FLA · Step 1, sensor-layout quick check · 15 Nov 2026 · all reports · summary

How many pressure points, and is shear worth it?

A quick check after the second simulation run of step 1. The simulated pressure map of the Booster K1's sole was pooled into 4, 8 or 18 pads per foot or kept at full resolution (152 points), each with and without four shear vectors per foot. Each layout was scored on gaits it saw, on gaits it did not see, and under eight changes of the sensor model. Two real legged-robot datasets were used as a distant check.

Simulation, plus real data from quadrupeds only. A reduced check: the full planned experiment was not completed.

In short

  1. No single number of pressure points is best. More points help on seen gaits (0.782 at four pads to 0.856 at full resolution) and on an unseen gait, but under a changed sensor model the ranking reverses: 0.727 at four pads against 0.084 at full resolution with a foot-shaped mask.
  2. Four shear vectors help every tested layout. At four pads: 0.782 → 0.821 on seen gaits, and +0.116 and +0.142 on the two unseen gaits. The smallest useful shear count is unknown: one- and two-vector layouts were not tested.
  3. Shear does not help more on the human-like gait than on the flat-foot one; that hypothesis is not supported.
  4. No layout earns a robust recommendation. Four pads are too coarse; finer layouts are fragile to a change of sensor.
  5. On real quadruped data, three-axis force beats normal force alone at an unseen speed on one dataset (89.0 % against 83.4 %) but not on the second. Neither validates a humanoid pressure layout.
1

What was done

2

All layouts at a glance

Layout per footSensing elementsSeen gaitsUnseen SONICUnseen G113Two-gait meanChanged sensor, 8-set meanLabel
4 pads40.782 ± 0.0030.223 ± 0.0170.460 ± 0.0100.342 ± 0.0140.693 ± 0.003insufficient
4 pads + shear80.821 ± 0.0060.340 ± 0.0190.602 ± 0.0200.471 ± 0.0190.742 ± 0.003insufficient
8 pads80.817 ± 0.0010.300 ± 0.0200.564 ± 0.0180.432 ± 0.0180.617 ± 0.004fragile
8 pads + shear120.851 ± 0.0030.407 ± 0.0060.721 ± 0.0100.564 ± 0.0060.686 ± 0.006fragile
18 pads180.839 ± 0.0030.318 ± 0.0160.570 ± 0.0120.444 ± 0.0030.576 ± 0.008fragile
18 pads + shear220.866 ± 0.0010.404 ± 0.0110.711 ± 0.0100.557 ± 0.0030.632 ± 0.018fragile
Full map, 152 points1520.856 ± 0.0010.418 ± 0.0240.653 ± 0.0280.535 ± 0.0120.478 ± 0.013fragile
Full map + shear1560.881 ± 0.0020.526 ± 0.0170.717 ± 0.0470.622 ± 0.0230.530 ± 0.007fragile

Reading the table. Read across a row: every layout loses most on an unseen gait, and finer layouts lose most under a changed sensor. Read down a column: seen-gait and unseen-gait scores rise with resolution, while the changed-sensor score falls. Adding shear raises every row in the seen and unseen columns. The label is the verdict of a rule set in advance, applied here at reduced scope: four-pad layouts are "insufficient", all finer ones "fragile".

3

How many pressure points

Without shear, seen-gait F1 rises from 0.782 ± 0.003 at four pads to 0.856 ± 0.001 at full resolution, and unseen SONIC from 0.223 ± 0.017 to 0.418 ± 0.024. A changed sensor reverses that ranking. With a foot-shaped mask, four pads still score 0.727 ± 0.007, the full map 0.084 ± 0.018. So no number of pressure points can be recommended from this simulation: the answer depends on a sensor model that has not been measured.

Line chart of terrain macro-F1 against pressure points per foot (4, 8, 18, 152) without shear: seen gaits rise slowly, the unseen-gait mean rises, the changed-sensor mean falls
Without shear. Blue: seen gaits; orange: unseen-gait mean; green: changed-sensor mean over eight test sets.

An earlier draft quoted 0.80 for four pads without shear; that was the score on the nominal sensor (0.799 ± 0.002), not the seen-gait score on the training data (0.782 ± 0.003) used here.

4

Is shear worth it

Gain in terrain F1 from adding four shear vectors, paired by seed, for each pressure layout.

Pressure layoutSeen gaitsUnseen SONICUnseen G113Changed sensor, mean
4 pads+0.039 ± 0.005+0.116 ± 0.034+0.142 ± 0.030+0.049 ± 0.005
8 pads+0.034 ± 0.004+0.107 ± 0.026+0.157 ± 0.027+0.069 ± 0.009
18 pads+0.027 ± 0.003+0.086 ± 0.015+0.140 ± 0.008+0.056 ± 0.026
Full map+0.026 ± 0.001+0.108 ± 0.021+0.065 ± 0.048+0.052 ± 0.007

Reading the table. Shear helps every tested layout: a little on seen gaits, much more on an unseen gait. At four pads, seen-gait F1 goes from 0.782 ± 0.003 to 0.821 ± 0.006, and the gain holds on unseen SONIC, on unseen G113 and on the changed-sensor mean. Whether shear helps on each changed-sensor test set individually was not established. The minimum useful number of shear vectors is unknown, because the one- and two-vector layouts were cut.

Three line charts of terrain macro-F1 against pressure points per foot, with and without four shear vectors, for seen gaits, unseen SONIC and unseen G113
With four shear vectors (orange) and without (blue): seen gaits, unseen SONIC, unseen G113.
5

Shear and the unseen gait

The first run raised a hypothesis: shear may buy robustness to a change of gait rather than accuracy on a known gait. Two questions follow.

Bar chart of terrain macro-F1 gained from shear on unseen SONIC and unseen G113 for 4, 8, 18 and 152 pressure points per foot, with error bars
Gain from shear on the two unseen gaits, by pressure layout. The full-map gain on G113 has a wide spread.
The hypothesis is not confirmed in its original strict form. The replication with the second run's block layout and the complete held-out-gait mean were cut, and the rule set in advance excludes pooled four-pad results from the strict verdict.
6

Which layout

No layout earns a robust recommendation under the original rule. The two candidates at either end:

Line chart of terrain macro-F1 against pressure points per foot with four shear vectors: seen gaits, unseen-gait mean and changed-sensor mean, the last two crossing between 18 and 152 points
With four shear vectors. The unseen-gait and changed-sensor objectives favour different layouts.

Reading the chart. The unseen-gait and changed-sensor lines cross: one objective asks for more points, the other for fewer. Which one matters more depends on how close the real sole is to the simulated one, which only a real recording can show. The course metrics and the full recommendation grid were not completed.

7

Real robot data

No humanoid pressure-sole dataset was available, so the shear question was tested by analogue on QCAT: a public terrain dataset from a real quadruped with one three-axis force vector per foot. "Normal" uses only the vertical force, "three-axis" the full vector (the analogue of shear against no shear). Testing at an unseen speed stands in for an unseen gait. The same temporal network as in simulation, five seeds, accuracy in percent.

Input10-fold by step10-fold by trialLeave one speed outLeave one step frequency out
Normal force92.1 ± 0.388.6 ± 0.383.4 ± 0.671.6 ± 1.3
Three-axis force94.3 ± 0.492.7 ± 0.489.0 ± 0.479.8 ± 1.6
IMU97.7 ± 0.292.5 ± 1.178.1 ± 0.973.9 ± 0.8
Three-axis force + IMU98.2 ± 0.296.3 ± 0.488.8 ± 0.583.8 ± 1.4

Reading the table. The shear analogue holds on QCAT: at an unseen speed the three-axis force reaches 89.0 ± 0.4 % against 83.4 ± 0.6 % for normal force alone. The sole-only analogue also points the same way as in simulation: at an unseen speed the motion sensor alone drops to 78.1 %, while force alone (89.0 %) does as well as force with IMU (88.8 %). The 10-fold step-level column is the one comparable with published results (force 96.6 %, IMU 96.63 %, force + IMU 97.64 %, Ahmadi et al. 2021).

8

Limits and what was cut

Scores are terrain macro-F1 on simulated walking test windows over six terrains, mean ± population sd over three seeds, with sole-only models; real-data scores are accuracy in percent over five seeds. Related: step 1, first simulation run, step 1, second simulation run, the simulated sole up close, step 1 summary.