calibrate spike sensors and analyze logs for pybricks competition robots. use when the user wants black and white thresholds, color detection, distance cutoffs, imu-based headings, or interpretation of reflection and timing data from repeated tests.
インストール
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
calibrate spike sensors and analyze logs for pybricks competition robots. use when the user wants black and white thresholds, color detection, distance cutoffs, imu-based headings, or interpretation of reflection and timing data from repeated tests.
Sensor Calibration Logger
Overview
Turn raw sensor samples into usable thresholds, noise margins, and logging plans that survive real lighting and field conditions.
Workflow
Identify the sensor and event context.
color sensor reflection or color detection.
distance trigger.
force trigger.
heading or orientation using hub sensors.
Ask for samples, not guesses.
at least 5 black and 5 white reflection samples for line work.
multiple measurements at match-like distance and lighting.
If the user provides labeled reflection csv, run scripts/reflection_stats.py.
Recommend the smallest robust trigger rule.
threshold.
hysteresis or deadband if needed.
recheck or confirmation sample if noise is high.
Return the result in this order.
observed ranges.
recommended thresholds.
logging plan for next test.
what to recalibrate on event day.
Rules
Never recommend a threshold from one black and one white sample.
Mention lighting sensitivity whenever reflection ranges overlap more than expected.
Prefer reflection over named color for line following unless the task truly depends on color classes.
Encourage event-day spot checks on the real table or mat.
Resources
Use references/logging-patterns.md for sample collection and interpretation.
Use scripts/reflection_stats.py on labeled csv data.
Use assets/reflection_log.csv as the default csv format.
