Why it matters
Without this system, misreading ground conditions can lead to poor entries, underperforming performances on unsuitable surfaces, or even welfare concerns if sprinting or jumping on risky footing is forced.
From design to safety
Design
AI groundwork for race planning
The AI design phase defines data sources: going history, forecast accuracy, moisture profiles, drainage notes, surface proofs, and rider interviews; it establishes decision rules for race selections and training emphasis with explicit thresholds and confidence levels, plus privacy and consent workflows.
- gallops surface
- stable blocks
- treadmill
- walker
Build
Data pipelines and sensor network
Install ground sensors, GPS nodes, moisture probes, and flagging beacons; calibrate instruments against standard tests, build secure data pipelines, schema catalogs, and a versioned data lake with audit trails.
- gallops surface
- stable blocks
- walker
- ground reporting posts
Operation
Ground-aware training and entries
AI analyzes daily going reports, sectional times, and past form to adjust training focus and propose race entries, with trainer approvals; it generates daily briefings for staff and owners.
- gallops surface
- paddock boards
- ground reporting posts
- scale and moisture probes
Maintenance
Ground health stewardship
Ongoing checks on surface compaction, moisture, drainage performance, and seasonal resurfacing; calibrate sensors, update models with fresh race data, and log surface modifications with time stamps.
- gallops surface
- drainage pipes
- ground testing kit
- paddock fences
Safety
Ground suitability and welfare assurance
Regulatory welfare checks confirm withdrawal times, passport updates, and declarations; safety audits verify that going aligns with rider skill, hoove health, and veterinary guidance before race entries.
- stable blocks
- gallops surface
- ground reporting posts
- tack room
Part by part
- 1
gallops surface
- Watch for
- Ground hardness variance or unexpected soft patches indicating unsafe running conditions
- AI action
- Read surface stiffness and moisture; trigger alert if thresholds exceed set limits.
- 2
stable blocks
- Watch for
- Ventilation inefficiency or bedding compaction affecting hoof health and turnout
- AI action
- Assess air quality and bedding moisture; flag need for aeration or bedding refresh.
- 3
treadmill
- Watch for
- Inconsistent belt wear or motor heat suggesting mechanical drift
- AI action
- Monitor belt temperature and wear rate; schedule maintenance if deviation detected.
- 4
walker
- Watch for
- Ineffective speed control or abnormal stride patterns in rehab horses
- AI action
- Analyze gait data; adjust rehab pace and volume in training plan.
- 5
ground reporting posts
- Watch for
- Delayed or inaccurate going notes from staff
- AI action
- Prompt staff for standardized going entries and timestamp annotations.
- 6
scale and moisture probes
- Watch for
- Moisture and depth shifts altering surface consistency
- AI action
- Flag surface density and moisture changes; trigger surface management actions.
Worked example
Adjusting a flat yard’s entries after a soggy week
Illustrative scenario with sample data.
A yard with twenty horses needs a revised race plan when forecast shows soft to heavy going midweek and potential firm later in the week.
Signal in
goingForecast: soft, moisture: 18mm, pastForm: H1, groundHistory: mixed on soft vs firm, sectionalTimes: 102.5s at 1200m, horses: 20, trainerNote: prefer turf/soft targets
System out
AI recommends two entries on soft ground for two horses, revises training load to suit softer footing, and flags a potential re-route if forecast firming persists.
Human sign-off: Licensed trainer approves entries and training adjustments; yard veterinary surgeon reviews welfare impact and withdrawal times as needed.
How to roll it out
Phase 1
Weeks 1 to 4: Baseline
Install data feeds, calibrate sensors, and establish baseline going profiles and decision rules.
Phase 2
Weeks 5 to 8: Calibration
Tune models with live data, validate going readings against race results, and refine race selection logic.
Phase 3
Weeks 9 to 12: Operational
Implement daily going analyses, auto-suggest entries, and human approve workflow across staff.
Where people stay in charge
- All race entries and welfare decisions remain signed off by the licensed trainer or yard veterinary surgeon as required by regulation.
- Data access is restricted to authorised personnel with full audit trails.
- Any significant model drift requires immediate review and revalidation by a named regulatory liaison.
Questions owners and crews ask
What sensors collect going and ground data?
The system uses gallops surface sensors, moisture probes, and ground reporting posts to build a ground health picture and inform decisions.
How does AI influence race entries?
AI analyzes going, forecast, past form, and sectional times to propose or veto race entries, with final approval by the trainer.
Who signs off on welfare implications?
The licensed trainer and yard veterinary surgeon must sign off on welfare changes and race entries where riders or hoof health could be affected.
How is data kept safe and compliant?
Data governance ensures audit trails, data access controls, and adherence to regulatory reporting standards for stability and integrity.