Actuator & Kinetic Command Safety
Authorization, validation and constraint of commands that produce physical action.
Domain controls
The following identifiers, titles, objectives and implementation expectations reproduce the canonical public PAI-SF™ v1.0 Control Catalogue. Evidence requirements, assurance expectations, dependencies, exclusions and conformance relevance remain available in the full catalogue.
Objective: Ensure kinematic limits (force, torque, speed, position) are enforced independent of the AI decision path. Requirement: Limits enforced at hardware/firmware level, architecturally independent from primary AI compute; configurable only through authenticated process; defaults to most restrictive limits on integrity failure.
Objective: Prevent replayed or timing-manipulated commands from reaching actuators. Requirement: Sequence numbering or timestamping with freshness checks on the command path.
Objective: Detect actuator behavior diverging from commanded behavior (mechanical wear, unexpected resistance) before it becomes unsafe. Requirement: Closed-loop comparison of commanded vs. actual actuator response, flagging discrepancy beyond a defined threshold.