Data and AIP-Based Strategy Platform
A pure-electric Le Mans car needs more than telemetry. It needs a strategy platform.
The car's performance depends on real-time decisions about energy, charging, thermal control, regeneration, traffic, Safety Car probability, and pit timing. No single engineer can manually integrate all of those signals fast enough across 24 hours without decision support.
This is where an AIP-based strategy platform becomes valuable. In this context, AIP means an AI-powered decision platform that connects models, data, operational rules, and human approval into one race system.
The platform should not replace race engineers. It should make their decisions faster, clearer, and more evidence-based.
The Data Problem
The car produces many streams:
- battery cell voltages
- cell temperatures
- module temperatures
- pack current
- SOC estimate
- SOH estimate
- inverter temperatures
- motor temperatures
- coolant flows
- connector temperatures
- brake temperatures
- tire pressures
- regen limits
- driver inputs
- GPS position
- lap sector data
- traffic gaps
- race control status
- weather and track condition
The platform must convert these streams into decisions.
The key question is not "what is the data?" The key question is:
What should the team do now?
Core Platform Functions
The strategy platform should provide:
- real-time energy forecast
- stint completion probability
- charge acceptance forecast
- thermal derating risk
- Safety Car charging recommendation
- minimum-charge and deep-charge decision support
- driver pace target
- regen setting recommendation
- high-voltage system health status
- anomaly detection
- failure mode guidance
The platform must show uncertainty. A forecast with false precision is dangerous. Race engineers need confidence intervals, scenario comparisons, and clear reasons.
Digital Twin Layer
The project should build a digital twin that updates throughout the race.
The twin should model:
- energy per lap
- battery temperature evolution
- charging curve
- pack degradation
- tire impact
- driver pace
- traffic energy effect
- cooling recovery
- derating thresholds
The model should be calibrated during testing and corrected during the race. If the car consumes more energy than expected, the platform should update the remaining stint forecast immediately. If charging tapers earlier than predicted, the next pit decision should change.
Human-in-the-Loop Decisions
The system should recommend, not command.
Race strategy remains a human responsibility because the context includes judgment: driver confidence, competitor behavior, weather uncertainty, pit-lane congestion, and risk appetite. The platform should present options:
- stay out and save energy
- pit now for minimum charge
- pit now for deep charge
- push for two laps and pit under expected yellow
- slow for thermal recovery
- switch regen mode
- prepare emergency high-voltage procedure
Each recommendation should include:
- expected benefit
- risk
- assumptions
- confidence level
- what would change the recommendation
Safety Integration
The platform should also support high-voltage safety.
It can track:
- isolation status
- crash sensor state
- battery enclosure status
- connector temperature
- charge lock status
- emergency stop state
- safe-to-touch indicators
- marshal notification state
This information must be presented in a form that pit crew and race control can understand quickly.
Strategy Memory
A 24-hour race is long enough for human memory to become unreliable under pressure. The platform should maintain a structured event memory:
- each charge event
- energy added
- peak charge power
- charge taper reason
- maximum cell temperature
- driver comments
- derating events
- mode changes
- Safety Car decisions
- anomalies and responses
This memory helps engineers avoid repeating mistakes and supports post-race IP development.
Public Storytelling Layer
The same data can also support public communication, but the storytelling layer must be separated from the operational layer.
Useful public metrics could include:
- energy recovered by regeneration
- number of charging events completed
- highest safe charge power
- total electric distance completed
- battery temperature stability
- total race hours completed
The public story should be accurate and controlled. It should not expose sensitive data or create pressure to make unsafe decisions for a better graphic.
The Platform Statement
The AIP-based strategy platform is successful if it helps the team make better decisions under uncertainty.
The core statement is:
Electric endurance racing is not only a hardware challenge. It is a real-time decision system where energy, heat, safety, and race timing must be optimized together.
The car needs intelligence in the pit wall as much as power in the drivetrain.
← Back to whitepaper