The robot is a fleet service wrapped around a physical body
Actuators and sensors create a capability envelope; manipulation policies turn contact into action; recovery protects uptime; technicians, spares, teleoperation, and workflow integration turn machines into dependable capacity. A demo usually shows the successful action and hides the reset queue.
Three numbers that locate the frontier
Installed units measure specialized automation, not task generality. The hourly example is derived and excludes finance, integration, maintenance, energy, downtime, tooling, and supervision.
Accepted task-hours per human intervention is the curve
A useful robot-hour counts only work completed inside quality and cycle-time limits. Generality improves when task coverage rises faster than integration effort and when one person can support a larger fleet without building a synchronous teleoperation workforce.
A 99% step success rate can still fail a long sequence frequently; autonomous recovery matters as much as nominal task success.
More degrees of freedom widen capability while adding mass, calibration, control burden, collision modes, wiring, and wear.
The headline metric sits on a system
Each layer can become the bottleneck even when the layer before it improves.
Body and actuation
Structure, joints, transmissions, power, mobility, and interchangeable end effectors define force, reach, speed, and service life.
- Measure
- Payload · lifetime cycles · Wh/h
- Failure mode
- Wear, heat, and maintenance
Perception and contact
Vision, force, tactile sensing, and proprioception estimate objects, constraints, slip, and hidden contact state.
- Measure
- Uncertainty · sensing bandwidth
- Failure mode
- Occlusion and ambiguous contact
Policy, dexterity, and recovery
Planning and learned behavior sequence compliant actions, detect failure, retry, regrip, or safely hand off.
- Measure
- Accepted sequences/intervention
- Failure mode
- Long-tail states
Safety and fleet operations
Validation, scheduling, teleoperation, spares, software updates, technicians, and workflow integration create delivered labor.
- Measure
- Availability · accepted hours/human-hour
- Failure mode
- Support and certification
A robot must move itself before it moves the work
Force, speed, reach, stiffness, battery energy, and contact information create unavoidable trade-offs. The useful floor is the least machine and supervision needed to complete a duty cycle safely for enough lifetime hours—not the lowest hardware bill.
Better policies expose hands, recovery, and service
As models broaden scene coverage, manipulation reliability, cable routing, connectors, contamination, impacts, charging, actuator life, autonomous resets, and field repair become the binding constraints.
Publish intervention ledgers
Count human minutes, resets, damaged work, and failure causes alongside task success.
Design for recovery
Use compliant motion, tactile feedback, safe retreat, retry, and asynchronous escalation.
Reuse task packages
Portable behaviors, tools, and safety cases must reduce engineering per deployment.
Pool fleet service
Standard modules, spares, remote expertise, and predictive maintenance amortize support.
An optimistic view, with conditions
General-purpose robotics becomes a service-level agreement
The credible product is a managed fleet quoting accepted throughput, availability, task-change cost, intervention burden, and repair response—not a humanoid silhouette with a catalog of demonstrations.
Sources, method, and boundaries
IFR deployment figures establish the industrial baseline. NIST performance and manipulation frameworks define comparable dimensions. Demonstrations are treated as capability evidence, not fleet economics.
