Resumption requires explicit manual confirmation
Aliases: manual reset · restart confirmation · resume interlock
What it is
Manual confirmation before resuming from a stop requires a dedicated reset by someone with adequate knowledge of the scene after the hazard has been checked and cleared. Reset only permits exit from the latched stop state; it is not task start. Releasing an actuator, restoring a network, or clearing an alarm must not resume motion automatically.
Why it happens
At the moment of a stop, the site often holds information the system never sensed: someone is still freeing a jammed part, a hand has not fully cleared the zone, or the alarm condition looks resolved on the surface while its root cause is not. If resumption is left to a sensor deciding "the obstruction is gone," the system can only verify the conditions it can measure — it cannot verify whether a person is actually clear or whether the task at hand is finished. Manual confirmation hands the judgment of whether it is safe to continue back to the person with full situational awareness, rather than to an assumption outside the sensor's coverage.
Studying it
Simulated stop scenarios can compare automatic resumption against a design that requires a confirming keypress, tracking operators' hand position, gaze target, and whether they check the site before resuming. This can reveal anticipatory behavior — stepping back into position early because resumption is expected to happen automatically. Log analysis can compare reset intervals against whether an on-site check preceded the reset, as evidence of whether the confirmation step is actually exercised or has become a rubber-stamp button.
Where it stops holding
Manual confirmation guards against blind resumption from sensor coverage gaps, but if the confirming action is required too often — for every minor disturbance, through the same full reset procedure — operators learn to press confirm without checking, and the step degrades into noise. Placement of the confirmation control also matters: a reset button positioned inside the hazard zone re-exposes the confirming person to a space that has not yet been verified safe.
Applying it
- Make actuator release, stop-state reset, and task start three distinct transitions; no transition should trigger the next.
- Before reset, show the trigger source, remaining faults, affected axes, and interrupted task. Place confirmation where the hazard area is visible or use a validated equivalent clearance process.
- Track reset intervals against whether an on-site check preceded them; if operators confirm quickly and without pause even after the information is shown, redesign the flow rather than adding more warning text.