Skip to content
Talk to an engineer

RoboticsProve one before you buy ten.

Most hardware projects are sold as a device and turn out to be everything around it: what it has to sense, what it moves or controls, which system it reports to, and what a person does when it stops. This is the work of deciding whether a task is worth building for, designing the sensors, actuators and controllers that do it, connecting them to the software you already run, and proving one working version on site before anything is rolled out. We specify, integrate and program the hardware; we do not manufacture it.

Talk about thisSee work using this

Commitments

One working version on site before any rollout
Proven
Every unit, against what it was built to do
Justified
The design, the programs and the configuration
Yours

Robotics

We design custom hardware that automates a physical action. We study the task and the space first, then build one working version and prove it on site before anything is rolled out.

Each design brings together three kinds of instrumentation:

Sensors
Measuring what is happening in a space, on a machine or in a process
Actuators
The motors, valves and grippers that carry out the action
Controllers and processing
The logic that reads the sensors and decides what the actuators do

We choose proven parts wherever they fit and connect the result to the systems you already run. We specify, integrate and program hardware; we do not manufacture it.

When would I want this?

  • You want to automate a physical action
  • You want to collect data about your environment
  • You want to deploy physical monitoring systems
  • You want an interactive system your customers use in person
  • You want to create a custom smart product

Sounds like

You might recognize one of these.

  • We are paying people to do the same lift, pick or scan hundreds of times a shift.

  • We want to know what is happening in the building without sending someone to look.

  • The machines on the floor have data on them and none of it reaches anything we can query.

  • We have an idea for a product with a sensor in it and nobody to build the first one.

  • The pilot worked. Now we need forty of them and nobody knows how.

What you get

What lands on your side, and stays there.

  • A feasibility study naming the task, what it costs today, and what building for it would take
  • A bill of materials with the reasoning for each choice and the alternatives rejected
  • One working version on site, measured against what it was built to do
  • Operator screens and procedures the people running it can follow
  • Integration with the systems the hardware reports to
  • Documentation, a spares list and an update process for the rollout

Shapes

How this usually runs.

  1. Feasibility and automation plan

    2–3 weeks

    What the task is, what it costs today, what building for it would take and whether the answer justifies the build. This phase can end in a recommendation to build less than was asked for.

  2. Pilot build

    6–12 weeks

    One working version built end to end and running on site: the sensors, actuators, controllers, integration and operator screens, measured against what it was built to do.

  3. Rollout

    Ongoing

    The remaining units, with updates, spares and fault handling run as a process rather than as a series of favors.

Proof

Where we’ve done this.

All case studies

What this includes

The work, specifically.

Not every engagement needs all of it. This is the range we cover and what each part is actually for.

  • Task and feasibility study

    What the task is, how often it happens, what it costs today and what building for it would take. The most useful answer this step gives is sometimes that the task needs no new hardware, or that one sensor on existing equipment answers most of it.

  • Sensors

    Measuring what is happening in a space, on a machine or in a process, with each sensor placed and calibrated so a reading means what it says. A correctly calibrated sensor in the wrong position produces confidently wrong data forever.

  • Actuators

    The motors, valves and grippers that carry out the action, chosen from proven parts wherever they fit. When nothing off the shelf does the job, we design the part that does and hand the production package to a contract manufacturer; we do not manufacture.

  • Controllers and integration

    The logic that reads the sensors and decides what the actuators do, on a controller or an edge processor sized for the job. It connects to the equipment around it and the software above it, whether that equipment speaks Modbus, OPC UA or a serial protocol from 1997, and it knows what to do when the network is down.

  • Operator screens

    Interfaces for the person who runs the hardware or stands in front of it, often busy and sometimes a customer: start it, stop it, clear a fault and see why it stopped, in as few steps as the job allows.

  • Safety and handover

    Guarding, stops and a written risk assessment before anything that moves runs with people nearby, and the documentation, spares and update process your team needs to keep it running after we leave.

Tooling

What we build it with.

No tool here was picked because it was new. Where we do reach for something novel, it is in one place, for a stated reason, and it is written down.

Sensing
  • Machine vision
  • Load cells
  • Current transformers
  • Thermocouples and RTDs
Actuation
  • Motors and motor control
  • Valves
  • Grippers and end effectors
  • Linear actuators
Control and integration
  • Microcontrollers
  • Linux on ARM
  • Modbus TCP and RTU
  • OPC UA
  • MQTT

Questions

Robotics, honestly.

  • No. We choose proven sensors, actuators and controllers, integrate them, and write the programs and screens that run them. When nothing off the shelf does the job, we design the part that does and hand the production package to a contract manufacturer.

  • That is what the feasibility study answers, before you buy anything. It compares what the task costs today with what the hardware would cost to build and run, and it can end in a recommendation to build less than was asked for.

  • Usually some of it, and it is the first thing we check. A machine that already reports its state, or a conveyor that already has a sensor on it, is cheaper to connect than to replace.

  • You do: the study, the bill of materials, the programs and the configuration, on the same terms as every other engagement here.

Further reading

What we think about this, at length.

All posts

Often paired with

Next step

Tell us what’s breaking.

Forty-five minutes, no charge, no deck. We’ll tell you what we’d do, what it would likely cost, and whether the problem is in the hardware at all.

Talk to an engineerSee how we work
Reply
A person replies, not a sequence: within one business day, from someone who would be on the engagement.