Read automatically from the scorecard below — every claim can be checked against it.
75 % of the score comes from charging (T4, 46.98 % × weight 0.100) and initial soc error (T10, 26.42 % × weight 0.100) — improving there moves the leaderboard position most.
Accuracy holds on the never-released cell (0.84 % vs 0.84 % on the open cells) — the model generalises rather than memorising the released data.
Cold is the weak spot: 1.09 % at −20 °C against 0.56 % 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 (46.98 % vs 0.84 % overall) — typical of estimators tuned only on discharge behaviour.
A wrong initial SOC is not corrected: 26.42 % 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)
0.84 %
0.0000
0.0000
T2Blinded cell (m448)
0.84 %
0.1000
0.0838
T3Non-blinded cells
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 0.85 %, max 1.1 %
Actual SOC
Torino_Sara_NARX (Previous Contest)
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 6.20 %, max 9.4 %
Actual SOC
Torino_Sara_NARX (Previous Contest)
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 13.89 %, max 20.9 %
Actual SOC
Torino_Sara_NARX (Previous Contest)
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
0.33
0.26
0.66
m80
-10 °C
HWFET
3.02 h
0.68
0.59
1.18
m80
-10 °C
LA92
4.74 h
0.42
0.30
0.90
m80
-10 °C
US06
2.30 h
0.52
0.45
0.93
m80
-10 °C
HWCUST
1.69 h
0.24
0.20
0.45
m80
-10 °C
HWGRADE
1.24 h
0.44
0.34
0.85
m80
-20 °C
UDDS
4.62 h
0.85
0.75
1.13
m80
-20 °C
HWFET
2.34 h
1.52
1.35
2.13
m80
-20 °C
LA92
3.54 h
1.33
1.07
2.34
m80
-20 °C
US06
1.86 h
1.30
1.05
2.24
m80
-20 °C
HWCUST
1.37 h
1.02
0.90
1.45
m80
-20 °C
HWGRADE
1.25 h
0.51
0.46
0.74
m80
0 °C
UDDS
8.75 h
0.67
0.48
1.44
m80
0 °C
HWFET
3.84 h
0.44
0.33
0.90
m80
0 °C
LA92
6.39 h
0.76
0.52
1.58
m80
0 °C
US06
2.93 h
0.42
0.33
0.82
m80
0 °C
HWCUST
2.08 h
0.19
0.12
0.44
m80
0 °C
HWGRADE
1.64 h
0.65
0.50
1.31
m80
10 °C
UDDS
11.32 h
0.60
0.40
1.40
m80
10 °C
HWFET
4.49 h
0.81
0.69
1.40
m80
10 °C
LA92
7.88 h
0.51
0.42
0.97
m80
10 °C
US06
3.42 h
1.00
0.83
1.79
m80
10 °C
HWCUST
2.42 h
0.11
0.09
0.19
m80
10 °C
HWGRADE
1.80 h
0.63
0.52
1.20
m80
25 °C
UDDS
16.13 h
0.93
0.72
1.87
m80
25 °C
HWFET
5.52 h
0.49
0.45
0.87
m80
25 °C
LA92
10.58 h
0.38
0.31
0.81
m80
25 °C
US06
4.22 h
0.73
0.67
1.31
m80
25 °C
HWCUST
2.86 h
0.50
0.47
0.75
m80
25 °C
HWGRADE
2.33 h
0.32
0.31
0.47
m80
40 °C
UDDS
12.40 h
1.46
1.20
3.57
m80
40 °C
HWFET
5.41 h
1.50
1.35
2.84
m80
40 °C
LA92
9.07 h
1.39
1.20
3.12
m80
40 °C
US06
4.17 h
1.64
1.47
2.83
m80
40 °C
HWCUST
2.95 h
0.92
0.86
1.49
m80
40 °C
HWGRADE
2.73 h
0.84
0.80
1.24
m448
-10 °C
UDDS
6.23 h
0.35
0.27
0.65
m448
-10 °C
HWFET
3.01 h
0.80
0.69
1.37
m448
-10 °C
LA92
4.59 h
0.64
0.51
1.33
m448
-10 °C
US06
2.25 h
0.87
0.70
1.52
m448
-10 °C
HWCUST
1.69 h
0.42
0.35
0.73
m448
-10 °C
HWGRADE
1.22 h
0.35
0.28
0.66
m448
-20 °C
UDDS
4.56 h
0.87
0.76
1.25
m448
-20 °C
HWFET
2.34 h
1.70
1.50
2.44
m448
-20 °C
LA92
3.46 h
1.43
1.17
2.38
m448
-20 °C
US06
1.80 h
1.49
1.20
2.49
m448
-20 °C
HWCUST
1.39 h
0.90
0.80
1.36
m448
-20 °C
HWGRADE
1.23 h
0.33
0.31
0.51
m448
0 °C
UDDS
8.60 h
0.81
0.61
1.65
m448
0 °C
HWFET
3.83 h
0.39
0.29
0.83
m448
0 °C
LA92
6.16 h
0.56
0.41
1.17
m448
0 °C
US06
2.87 h
0.50
0.41
0.95
m448
0 °C
HWCUST
2.09 h
0.36
0.28
0.73
m448
0 °C
HWGRADE
1.51 h
0.46
0.35
0.98
m448
10 °C
UDDS
11.08 h
0.73
0.51
1.61
m448
10 °C
HWFET
4.47 h
0.69
0.59
1.21
m448
10 °C
LA92
7.60 h
0.36
0.32
0.62
m448
10 °C
US06
3.35 h
0.75
0.66
1.26
m448
10 °C
HWCUST
2.42 h
0.10
0.09
0.21
m448
10 °C
HWGRADE
1.68 h
0.65
0.52
1.22
m448
25 °C
UDDS
15.59 h
0.84
0.65
1.74
m448
25 °C
HWFET
5.46 h
0.39
0.35
0.80
m448
25 °C
LA92
10.06 h
0.32
0.28
0.61
m448
25 °C
US06
4.07 h
1.46
1.11
3.17
m448
25 °C
HWCUST
2.86 h
0.51
0.47
0.80
m448
25 °C
HWGRADE
2.19 h
0.37
0.36
0.58
m448
40 °C
UDDS
12.09 h
1.59
1.31
3.90
m448
40 °C
HWFET
5.21 h
1.51
1.35
2.90
m448
40 °C
LA92
8.79 h
1.41
1.25
2.99
m448
40 °C
US06
4.06 h
3.35
2.69
6.24
m448
40 °C
HWCUST
2.92 h
0.95
0.89
1.54
m448
40 °C
HWGRADE
2.32 h
0.94
0.88
1.48
m448-N
-10 °C
UDDS
10.80 h
2.05
1.67
3.48
m448-N
-10 °C
HWFET
3.85 h
0.32
0.23
0.69
m448-N
-10 °C
LA92
6.66 h
0.59
0.42
1.20
m448-N
-10 °C
US06
2.70 h
0.48
0.38
0.96
m448-N
-10 °C
HWCUST
1.89 h
0.09
0.07
0.20
m448-N
-10 °C
HWGRADE
1.36 h
0.79
0.65
1.42
m448-N
-20 °C
UDDS
8.67 h
1.86
1.72
2.67
m448-N
-20 °C
HWFET
3.09 h
1.17
1.02
1.66
m448-N
-20 °C
LA92
5.21 h
0.36
0.30
0.69
m448-N
-20 °C
US06
2.23 h
1.35
1.19
2.25
m448-N
-20 °C
HWCUST
1.65 h
1.05
0.90
1.57
m448-N
-20 °C
HWGRADE
1.50 h
0.31
0.27
0.61
m448-N
0 °C
UDDS
12.79 h
2.10
1.79
3.57
m448-N
0 °C
HWFET
4.47 h
0.28
0.20
0.62
m448-N
0 °C
LA92
7.99 h
0.54
0.36
1.22
m448-N
0 °C
US06
3.22 h
0.21
0.16
0.49
m448-N
0 °C
HWCUST
2.31 h
0.14
0.11
0.35
m448-N
0 °C
HWGRADE
1.65 h
0.97
0.78
1.85
m448-N
10 °C
UDDS
13.92 h
1.47
1.18
2.76
m448-N
10 °C
HWFET
4.84 h
0.33
0.28
0.63
m448-N
10 °C
LA92
8.80 h
0.46
0.33
1.09
m448-N
10 °C
US06
3.56 h
0.41
0.33
0.78
m448-N
10 °C
HWCUST
2.53 h
0.41
0.34
0.73
m448-N
10 °C
HWGRADE
1.79 h
1.06
0.87
1.94
m448-N
25 °C
UDDS
15.61 h
1.13
0.88
2.07
m448-N
25 °C
HWFET
5.47 h
0.22
0.19
0.43
m448-N
25 °C
LA92
10.09 h
0.59
0.49
1.18
m448-N
25 °C
US06
4.08 h
1.68
1.32
3.57
m448-N
25 °C
HWCUST
2.86 h
0.25
0.23
0.35
m448-N
25 °C
HWGRADE
2.30 h
0.14
0.14
0.25
m448-N
40 °C
UDDS
16.75 h
1.37
1.01
3.86
m448-N
40 °C
HWFET
5.89 h
1.25
1.12
2.52
m448-N
40 °C
LA92
11.03 h
1.08
0.89
2.62
m448-N
40 °C
US06
4.44 h
1.84
1.54
3.38
m448-N
40 °C
HWCUST
3.13 h
0.69
0.65
1.17
m448-N
40 °C
HWGRADE
2.48 h
0.72
0.68
1.09
m1000
-10 °C
UDDS
5.91 h
0.40
0.29
0.95
m1000
-10 °C
HWFET
2.91 h
0.84
0.73
1.42
m1000
-10 °C
LA92
4.23 h
0.86
0.69
1.67
m1000
-10 °C
US06
2.09 h
0.84
0.68
1.49
m1000
-10 °C
HWCUST
1.67 h
0.42
0.36
0.71
m1000
-10 °C
HWGRADE
1.20 h
0.55
0.42
1.02
m1000
-20 °C
UDDS
4.31 h
0.95
0.80
1.57
m1000
-20 °C
HWFET
2.27 h
1.56
1.39
2.20
m1000
-20 °C
LA92
3.20 h
1.61
1.29
2.81
m1000
-20 °C
US06
1.74 h
3.45
3.41
4.02
m1000
-20 °C
HWCUST
1.37 h
0.82
0.74
1.24
m1000
-20 °C
HWGRADE
1.22 h
0.34
0.30
0.61
m1000
0 °C
UDDS
8.09 h
0.62
0.42
1.35
m1000
0 °C
HWFET
3.70 h
0.44
0.34
0.90
m1000
0 °C
LA92
5.60 h
0.73
0.51
1.57
m1000
0 °C
US06
2.66 h
0.54
0.47
0.96
m1000
0 °C
HWCUST
2.05 h
0.36
0.29
0.73
m1000
0 °C
HWGRADE
1.36 h
0.67
0.49
1.38
m1000
10 °C
UDDS
10.51 h
0.54
0.35
1.25
m1000
10 °C
HWFET
4.37 h
0.67
0.58
1.16
m1000
10 °C
LA92
7.06 h
0.45
0.40
0.74
m1000
10 °C
US06
3.11 h
1.17
1.01
2.05
m1000
10 °C
HWCUST
2.41 h
0.10
0.09
0.20
m1000
10 °C
HWGRADE
1.52 h
0.81
0.64
1.51
m1000
25 °C
UDDS
14.69 h
0.66
0.51
1.47
m1000
25 °C
HWFET
5.37 h
0.40
0.36
0.77
m1000
25 °C
LA92
9.30 h
0.52
0.45
1.23
m1000
25 °C
US06
3.79 h
2.55
1.97
5.31
m1000
25 °C
HWCUST
2.83 h
0.52
0.49
0.77
m1000
25 °C
HWGRADE
1.90 h
0.29
0.27
0.40
m1000
40 °C
UDDS
11.69 h
1.48
1.25
3.47
m1000
40 °C
HWFET
5.15 h
1.53
1.38
2.83
m1000
40 °C
LA92
8.33 h
1.54
1.34
3.16
m1000
40 °C
US06
3.89 h
3.11
2.51
5.75
m1000
40 °C
HWCUST
2.91 h
0.89
0.84
1.43
m1000
40 °C
HWGRADE
2.17 h
0.91
0.86
1.34
m80 UDDS at -20 °C
Whole cycle · RMSE 0.85 %, max 1.1 %
Actual SOC
Torino_Sara_NARX (Previous Contest)
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.
Aug 31, 2026, 04:15 a.m. EDTEvaluation failed · realattempt 1: Evaluation exceeded the 180 minute limit.failed
How a other worksThe canonical structure for this model family; the author's description below gives the specific architecture.
Other — standardized view
Unspecified
Governing equation
[Y,z]=Model(X,z)
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).
Nonlinear Autoregressive Network with Exogenous Inputs
Package
Torino_Sara_Updated.zip (14.3 KB) — deleted after evaluation
Visibility
Public
Evaluator
socbench-eval-0.1.0/matlab
Completed
Aug 31, 2026, 10:51 a.m. EDT
Submission ID
cmtgsc6h40003ky04cu7bands
0.84 %
0.1000
0.0843
Operating conditionsT4Charging
46.98 %
0.1000
4.6981
T580 kg payload
0.75 %
0.0333
0.0250
T5448 kg payload, HVAC on
0.84 %
0.0667
0.0559
T6448 kg payload, HVAC off
0.83 %
0.0667
0.0551
T51000 kg payload
0.95 %
0.0333
0.0316
T7Standard drive cycles
0.99 %
0.1000
0.0991
T8Non-standard drive cycles
0.54 %
0.1000
0.0543
T9−20 °C
1.09 %
0.0167
0.0182
T9−10 °C
0.44 %
0.0167
0.0073
T90 °C
0.52 %
0.0167
0.0087
T910 °C
0.61 %
0.0167
0.0101
T925 °C
0.56 %
0.0167
0.0093
T940 °C
1.29 %
0.0167
0.0215
RobustnessT10Initial SOC error
26.42 %
0.1000
2.6422
T11Current sensor offset
18.83 %
0.1000
1.8828
Weighted error = Σ (weight × RMSE)
9.79 %
Max error 6.2 % — largest instantaneous |estimate − truth| over every blinded cycle (on m448 US06 at 40 °C — plotted under Key cases for new evaluations); not scoredComplexity 10 ±1 · Extreme+ (informational, not scored)
Ahnaf Rahman · McMaster University