Label joint power bursts with Winter nomenclature

Segments a single-gait-cycle joint power time series into contiguous same-sign bursts and assigns Winter's power-phase labels: A1/A2 for the ankle, H1-H3 for the hip, and K1-K4 for the knee. Each canonical burst has an expected sign (generation vs absorption) and an approximate location within the gait cycle; detected bursts are matched to the nearest unused canonical burst of the same sign.

Usage

labelPowerBursts(
  power,
  sampling_rate,
  joint = c("ankle", "knee", "hip"),
  body_mass = NULL
)

Arguments

power: Numeric vector of joint power (W) over one gait cycle, or a data frame with a power column.
sampling_rate: Sampling rate in Hz.
joint: One of "ankle", "knee", or "hip".
body_mass: Optional body mass in kilograms; if supplied, a work_per_kg column (J/kg) is added.

Value

A data frame (class power_bursts) with one row per detected burst: label (canonical burst name or NA if unmatched), joint, type ("generation" or "absorption"), start_pct, end_pct, peak_power (W, signed), and work (J), ordered by time.

Details

The power vector is assumed to span exactly one gait cycle (initial contact to the next ipsilateral initial contact), so burst timing is reported as a percentage of the cycle.

References

Winter DA (2009). "Biomechanics and Motor Control of Human Movement." 4th ed. John Wiley & Sons. Power-phase nomenclature A1-A2 / H1-H3 / K1-K4.

Examples

t <- seq(0, 1, length.out = 101)
ankle_power <- ifelse(t < 0.4, -50 * sin(pi * t / 0.4),
                      120 * sin(pi * (t - 0.4) / 0.2) * (t < 0.6))
labelPowerBursts(ankle_power, sampling_rate = 100, joint = "ankle")
#>   label joint       type start_pct end_pct peak_power      work
#> 1    A1 ankle absorption         1      39        -50 -12.72585
#> 2    A2 ankle generation        41      59        120  15.24745