Battery management software
Turning battery telemetry into monitoring and diagnostic tools that help owners understand their batteries.
AI battery intelligence · Tesla only
Know your Tesla battery's real health, and what it's really worth.
Dr.EV reads your Tesla's data, straight from Tesla fleet telemetry, and returns an honest, AI-verified battery report, state of health, range, hidden risks, and resale value, before they cost you money.
Untracked degradation costs the average owner~$4,200/yr
★★★★★ 4.8 App Store1-day free trial · cancel anytime
Built by battery management engineers.
See our engineering & researchBattery Health● Live
94%
State of Health
Est. range291 mi
Degradation6.0%
Resale impact+$3,100
4.8★
App Store
4.7★
Google Play
9
Registered patents
15
Languages
Engineering behind Dr.EV
Built by battery management engineers.
Dr.EV is built by BatterMachine's battery management system (BMS) engineering team, combining battery management software, battery state algorithms, and real-world vehicle data.
Battery state algorithms
Developing methods to estimate battery state, track degradation, and examine abnormal patterns.
Real-world data
Using vehicle and charging records to study battery behavior in everyday operation.
12,000+
Registered vehicles
Dr.EV platform coverage
400,000+
Charging sessions
Real-world operating data
9
Korean patent registrations
BatterMachine portfolio
14
PCT international applications
International patent filings
4
US patent applications
Pending · not yet granted
1
Chinese patent application
Pending · not yet granted
Company figures as of October 2026. Platform totals describe overall data coverage, not the sample size of the battery-failure study.
Battermachine research
Research behind battery risk detection.
Can battery data help identify vehicles with pack failures? A retrospective study compared detection results with the number of healthy vehicles incorrectly flagged.
Two-model ensemble · 1% target false-positive rate
- Failure vehicles identified
- 79.9%
- 123 of 154 failure vehicles
- Healthy vehicles incorrectly flagged
- 0.990%
- 34 of 3,435 healthy vehicles
- Vehicles in the study
- 3,589
- 154 failure + 3,435 healthy vehicles
This retrospective comparison used previously explored labels. It is not independent validation and does not establish DrEV’s live app performance.
Research chart and full results
| Target false-positive rate | Observed false-positive rate | Two-model ensemble | Three-model ensemble |
|---|---|---|---|
| 0.1% | 0.087%3 / 3,435 | 42.2%65 / 154Median interval: 30.9 days | 50.6%78 / 154Median interval: 35.8 days |
| 1% | 0.990%34 / 3,435 | 79.9%123 / 154Median interval: 30.7 days | 76.6%118 / 154Median interval: 30.3 days |
What the warning interval measures
Median early-warning intervals were 30.3–35.8 days across the two ensembles at the 0.1% and 1% false-positive targets.
Among detected failure vehicles, this interval runs from the first alert to the last observed battery-pack session end. The actual fault onset was unknown, so it does not establish how long before a fault an alert occurred.
These results describe this study cohort, not all registered vehicles. The three-model ensemble combines four residual scores; it includes no measured-spread component.
Battery state & remaining-life model results
These results concern SOH and remaining useful life (RUL) models. They are separate from the retrospective battery-failure study and do not measure failure-warning performance.
NCM / NCAR² 0.838
SOH estimationMAE 1.84 pp
LFPR² 0.795
SOH estimationMAE 1.37 pp
NCM / NCAR² 0.916
RUL estimationMAE 6.14 mo
LFPR² 0.900
RUL estimationMAE 7.07 mo
The hidden cost
Your battery is quietly losing value. You just can't see it.
The dashboard shows a range number, not the truth. Degradation, cell imbalance, and charging habits erode resale value long before any warning light. Dr.EV surfaces it early, so you can act or sell at the right time.
✓See true State of Health, not the optimistic dashboard number.
✓Explore abnormal battery and charging patterns to support further investigation.
✓Get an AI-verified report that proves value when you sell.
Resale value over time8 years
Managed with Dr.EVUnmanaged
Everything in one app
A full diagnostic for the battery you can't open up.
AI Battery Health
A precise State of Health and remaining-life estimate from your Tesla's real fleet telemetry, updated weekly.
Charging Efficiency
See how much energy actually reaches the pack and which charging habits are costing you range.
Battery Anomaly Alerts
Alerts highlight abnormal battery patterns to support further investigation and informed maintenance decisions.
Range & Degradation
Track real-world range and degradation trends over time, benchmarked against similar Teslas.
Resale Certificate
Generate an AI-certified battery report buyers and insurers trust, worth real money at sale.
Built for Tesla
Deep, Tesla-only integration across every model. No dongle, no extra hardware, just connect.
See it on your phone
The engine inside
Powered by BatterMachine's ML core engine.
Dr.EV brings together BatterMachine's battery engineering, physics-informed machine learning, and real-world operating data. Our research examines battery state, degradation, and patterns associated with battery failure.
01 · Data Ingestion →
Charging sessions
Cell voltages
Temperature
State of charge
Data Foundation
unified per-vehicle session store
02 · Physics Layer →
Physics Signal Layer
Interpretable signatures extracted from raw telemetry, shared by both branches.
03 · Two-Branch AI →
Health & Lifetime
Degradation Trends → Monotonic Transformer ×3 → SOH · RUL
Behavior & Risk
Encoder → Decoder → Forecaster
autoencoder + recurrent forecaster
04 · Outputs
SOH
RUL
Alert
Anomaly
Example model outputs
Mileage · shaded = confidence band
78
months
19
cycles ×100
24
miles ×10k
Remaining life · point + interval
70.0
Normal
15.0
Watch
9.9
Early Warning
5.1
High Risk
Share of fleet by risk tier
Session #lowhigh
Representative model outputs. Battery state and remaining-life estimates are distinct from the battery-failure study.
What is Physics-Informed AI?
It fuses electrochemical physics with machine learning, the model learns how a battery actually behaves, not just data patterns, so it stays accurate even where data is thin.
Real-time ingestion
Charging telemetry streams in from active Teslas.
Continuous learning
Models refresh automatically as new data arrives.
Rising accuracy
More data means sharper diagnosis over time.
Cloud Dataset
Data collection
Fleet telemetry, battery chemistry, and degradation trajectories are continuously ingested.
Physics-Guided Learning
Model training
Physics-informed features combine electrochemical domain knowledge with machine learning.
Diagnosis & Prediction
AI inference
Risk engine, anomaly detection, and degradation models produce precise diagnostics.
Outputs & Feedback
Actionable results
Forecasts, alerts, and risk scores feed back to keep refining the model.
More on the engine at battermachine.ai →
Protected intelligence
Patented battery intelligence, built into Dr.EV.
BatterMachine's patent portfolio includes 9 Korean patent registrations, 14 PCT international applications, 4 pending US patent applications, and 1 pending Chinese patent application. Selected registrations below cover state estimation, safety, cloud intelligence, and adaptive monitoring.
Selected Korean patent registrations
Diverse input-vector DNN model generation for battery state estimation
KR 10-2881214
Safe battery management with limited data using a decoder
KR 10-2895371
Learning model generation for battery state estimation via conditional learning
KR 10-2911543
Battery management considering environmental variables from product usage
KR 10-2911550
Cloud-based battery management system and method
KR 10-2911556
Adaptive battery monitoring during product sleep mode
KR 10-2905098
Portfolio status · October 2026 · Selected examples
Explore the patent portfolioLoved by Tesla owners
Rated 4.8★ by Tesla owners
★★★★★
Professional battery management
“After using it for a day, I found there was a lot to explore. I keep opening the app because it provides so much information, including driving data, charging data, battery score, and estimated range. Vehicle alerts are easy to overlook until you tap them on the car screen, so it was great to receive them through the app as well. I would also appreciate more detailed explanations of what the various figures mean and which values are normal or abnormal.”
M3du11a · 2025. 12. 26. · App Store review · translated from Korean
★★★★★
Packed with varied and essential information
“While connecting and setting up the vehicle, I could easily check a variety of battery-related information, and the explanations made it easy to understand. It feels like a very convenient app to use, and I'm really looking forward to using it more. If possible, could you add manual preconditioning to the vehicle controls? It would be useful when using third-party fast chargers.”
사태야호 · 2025. 11. 12. · App Store review · translated from Korean
★★★★★
An essential EV management app
“You might not need it if you never think about these things, but if you want to manage your EV because of battery-related concerns, this is an essential app. It seems especially necessary for Tesla owners. The intuitive, detailed explanations and responsiveness make it feel trustworthy to use. I hope the app continues to receive consistent updates.”
이중삼중의 · 2025. 09. 21. · App Store review · translated from Korean
Your battery report is one tap away.
Download Dr.EV and run your first AI diagnosis free.