Robotics & Autonomous SystemsField engagement

An offline score is not a robot. The field is a distribution the simulator never modeled, on a clock the benchmark never measured.

Fractional CTO and lead engineer.
hardware
in the loop
every cycle
inside the deadline
real
distribution, not a benchmark
SensorsPerceptionStateControlActuationthe physical world
Perception feeds control on a clock, and the loop closes through the physical world. It only works if it fits inside the deadline, every cycle.
What was at stake

A robotics system that fails in the field does not produce a worse benchmark. Depending on the application it produces a collision, a stuck robot, a safety incident, or a deployment that cannot be trusted to run unattended, which means it is not deployed at all and the program has spent its budget on something that will not ship.

The constraint

Two realities shape everything and neither appears on a leaderboard. The distribution: the field is full of conditions the simulator never modeled, so a high offline score does not transfer, and the amount by which it fails to transfer is not predictable from the score itself. The clock: perception is bounded by the control loop period every cycle and has to leave headroom for jitter, and a stack that fits the budget on average and exceeds it occasionally does the wrong thing occasionally, which in a closed loop is the whole problem. The two constraints fight each other, because robustness usually costs computation and computation is exactly what the deadline lacks.

perception + inference
control + actuation
jitter headroom
where the design lives
must not be eaten
One control loop period is the whole budget. Perception that does not fit it is not deployable, whatever its offline score.
The fork

The reflex, and the fix.

Road not taken

Optimize offline accuracy

Pull

The standard metric. Improves cleanly with more model and more training, which makes it satisfying to work on.

Why not

A higher score survives neither the gap between simulation and reality nor the loop's deadline.

Road taken

Build for the real environment and its timing

Accepted

More work on robustness, sensor reality, latency, and integration. Possibly a lower offline score.

Bought

A system that holds on the real distribution, inside the loop's deadline.

Decision

Make it work on the robot, in the world, not on the leaderboard.

How it was built
01Sensors
02Perception
03State estimate
04Control
05Actuation
01

Perception and control as one real-time loop

Perception feeds control on a clock, and the loop closes through the physical world. It was built and tested with hardware in the loop, because that is the only place the real failure cases appear. A simulator will not generate the failure a particular sensor produces at a particular temperature, because nobody modeled it.

02

The timing budget allocated before anything is built

The control loop period is divided explicitly across sensing, perception and inference, state estimation, control, and actuation, with headroom reserved for jitter. This is an architectural input, not a performance concern to check at the end, and getting it wrong is not recoverable by optimizing later.

03

Sensor reality as a first-class concern

Calibration, unit-to-unit variation, and degradation over time, and what the system does as the real sensors diverge from the assumption the software was built against. This is a hardware property surfacing as a software failure, which is why it is so consistently nobody's ticket.

How it was measured

A single headline number hides where a system fails. This work was scored on the dimensions that actually decide whether it holds in production, measured on real, held-out cases rather than the demo path.

Behavior on the real distributionLatency within the control-loop deadlineJitter headroom under loadFailure cases that appear only with hardware in the loop
figures

What it produces
Without this discipline

A high benchmark score that does not survive contact with the field or the clock, and that looks like progress right up until deployment.

This system

Perception and action that hold reliably in the field, within the control loop's timing, including the failure cases that only showed up with hardware in the loop.

real distributionreal-timehardware-in-the-loopjitter headroom budgeted
The operating envelope

What it owns, and what it hands to a person.

Handled with confidence
Conditions represented in the real-distribution test set
Operation inside the designed envelope
Timing within the loop deadline
Flagged for review
Novel scenarios
Sensor calibration drift
Out of scope by design
Guarantees outside the tested envelope
The honest limit

It holds on the conditions and failure modes it was built and tested against. A genuinely novel scenario can still defeat perception, it depends on sensor calibration and the operating envelope it was designed for, and the gap between simulation and reality is narrowed rather than closed. Anyone claiming otherwise is describing a system that has not met the field yet.

What it generalizes to

The deciding constraints were a hard timing budget and real sensor behavior, both of which sit below the model and neither of which a perception specialist would have owned. The deadline is an architectural input, not a performance concern to address after the fact. The discipline carries to any system that closes a loop through the physical world, which includes the defense edge autonomy and the hydraulic control work on this site.

How we engage

You have a system like this one.
Tell us where it stands.

Whether it is failing, not yet built, or about to meet a scale it has never seen, we can tell you what we see.

Start a conversation
mostafa@opulion.dev · Response within 24 hours · By inquiry