Requirements Analysis & Specification, Pick & Place
Requirements analysis and functional specification for a pick-and-place cell, the scoping step that decides whether a robot is the right answer at all.
View detailsPick & Place
Perception, localisation, navigation and picking software for handling tasks.
What to buy first, and what you can skip
These six items are the parts of a pick-and-place deployment, priced separately because not every project needs all of them. Bought in the wrong order they overlap and you pay twice.
- Start with the requirements analysis.
- It is the cheapest item here and the only one that can tell you not to buy the others. It produces the written scope, the cycle-time target and the acceptance criteria, and without those the later work has nothing to be measured against. Teams that skip it usually end up paying for a second round of perception work because the tolerance was never agreed.
- Mapping and perception come next
- , and which one you need depends on what moves. If the parts arrive in a fixed fixture and the robot stays put, you may need neither: a taught position is cheaper and more repeatable than vision. If the robot has to travel, you need the map. If the part can be anywhere in a bin or on a belt, you need perception. Both, only if both are true.
- The navigation stack is only relevant to a mobile platform.
- A robot bolted to a bench does not need it, and it is the item most often bought unnecessarily because it sounds fundamental.
- The picking and drop-off process is where most of the value sits
- , and it is the item to spend on. Grasp selection, approach and retreat, placement tolerance and error handling are what decide whether the cell runs unattended or needs someone standing next to it.
- Order management
- is for when the cell has to take jobs from a system rather than a person. If a queue on a screen is enough, leave it out until it is not.
All of it is charged by the hour or by the engineering day, so the scope is yours to control. We would rather size it honestly and be asked for more later than pad it and be argued down.
How the cell gets measured, and what counts as done
Almost every argument about a pick-and-place cell traces back to nobody having written down what success meant. These are the numbers to agree before the engineering starts, because afterwards they become a negotiation.
- Cycle time, defined end to end.
- A pick takes a certain number of seconds, but the figure that matters includes the approach, the grip, the transfer, the placement, the retreat and the wait for whatever comes next. Measure the current process the same way, by hand and with a stopwatch, before comparing. A cell that is faster per pick and slower per part is a common and expensive surprise.
- Success rate, with the failure modes separated.
- A cell that picks 98 of 100 is excellent if the two failures are detected and set aside, and unacceptable if they are dropped on the floor or placed wrong without anyone noticing. Agree what the machine does when it is not sure: retry, skip and flag, or stop. Those three are different products and cost different amounts.
- The part population, written down.
- Which parts, in what condition, arriving how. A cell taught on clean parts in a fixture will not hold on the same parts covered in coolant, and a cell that handles the twelve current variants may not handle the thirteenth. If new variants appear every quarter, that is a requirement rather than a surprise, and it changes the design.
- Uptime and who intervenes.
- Over a shift, how many operator interventions are acceptable and what each of them is allowed to take. An unattended cell and a supervised cell are separated by a small amount of hardware and a large amount of error handling, and error handling is where the engineering days actually go.
- The acceptance test, agreed in writing before anything is ordered.
- A stated number of parts from your own production, run in your own building, on a day both sides attend, against the numbers above. It is the cheapest insurance either side can buy: it tells you when to pay, and it tells us when we are finished.
6 products
Order Management (hourly rate)
Order-management software for robot pick-and-place: it takes jobs from your system, sequences them and hands the robot a queue it can actually execute.
View details
Mapping (hourly rate)
Mapping of the working area a robot will operate in, delivered as the map layer that the navigation stack and the picking process both depend on.
View details
Perception & Localisation
Perception and localisation for humanoid and mobile platforms: recognise the part, know where the robot is, and keep both stable in a live environment.
View details
Navigation Stack
ROS-based navigation for humanoid and legged platforms, path planning, obstacle handling and recovery behaviour tuned to your floor, not a demo hall.
View detailsPicking & Dropping Off Process (hourly rate)
Application engineering for the pick and drop-off cycle itself: grasp selection, approach, placement tolerance and the error handling around them.
View detailsFrequently asked questions
Contact us nowYes. Most platforms can be demonstrated on a working unit, and several can be rented for a trial on your own site.
Every supply includes commissioning, operator briefing and European service handling. Engineering support is scoped per project.
It depends on the platform and configuration. Ask us for the current lead time on a specific model.