Mechanism
The mechanism addresses a governed unit confronted with a physical configuration in which no available actuation path avoids harm. Rather than treating that case as undefined, the harm-minimization deviation mechanism selects the actuation path that minimizes composite projected harm across all entities within the spatial region. A candidate-path generator produces a plurality of candidate actuation paths available to the unit given its current kinematic state, its capability envelope, and its observed environment. A harm projector projects the composite expected harm across entities over each candidate path, consuming entity-class harm coefficients per a governance-policy-defined entity-class harm ordering and consuming an empathy-weighting output of the integrity engine.
A composite-admissibility evaluator combines the per-path projected harm with per-path composite-admissibility determinations into a composite harm-admissibility score per path. A path selector selects the candidate path whose composite harm-admissibility score is the most favorable. A deviation-lineage recorder records the candidate-path set, the harm projections per path, the governance-policy-defined parameters applied, the selected path, and the actuation execution. The harm-minimization deviation evaluates self-damaging actuation paths as admissible candidates when they minimize composite harm, whereas prior architectures exclude self-damaging paths categorically.
The entity-class harm coefficient is a governance-policy-defined parameter ranking harm to entity classes in a governance-policy-configurable ordering. A non-limiting exemplary ordering assigns the highest harm coefficient to human life and bodily integrity, a lower coefficient to animal life, a lower coefficient to physical property, a yet lower coefficient to the governed unit itself, and the lowest coefficient to designated replaceable environmental fixtures including guardrails, shoulder berms, crumple zones, and emergency barriers. The ordering is configurable by the deploying authority per jurisdictional regulatory framework, per deployment-domain ethical framework, per organizational policy, and per operational context. The specific ordering is immaterial to the primitive; the primitive is the composite harm-admissibility evaluation across a governance-policy-defined entity-class harm ordering combined with the composite-admissibility evaluator.
Configurable Ordering by Deployment
The ordering is configured to the deployment rather than fixed by the system designer. In a non-military consumer-roadway deployment, the example ordering applies. In a military defense deployment, the ordering additionally differentiates combatant from non-combatant in accordance with applicable laws of armed conflict. In an industrial deployment, the ordering additionally differentiates authorized personnel from unauthorized entrants.
The ordering reaches the unit as a governance-policy update carried as a governed observation. The governed observation carries the policy version, the policy content, the deploying authority's signature, and an applicability-scope specification identifying the devices eligible to receive the policy update. At each receiving device the governed observation is evaluated through the composite admissibility evaluator, which verifies the authority credential of the deploying authority, the evidential weight assigned to that authority per the receiving device's authority taxonomy, and the consistency of the update with the device's prior policy state. The receiving device atomically transitions from the prior policy state to the updated policy state, with a rollback mechanism enabling reversion upon detection of admission failure or upon receipt of a subsequent revocation governed observation, and the policy-update event is recorded in the device's lineage field.
Coupling With Graduated-Actuation Modes
The harm-minimization deviation mechanism integrates with the graduated-actuation mode selector. A high-confidence harm-minimization deviation proceeds in full mode, in which the actuator physically executes the proposed actuation. A lower-confidence deviation proceeds in constrained mode, in which the actuator executes subject to one or more constraints. An uncertain harm-minimization scenario proceeds in consultative mode when a human operator or higher-authority agent is reachable within the decision horizon, in which the actuator emits a governance-credentialed consultation rather than acting unilaterally. A harm-minimization path exceeding per-path governance-policy-defined risk thresholds proceeds in disabled mode, in which the actuator does not execute the proposed actuation, the non-execution is lineage-recorded, and the agent's cognitive state is transitioned to a confidence-degraded operating mode.
Because disabled mode is itself a selectable governed outcome, the architecture permits the unit to decline to actuate when no path clears its governance-policy-defined risk thresholds, with the declination recorded in lineage rather than discarded.
Representative Deviations
The deviations within the scope of the disclosure include, without limitation, a ground vehicle intentionally steering into a guardrail, a shoulder berm, a vacant area, or a designated crumple zone to avoid contact with a pedestrian, cyclist, or occupant of another vehicle; an autonomous aircraft executing a controlled off-airport descent into unpopulated terrain to avoid overflight of a populated area during an in-flight emergency; and an autonomous underwater vehicle intentionally grounding in a non-navigable zone or executing controlled disabling to avoid collision with a submerged cable, a pipeline, a coral reef, or a protected habitat.
Further deviations include an industrial robot executing a controlled self-disabling arm retraction, arm collapse, or emergency-stop sequence to avoid contact with a human operator entering the work envelope; a medical infusion device aborting or reducing a dose upon projection that the administered interaction harm exceeds the non-administration harm; a weapon-system terminal actuator redirecting a munition into a less-harmful terminal location upon identification of non-combatants in the originally-targeted area or upon re-identification of the target as non-hostile; an autonomous delivery robot executing a controlled stop into a non-traversable zone rather than continuing onto an occupied pedestrian path; a building ventilation actuator accepting reduced ventilation in unoccupied zones to extend survivable-atmosphere duration in occupied zones during fire emergencies; and a grid-protection actuator executing controlled load-shed in a non-critical region to preserve supply to a critical-infrastructure region.
Prior-Art Distinctions
The harm-minimization deviation mechanism is structurally distinguished from prior collision-avoidance systems, prior pre-collision braking systems, prior emergency-steering systems, and prior academic discussions of trolley-problem ethics by the simultaneous presence of several properties. It uses a governance-policy-defined entity-class harm ordering rather than implicit per-system heuristics. It performs composite admissibility evaluation across cognitive primitives including the integrity engine's empathy weighting.
It contemplates self-damaging actuation paths as admissible candidates rather than excluding them categorically. It couples to the graduated-actuation mode selector. It provides lineage-recorded reconstructibility of each deviation decision. And it achieves cross-domain applicability through the same architectural chain across automotive, aerial, subsurface, industrial, medical, building, energy, and defense domains.
Disclosure Scope
This disclosure forms part of U.S. Provisional Application No. 64/049,409. The disclosure covers the harm-minimization deviation mechanism; the candidate-path generator, harm projector, composite-admissibility evaluator, path selector, and deviation-lineage recorder; the governance-policy-defined, configurable entity-class harm ordering and its example coefficient assignment; the configuration of the ordering per jurisdictional, ethical, organizational, and operational context; the integration with the graduated-actuation mode selector including full, constrained, consultative, and disabled modes; and the representative deviations enumerated across the automotive, aerial, subsurface, industrial, medical, building, energy, and defense domains.
The provisional, 64/049,409, situates this mechanism among the governed-actuation cluster alongside graduated-actuation mode selection, preemption-budget semantics, reversibility-aware commitment-point evaluation, and post-actuation verification. The specific ordering is immaterial to the primitive; the primitive is the composite harm-admissibility evaluation across a governance-policy-defined entity-class harm ordering combined with the composite-admissibility evaluator.
