Read automatically from the scorecard below — every claim can be checked against it.
73 % of the score comes from charging (T4, 80.77 % × weight 0.100) and initial soc error (T10, 31.35 % × weight 0.100) — improving there moves the leaderboard position most.
Accuracy holds on the never-released cell (2.30 % vs 2.44 % on the open cells) — the model generalises rather than memorising the released data.
Cold is the weak spot: 3.58 % at −20 °C against 1.31 % at 25 °C. Below zero the cell's resistance rises and the voltage curve distorts; models without temperature awareness lose the most here.
Charging is much worse than driving (80.77 % vs 2.40 % overall) — typical of estimators tuned only on discharge behaviour.
A wrong initial SOC is not corrected: 31.35 % on test 10. Pure current integration can never recover from a bad start; voltage feedback is what fixes this.
Scorecard
Average RMSE (% SOC) per test case, weighted into the leaderboard score. Lower is better.
Estimation accuracyT1All cells(reference only — every other test is a subset of it)
2.40 %
0.0000
0.0000
T2Blinded cell (m448)
2.30 %
0.1000
0.2300
T3
Key cases
The runs that separate estimators — cold and hot cycles, the blinded cell, wrong initial SOC, a biased current sensor. Each plot says why it is there.
Selected drive cycles
The thirteen cycles the original MATLAB tool plotted, one per test: blinded vs non-blinded cell (tests 1–3), drive vs charge (test 4), the three payloads on one cycle (tests 5–6), standard vs non-standard highway profiles (tests 7–8) and the temperature extremes (test 9).
m80 UDDS at -20 °C
Whole cycle · RMSE 3.28 %, max 5.7 %
Actual SOC
CC MATLAB EXAMPLE
Error axis
Why this case: Test 9 — coldest condition: high resistance and a strongly non-linear OCV response — the hardest cycle for most estimators.
Wrong initial SOC (test 10)
Three cycles restarted with the estimator told the battery is at 90, 60 or 30 % while it is really full. Coulomb counting can never recover; voltage-informed estimators converge at a rate that is very visible here.
Wrong initial SOC 90 % — m80 LA92 at 25 °C
Whole cycle · RMSE 10.86 %, max 11.8 %
Actual SOC
CC MATLAB EXAMPLE
Error axis
Why this case: The cycle is joined at the point where the true SOC has already fallen to 90 %, so a model that assumes a full battery at start-up begins 10 % off. What matters is how fast the estimate converges onto the truth.
Current-sensor offset (test 11)
Three cycles re-run with ±0.3 A added to the measured current. Integration-based estimators drift steadily; the plot shows whether and how fast the model corrects the bias.
Current-sensor offset -0.3 A — m1000 US06 at -10 °C
Whole cycle · RMSE 11.02 %, max 15.1 %
Actual SOC
CC MATLAB EXAMPLE
Error axis
Why this case: A constant -0.3 A is added to the measured current before it reaches the model. Pure current integration drifts linearly; estimators that also use voltage should correct it.
RMSE vs. sensor offset
Test 11 — the same three cycles re-run with a constant current offset; 0 A is the unmodified run. Only ±0.3 A enters the score.
US06 at −10 °C
HWFET at 10 °C
LA92 at 40 °C
Test-case chartsThe scorecard as bar charts — tests 1–8, and RMSE against temperature (test 9).
Error by test case
Average RMS error (% SOC) for each blinded test case. Lower is better.
Error vs. ambient temperature
Test 9 — m80 cell, all blinded cycles at each chamber temperature.
All 144 cyclesEvery blinded drive cycle: per-cycle errors, and any of the plotted cycles in the time domain.
Cell
Temp.
Cycle
Duration
RMSE
MAE
MAXE
m80
-10 °C
UDDS
6.32 h
3.23
2.77
5.56
m80
-10 °C
HWFET
3.02 h
3.39
2.92
5.67
m80
-10 °C
LA92
4.74 h
3.22
2.76
5.58
m80
-10 °C
US06
2.30 h
3.42
2.94
5.70
m80
-10 °C
HWCUST
1.69 h
3.86
3.36
6.18
m80
-10 °C
HWGRADE
1.24 h
3.88
3.35
6.26
m80
-20 °C
UDDS
4.62 h
3.32
2.87
5.67
m80
-20 °C
HWFET
2.34 h
3.46
2.98
5.73
m80
-20 °C
LA92
3.54 h
3.21
2.75
5.46
m80
-20 °C
US06
1.86 h
3.52
3.04
5.67
m80
-20 °C
HWCUST
1.37 h
3.84
3.35
5.94
m80
-20 °C
HWGRADE
1.25 h
4.10
3.64
6.01
m80
0 °C
UDDS
8.75 h
3.20
2.75
5.59
m80
0 °C
HWFET
3.84 h
3.36
2.89
5.71
m80
0 °C
LA92
6.39 h
3.20
2.74
5.56
m80
0 °C
US06
2.93 h
3.32
2.86
5.59
m80
0 °C
HWCUST
2.08 h
3.76
3.25
6.14
m80
0 °C
HWGRADE
1.64 h
4.00
3.46
6.23
m80
10 °C
UDDS
11.32 h
1.96
1.68
3.47
m80
10 °C
HWFET
4.49 h
2.17
1.87
3.72
m80
10 °C
LA92
7.88 h
2.03
1.73
3.60
m80
10 °C
US06
3.42 h
2.40
2.06
4.08
m80
10 °C
HWCUST
2.42 h
3.23
2.80
5.27
m80
10 °C
HWGRADE
1.80 h
3.45
2.98
5.57
m80
25 °C
UDDS
16.13 h
1.01
0.86
1.78
m80
25 °C
HWFET
5.52 h
1.12
0.96
1.96
m80
25 °C
LA92
10.58 h
1.08
0.92
1.92
m80
25 °C
US06
4.22 h
1.17
1.00
1.99
m80
25 °C
HWCUST
2.86 h
1.71
1.48
2.85
m80
25 °C
HWGRADE
2.33 h
1.80
1.55
3.03
m80
40 °C
UDDS
12.40 h
0.48
0.41
0.84
m80
40 °C
HWFET
5.41 h
0.38
0.32
0.65
m80
40 °C
LA92
9.07 h
0.29
0.25
0.58
m80
40 °C
US06
4.17 h
0.32
0.28
0.54
m80
40 °C
HWCUST
2.95 h
0.56
0.48
0.95
m80
40 °C
HWGRADE
2.73 h
0.81
0.70
1.29
m448
-10 °C
UDDS
6.23 h
2.93
2.53
4.99
m448
-10 °C
HWFET
3.01 h
3.03
2.61
5.07
m448
-10 °C
LA92
4.59 h
2.76
2.36
4.74
m448
-10 °C
US06
2.25 h
2.91
2.51
4.87
m448
-10 °C
HWCUST
1.69 h
3.48
3.02
5.58
m448
-10 °C
HWGRADE
1.22 h
3.63
3.15
5.70
m448
-20 °C
UDDS
4.56 h
2.97
2.56
5.06
m448
-20 °C
HWFET
2.34 h
3.11
2.68
5.16
m448
-20 °C
LA92
3.46 h
2.82
2.41
4.80
m448
-20 °C
US06
1.80 h
3.04
2.62
5.00
m448
-20 °C
HWCUST
1.39 h
3.48
3.04
5.38
m448
-20 °C
HWGRADE
1.23 h
3.75
3.35
5.43
m448
0 °C
UDDS
8.60 h
3.04
2.60
5.34
m448
0 °C
HWFET
3.83 h
3.20
2.75
5.43
m448
0 °C
LA92
6.16 h
2.97
2.53
5.15
m448
0 °C
US06
2.87 h
3.10
2.66
5.21
m448
0 °C
HWCUST
2.09 h
3.39
2.94
5.54
m448
0 °C
HWGRADE
1.51 h
3.53
3.05
5.58
m448
10 °C
UDDS
11.08 h
1.89
1.61
3.37
m448
10 °C
HWFET
4.47 h
2.07
1.78
3.56
m448
10 °C
LA92
7.60 h
2.01
1.72
3.61
m448
10 °C
US06
3.35 h
2.15
1.83
3.70
m448
10 °C
HWCUST
2.42 h
3.13
2.72
5.12
m448
10 °C
HWGRADE
1.68 h
3.30
2.85
5.36
m448
25 °C
UDDS
15.59 h
1.00
0.84
1.82
m448
25 °C
HWFET
5.46 h
1.07
0.92
1.85
m448
25 °C
LA92
10.06 h
1.02
0.85
1.85
m448
25 °C
US06
4.07 h
1.44
1.27
2.36
m448
25 °C
HWCUST
2.86 h
1.64
1.41
2.75
m448
25 °C
HWGRADE
2.19 h
1.71
1.48
2.87
m448
40 °C
UDDS
12.09 h
0.59
0.51
1.05
m448
40 °C
HWFET
5.21 h
0.49
0.41
0.85
m448
40 °C
LA92
8.79 h
0.50
0.45
0.80
m448
40 °C
US06
4.06 h
0.43
0.39
0.69
m448
40 °C
HWCUST
2.92 h
0.51
0.43
0.86
m448
40 °C
HWGRADE
2.32 h
0.71
0.62
1.19
m448-N
-10 °C
UDDS
10.80 h
3.00
2.55
5.30
m448-N
-10 °C
HWFET
3.85 h
3.37
2.90
5.70
m448-N
-10 °C
LA92
6.66 h
3.03
2.59
5.30
m448-N
-10 °C
US06
2.70 h
3.20
2.75
5.34
m448-N
-10 °C
HWCUST
1.89 h
3.82
3.31
6.22
m448-N
-10 °C
HWGRADE
1.36 h
4.06
3.54
6.32
m448-N
-20 °C
UDDS
8.67 h
3.00
2.56
5.35
m448-N
-20 °C
HWFET
3.09 h
3.40
2.93
5.75
m448-N
-20 °C
LA92
5.21 h
2.93
2.50
5.13
m448-N
-20 °C
US06
2.23 h
3.52
3.04
5.73
m448-N
-20 °C
HWCUST
1.65 h
3.90
3.41
6.03
m448-N
-20 °C
HWGRADE
1.50 h
4.23
3.81
6.07
m448-N
0 °C
UDDS
12.79 h
3.31
2.84
5.85
m448-N
0 °C
HWFET
4.47 h
3.53
3.04
6.03
m448-N
0 °C
LA92
7.99 h
3.32
2.82
5.84
m448-N
0 °C
US06
3.22 h
3.44
2.96
5.82
m448-N
0 °C
HWCUST
2.31 h
3.84
3.33
6.22
m448-N
0 °C
HWGRADE
1.65 h
4.07
3.54
6.29
m448-N
10 °C
UDDS
13.92 h
2.13
1.82
3.77
m448-N
10 °C
HWFET
4.84 h
2.36
2.03
4.07
m448-N
10 °C
LA92
8.80 h
2.20
1.87
3.95
m448-N
10 °C
US06
3.56 h
2.53
2.17
4.26
m448-N
10 °C
HWCUST
2.53 h
3.52
3.05
5.77
m448-N
10 °C
HWGRADE
1.79 h
3.77
3.26
5.99
m448-N
25 °C
UDDS
15.61 h
1.31
1.13
2.32
m448-N
25 °C
HWFET
5.47 h
1.35
1.15
2.37
m448-N
25 °C
LA92
10.09 h
1.31
1.10
2.34
m448-N
25 °C
US06
4.08 h
1.34
1.13
2.36
m448-N
25 °C
HWCUST
2.86 h
1.94
1.68
3.26
m448-N
25 °C
HWGRADE
2.30 h
2.10
1.82
3.41
m448-N
40 °C
UDDS
16.75 h
0.37
0.32
0.65
m448-N
40 °C
HWFET
5.89 h
0.27
0.23
0.46
m448-N
40 °C
LA92
11.03 h
0.23
0.20
0.44
m448-N
40 °C
US06
4.44 h
0.07
0.05
0.14
m448-N
40 °C
HWCUST
3.13 h
0.74
0.64
1.26
m448-N
40 °C
HWGRADE
2.48 h
0.97
0.84
1.62
m1000
-10 °C
UDDS
5.91 h
2.85
2.44
4.93
m1000
-10 °C
HWFET
2.91 h
3.08
2.65
5.20
m1000
-10 °C
LA92
4.23 h
2.73
2.34
4.67
m1000
-10 °C
US06
2.09 h
3.01
2.59
5.01
m1000
-10 °C
HWCUST
1.67 h
3.57
3.11
5.69
m1000
-10 °C
HWGRADE
1.20 h
3.89
3.43
5.81
m1000
-20 °C
UDDS
4.31 h
2.97
2.56
5.06
m1000
-20 °C
HWFET
2.27 h
3.19
2.75
5.30
m1000
-20 °C
LA92
3.20 h
2.83
2.42
4.78
m1000
-20 °C
US06
1.74 h
1.75
1.50
2.95
m1000
-20 °C
HWCUST
1.37 h
3.53
3.08
5.45
m1000
-20 °C
HWGRADE
1.22 h
3.92
3.54
5.56
m1000
0 °C
UDDS
8.09 h
3.01
2.58
5.27
m1000
0 °C
HWFET
3.70 h
3.22
2.77
5.46
m1000
0 °C
LA92
5.60 h
2.89
2.46
5.12
m1000
0 °C
US06
2.66 h
3.13
2.69
5.23
m1000
0 °C
HWCUST
2.05 h
3.44
2.98
5.62
m1000
0 °C
HWGRADE
1.36 h
3.62
3.13
5.69
m1000
10 °C
UDDS
10.51 h
1.93
1.64
3.48
m1000
10 °C
HWFET
4.37 h
2.21
1.89
3.79
m1000
10 °C
LA92
7.06 h
2.05
1.72
3.71
m1000
10 °C
US06
3.11 h
2.13
1.80
3.71
m1000
10 °C
HWCUST
2.41 h
3.21
2.78
5.23
m1000
10 °C
HWGRADE
1.52 h
3.42
2.96
5.45
m1000
25 °C
UDDS
14.69 h
1.02
0.86
1.76
m1000
25 °C
HWFET
5.37 h
1.21
1.03
2.05
m1000
25 °C
LA92
9.30 h
1.07
0.89
2.00
m1000
25 °C
US06
3.79 h
1.40
1.19
2.42
m1000
25 °C
HWCUST
2.83 h
1.77
1.53
2.98
m1000
25 °C
HWGRADE
1.90 h
1.76
1.52
3.00
m1000
40 °C
UDDS
11.69 h
0.45
0.39
0.79
m1000
40 °C
HWFET
5.15 h
0.38
0.33
0.67
m1000
40 °C
LA92
8.33 h
0.24
0.22
0.39
m1000
40 °C
US06
3.89 h
0.08
0.07
0.16
m1000
40 °C
HWCUST
2.91 h
0.62
0.53
1.05
m1000
40 °C
HWGRADE
2.17 h
0.81
0.71
1.31
m80 UDDS at -20 °C
Whole cycle · RMSE 3.28 %, max 5.7 %
Actual SOC
CC MATLAB EXAMPLE
Error axis
One hour of padded data precedes every cycle in the evaluator and is excluded from the error metrics. Charts are down-sampled for display (peaks preserved); the full 1 Hz data is in the traces download below.
Score historyEvery evaluation attempt and every change to how this submission is scored; the current score is the last row.
How a coulomb counter worksThe canonical structure for this model family; the author's description below gives the specific architecture.
Coulomb counter — standardized view
Open-loop current integration
Governing equation
SOCk=SOCk−1+3600CnIkΔt
Downloads
PDF report (summary, every test case with the arithmetic, key plots, per-cycle table) · results as JSON · full 1 Hz traces of all 195 runs as a MATLAB v7 file (0.01 % SOC steps; readable with scipy.io.loadmat, see the readme variable inside).
CC MATLAB EXAMPLE.....................................
Package
coulomb-counter.matlab.zip (679 B) — deleted after evaluation
Visibility
Public
Evaluator
socbench-eval-0.1.0/matlab
Completed
Sep 21, 2026, 04:49 p.m. EDT
Submission ID
cmubpwj8o0005yo3gnqpbx1d5
Non-blinded cells
2.44 %
0.1000
0.2436
Operating conditionsT4Charging
80.77 %
0.1000
8.0767
T580 kg payload
2.48 %
0.0333
0.0826
T5448 kg payload, HVAC on
2.30 %
0.0667
0.1533
T6448 kg payload, HVAC off
2.54 %
0.0667
0.1694
T51000 kg payload
2.29 %
0.0333
0.0763
T7Standard drive cycles
2.17 %
0.1000
0.2168
T8Non-standard drive cycles
2.87 %
0.1000
0.2871
T9−20 °C
3.58 %
0.0167
0.0596
T9−10 °C
3.50 %
0.0167
0.0583
T90 °C
3.47 %
0.0167
0.0579
T910 °C
2.54 %
0.0167
0.0423
T925 °C
1.31 %
0.0167
0.0219
T940 °C
0.47 %
0.0167
0.0079
RobustnessT10Initial SOC error
31.35 %
0.1000
3.1350
T11Current sensor offset
24.47 %
0.1000
2.4470
Weighted error = Σ (weight × RMSE)
15.37 %
Max error 6.3 % — largest instantaneous |estimate − truth| over every blinded cycle (on m448-N HWGRADE at -10 °C — plotted under Key cases for new evaluations); not scoredComplexity 1 ±1 · Trivial (informational, not scored)