended to capture provenance-aware field observations for longitudinal research.
Access
Publicly released records link directly to their artifacts and terms. Specifications, results and access information for unreleased models will be published after internal and external evaluation.
Read the open research approach