01 Route
How do we compose many models and providers behind one reliable interface?
Research
The HOPN Lab studies one question:
How do we build AI systems that organisations can trust, audit and afford to run themselves?
Programmes
Five programmes break the question down. Each one names what it will measure before any result is published.
02 Ground
What does the system know, and how do we prove where it came from?
03 Assure
How do we know each output is good enough, at runtime, before it ships?
04 Foundation
What quality per watt and per euro is possible when data never leaves the organisation?
05 Physical
How do Route, Ground and Assure apply where errors have physical consequences?
Projects
Projects connected to the programmes. Funders and partners are listed only once they have agreed in writing.
| Project | Type | Programme | HOPn role | Status |
|---|---|---|---|---|
| SWARMind Software for coordinating groups of autonomous drones: a shared operating layer, on-board algorithms and mission planning. | Research project | Physical | Coordinator | Concept |
| SAFE-CARE An architecture for AI-supported assistance in elder care. Every planned action passes a safety gate before it is carried out. | Research project | Physical | Lead | Concept |
| RefereeX AI Decision support for football referees. Several camera feeds are combined with computer vision to give officials replays and a confidence score. The final decision stays with the referee. | Prototype | Developer | Concept | |
| AI-Pass Infrastructure layer for the other programmes. | Research project | Route | Developer | In progress |
| TEAOA Trust model for orchestration. | Research project | Route | Developer | In progress |
| SLDE-AFT Corroborated extraction and offline LoRA extraction feeding the knowledge graph. | Research project | Ground, Foundation | Developer | In progress |
| Semvec Temporal and versioned memory. | Research project | Ground | Developer | In progress |
| ServAlign Constraints and policy as machine-checkable objects. | Research project | Assure | Developer | Planned |
| Invoice automation Shared testbed for the Route, Ground and Assure programmes. | Research project | Route, Ground, Assure | Developer | In progress |
| Sovra AI Compute efficiency across hardware sizes, private retrieval over documents and graph pruning. | Research project | Foundation, Ground | Developer | In progress |
Roadmap
Planned work for the next twelve months, counted from the start of the programme.
| months | Programme | Summary | Status |
|---|---|---|---|
| 0–2 | Route | AI-Pass submission | In progress |
| 0–3 | Foundation | Sovra raw data and scoping | In progress |
| 0–3 | Assure | Quality orchestration charter and shared invoice benchmark | Planned |
| 0–3 | Physical | Simulation baseline with fault injection (ROS 2, Gazebo) | Planned |
| 3–6 | Ground | Graph-pruning study | Planned |
| 3–6 | Assure | First quality-routing results | Planned |
| 3–6 | Physical | Degradation curves and verification methodology note | Planned |
| 6–12 | Route | Integrated pipeline paper | Planned |
| 6–12 | Foundation | Open benchmark for extraction or orchestration under constraint | Planned |
Partners and grants
We list academic and funding partners on a separate page, grouped by relationship.
Contact a researcher
For research collaboration, write to us or use the partner application and choose Research or academic collaboration.