STOKA
ESEN中文
Contact usJoin STOKA

STOKA / Systems

Moving the load without a person pushing it

Moving loads between distant points of the warehouse is the stage that adds least value and consumes most hours: somebody walks a hundred metres with a forklift or a pallet truck to drop off one pallet and comes back empty. Autonomous vehicles make that journey on their own. STOKA designs, imports, installs, integrates and supports both families: the automated guided vehicle, which follows a fixed path, and the autonomous mobile robot, which navigates under its own judgement. Which one suits you depends on how much your warehouse changes, not on a catalogue.

Tell us your problem

How many hours per shift go into carrying loads from one point to another?

More than any dashboard shows. Moving goods neither transforms nor prepares them: it only changes where they sit. A warehouse operative costs the company USD 965 per month, USD 11,580 a year (CCT 40/89 logistics branch, June 2026), and social contributions account for 34.6% of labour cost (IARAF, February 2026).

What an AGV is, and what an AMR is

In plain terms: these are vehicles that carry loads from one point of the warehouse to another with nobody driving them. There are two families, and the difference matters. The AGV, automated guided vehicle, follows a fixed path marked on the floor or in the building: a magnetic tape stuck to the floor, a rail, or a laser that orients itself against reflectors mounted on the columns. It always runs the same route, it is predictable and it is the cheaper option per unit; in exchange, changing the route is building work. The AMR, autonomous mobile robot, builds its own map of the warehouse, locates itself within it, avoids whatever crosses its path and recalculates the route as it goes. It costs more per unit, and the route is changed in software, without touching the floor.

AGV: the path is marked once

The automated guided vehicle runs on a physical reference installed in the building. It can be a magnetic tape stuck to the floor, an embedded wire, a rail, or a rotating laser that takes its position against reflectors mounted on columns and walls.

Keep reading

The vehicle follows that reference and always runs the same route, with the same stops. The rigidity is the virtue: behaviour is predictable, commissioning is straightforward to audit, and cost per unit is lower than that of an autonomous vehicle. Where the route was settled years ago and never touched, from the end of the production line to the despatch area, from the dock to the goods-in floor, it does that job every day without surprises. The limit appears when the route changes: moving the path means lifting the tape or relocating reflectors and recalibrating. That is building work, with the warehouse disrupted, not a software setting.

AMR: the route is changed in software

The autonomous mobile robot needs no mark on the floor. It travels the warehouse once, builds a map with its sensors and then locates itself in that map by comparing what it sees against what it has stored.

Keep reading

With that it goes from one point to another by whichever route suits, stops if a person appears, drives around a badly placed pallet and carries on. When the layout changes, the map is updated and that is that: the new path is a setting, not a building job. Cost per unit is higher than for a fixed-path vehicle, and that difference pays for itself in warehouses where the layout moves, where building work is under way, or where the vehicle has to share an aisle with people and forklifts. It also suits a project that starts small and will grow, because adding units does not force you to redo the floor infrastructure. In exchange it demands more of the environment: lighting, stable references and a floor kept in good condition.

What the warehouse itself has to provide

An autonomous vehicle is not plug and play. First, the floor: joints, level changes, damage and ramps condition speed and even the viability of the project, so flatness and permissible load are verified on site during the survey.

Keep reading

Second, signage and order: clear traffic zones, marked crossings, defined pick-up and drop-off points, and a firm rule about where goods may be set down and where they may not, because a pallet left in the wrong place is, to the vehicle, a wall. Third, connectivity: the wireless network has to cover the whole route with no blind spots. And fourth, the software. One vehicle needs only simple task logic; several vehicles mean somebody has to decide who goes first at a crossing, who charges and when, and how orders are shared out. That is a traffic manager, integrated with your warehouse management system (WMS).

When each one suits

The useful question is not which is better, but how much your warehouse moves. If routes are stable and repetitive, if the stops are always the same and the layout has not changed in years, the fixed-path vehicle does that job with less investment per unit and fewer things that can go wrong.

Keep reading

If the layout is rearranged by season, if building work is under way, if aisles are shared with people and forklifts, or if the project will grow in stages, the autonomous mobile robot is the answer, because every change is settled in software. There is a third scenario, and it is common: both together. The trunk route, long and always the same, on fixed-path vehicles; the variable legs on autonomous ones. The decision comes out of mapping a month of real movements: origin, destination, frequency, time of day and load unit. That map is drawn during the survey, and it is what then organises the investment.

What decides whether an autonomous vehicle suits, and which one?

We do not publish model-by-model specification tables: the detailed sheet comes with the assessment, sized against your operation. These are the criteria the choice is made on.

Distance and repetitiveness of the route
A vehicle is justified when the journey is long and repeats many times per shift. Short legs, which the operative covers on the way while completing another task, almost never return the investment.
Stability of the layout
If the route has not changed in years, a fixed path is enough and costs less per unit. If the warehouse is rearranged by season, by building work or by growth, choose a vehicle whose route changes in software.
Volume of movements per shift
This settles how many units are needed and whether the project adds up. Count real movements, not the average somebody recalls in a meeting: origin, destination, time of day and frequency of every single journey.
Sharing space with people and forklifts
A shared aisle calls for a vehicle that brakes, avoids and carries on. If traffic is dense throughout the shift, review the layout before the vehicle, because no technology fixes a circulation bottleneck.
Floor, network and coordination software
Joints, level changes and ramps condition speed; the wireless network has to cover the whole route; and more than one vehicle calls for a traffic manager integrated with your management system. All of it is verified during the survey.

Engineering for your operation

When is an autonomous vehicle the wrong answer?

Our approach

When distances are short, there is no case. If the typical journey is twenty metres that the operative covers on the way while finishing another task, no vehicle improves it: it adds a wait. Nor does it suit a low volume of movements, because the equipment stands idle for most of the shift and the investment finds nowhere to pay itself back. The third scenario is the most frequent and the least discussed: aisles where traffic from people and forklifts is so dense that the vehicle would spend more time braking than moving. There the problem is not transport but circulation, and it is solved with layout, not with robots. The same applies when disorder is the bottleneck: if locations are undefined, if stock counts do not match, or if every pallet turns up where it should not be, automating the journey only moves the problem faster. Process first; the vehicle afterwards, and only if the numbers add up.

Frequently asked questions

What is the real difference between an AGV and an AMR?

The automated guided vehicle (AGV) follows a fixed path marked on the floor or in the building: it costs less per unit and is highly predictable, but changing the route is building work. The autonomous mobile robot (AMR) navigates on its own, avoids obstacles and recalculates the route; it costs more per unit and the route is changed in software.

Can they run in an aisle shared with people and forklifts?

Yes, and that is precisely the scenario where an autonomous vehicle suits better than a fixed-path one, because it brakes, avoids and carries on. That said, if traffic is dense throughout the shift the vehicle spends more time braking than moving, and the real problem is one of layout and traffic timing.

Does it work if my layout changes every month?

That is the case for the autonomous mobile robot. You update the map and the new path is live without touching the floor. With a fixed-path vehicle, by contrast, every change means lifting the magnetic tape or relocating the reflectors and recalibrating the system, with the warehouse disrupted while the work lasts.

How long does the investment take to pay back?

As an indicative reference, in automation projects we work with a payback of 18 to 36 months, and the real figure depends on movements, shifts and the labour cost of each operation. Installation takes 3 to 4 months from the purchase order. The sums are settled with the survey, not before it.

Other systems

Map the movements before buying the vehicle

The survey counts a month of real journeys, measures the floor and the network, and settles whether a fixed-path vehicle, an autonomous one or a change of layout is what you need. With that we deliver the diagnosis and the quotation.