Skip to content
Koidra
Book a demo

Autonomous control

KoPilot

Closed-loop control of the process, minute by minute, with no human at the panel.

How it works

The closed loop, and the envelope it obeys The process is read by a physics-aware model, whose proposed setpoints pass through an operating envelope that states the limits it may not cross, then reach the controllers, which write them back to the process. The loop runs every minute. The process dryer, boiler, hall, crop Physics-aware model knows what the equipment can and cannot do Operating envelope the limits it may not cross Controllers setpoints written back every minute, not every shift
The envelope is a separate box on purpose. It states what the loop may not do, and it is engineered for the site rather than learned from it, which is why a model that has never seen your plant is still safe to run on it.

Everything on this page is also reachable over our API, including from the AI tools your team already uses, so Claude or ChatGPT can query the same record directly. We send the reference on request.

What it takes over from

  • Setpoints tuned by hand, differently on every shift
  • Comfort margins nobody has revisited

Pays for itself by

Energy per tonne, yield, and consistency that no longer depend on who is on shift.

The number a buyer checks: Energy cost per unit of output, and output variance between shifts.

Proof

2×
Winner, Autonomous Greenhouse Challenge 2018 and 2022, a live crop each time
17%
Higher net profit than the expert growers 2018 competition, with 6% higher yield
19.5%
Yield increase on cucumber against a control zone grown manually alongside it
17%
Energy savings in the same commercial trial

A control zone grown alongside separates the effect of the strategy from the season, the weather and the crop. These results are why we say autonomous control: KoPilot does not only recommend a setpoint, it calculates and applies control decisions continuously.

What KoPilot does

KoPilot continuously adjusts process setpoints to keep operations close to the desired operating point. It reads real-time plant data, determines the appropriate setpoints, and sends those setpoints back to the existing control system.

Instead of waiting for an operator to review conditions and make the next adjustment, KoPilot responds continuously as conditions change.

How KoPilot runs a greenhousePlant and climate sensors in the greenhouse feed the climate computer, which sends the greenhouse data to KoPilot. KoPilot reads the weather forecast directly, and the grower adds crop registration, the strategy and the limits. KoPilot decides the growing conditions round the clock and sends setpoints back to the climate computer, which moves the screens, vents, heating, lighting and CO2.greenhouse datadataforecastcrop registration, strategyand limitssetpointsPlant and climate sensorsWeather forecastGrowerClimate computerKoPilot, round the clockScreens, vents, heating, lighting, CO2
The loop in a greenhouse, where KoPilot started. The climate computer stays in charge of the equipment; KoPilot decides what it should be holding, and the grower sets the strategy and the limits.

Physics-informed control

KoPilot combines process data with models of how the physical system behaves, so a control decision stays within what the equipment can actually do rather than only within patterns seen in historical data.

How quickly equipment can heat or cool
A dryer cannot heat faster than its mass allows, so the model never proposes it.
How humidity responds to ventilation
Opening a vent changes humidity and temperature together, and the model knows by how much.
How one control action affects conditions later
A move that looks good this minute is rejected when it costs more over the next four hours.

Historical data remains useful, but autonomous control also has to behave sensibly when the process enters conditions it has rarely seen before.

Why the model is built from physicsA decision is checked against how the equipment behaves, how fast it heats or cools, how humidity answers ventilation, what one action does later, so it stays inside what the plant can do, even in conditions the history never showed.Process datawhat is happening nowKoPilot's modelPhysics of the planthow fast it heats or cools,how humidity answers ventilation,what an action does laterA setpoint the equipment can reacheven in conditions never seen before
The physics is the guard rail. It is what lets the model be trusted in a week the history never showed.

Start with the right foundation

Closed-loop control is not appropriate for every process on day one.

It depends on the quality and coverage of existing instrumentation, the reliability of the underlying control system, how well the process can be modeled, the operating constraints that must be respected, and whether a supervised trial can be run safely before autonomous operation.

Most customers start with Maintenance Management or DataPilot, and add AI Advisory before moving to closed-loop control. Those products create the data, operational history and process understanding needed to make autonomous control more reliable.

Tell us about your process and we can assess where autonomous control is practical, what should come first, and how a supervised deployment could be structured.

What has to be true before closed-loop controlInstrumentation coverage, a reliable control system, a process that can be modelled, defined constraints and a safe supervised trial. Most customers build these with Maintenance Management or DataPilot, then AI Advisory, before KoPilot.Maintenance Management or DataPilotthen AI AdvisoryFive things to have firstinstrumentation that covers the processa reliable control systema process that can be modelleddefined operating constraintsa place for a supervised trialSupervised trialKoPilot, closed loop
The products before KoPilot are not a queue. They are where the data and the process understanding that make a closed loop safe come from.

Where this sits

Two ways into one platform, a layer that needs both, and control on top. Each is a complete product with its own payback. How the pieces fit.

  1. Start here

    Maintenance Management

    The system decides what maintenance is due, dispatches it, and tracks what it cost.

  2. Start here

    DataPilot

    Every signal the plant produces, in one place, live and historical.

  3. Comes with both

    AI Advisory

    Reports on what changed, answers in plain language, what to change to move the KPIs you set, and predictive maintenance: failures seen before they arrive.

  4. Autonomous control

    KoPilot

    Closed-loop control of the process, minute by minute, with no human at the panel.

See Koidra on your own plant data

Bring a week of readings from one line. We will show you what the platform sees, what it would have flagged, and what it would have scheduled.

Book a demo