Who are Gardasoft and what do they do?
Gardasoft specializes in accurate control technology for the machine vision market.
We're very well known for our top-quality LED controllers. But we also design and manufacture some very high-performance strobe lights which are used for traffic and infrastructure management.
How do your controllers fit into the machine vision industry?
Our controllers were originally developed specifically for machine vision applications, and as the machine vision industry has grown, Gardasoft has grown with it.
In machine vision, you always need good, consistent lighting. You might be running quality control measurements, for example, or tracking items through production, but you’ll always need a good image to get the best results from your software. Good lighting is absolutely key to getting useful, high-quality, and consistent images from your setup.
There are a number of lighting suppliers on the market, and you do have various choices on how to control the light. You could, for example, use a very basic fixed voltage supply. However, the problem is that the measurement results could change with even small alterations in the light intensity. That’s how Gardasoft can help you.
We produce lighting controllers that will supply very accurate drive for your lights, and because it’s based on current, rather than voltage, it will be much more dependable. The brightness of LEDs depends on the current through them, and not the voltage drive, and LEDs are very sensitive to tiny changes in current.
Many people still think that the easiest way to drive a light is to supply it with a fixed voltage. If you ever look at a graph of voltage against brightness for an LED, you’ll see that the brightness varies massively with voltage. If the voltage changes just a little bit, the current through the device will change a lot, and therefore the intensity will change a lot. This means that the only really good way of controlling an LED is to use an accurate current source, such as a Gardasoft controller.
A current source is harder to design, so if people are constructing the system themselves, it's much easier to use our expertise in that area. We specialize in current controllers to ensure very stable lighting, which will give you reliable, repeatable images. Then you can also be sure that your quality control will also be good.
What does Gardasoft supply in hardware and software?
Our controllers are hybrid devices and make use of both hardware and software to accomplish the functionality. We design them in a way that means that we can customize them if need be. So if a customer has a very specific application that needs something a bit different, we can do that quickly for them.
We also have devices, such as the CC320 and the RTCC4, that have inbuilt sequencing control. It's like a channel sequencing device, but the best way to think of it is that it's like a PLC. But, PLCs require a lot of expert knowledge to use. You need to understand concepts such as ladder logic, and the programming to use them.
On the other hand, our CC320 and RTCC4 have a web interface so you can easily set up complicated machine vision sequences, such as tracking defect products down production lines and kicking them off a conveyor belt if they fail an inspection. The controller can communicate with a machine vision system, which can tell it whether something's failed or not. You can use it to trigger lights and cameras through the production line.
It can also trigger those lights and cameras on the production line so that they track the speed of conveyor belt, and we've got an encoder input to do that. The CC320 is a great example of us taking our customers’ problems and finding a way to fix the problem more easily. The main aim is to make vision easy for people to set up and use.
Could it be described as a 'plug-and-play' system - i.e., it’s able to have different scripting languages in one area?
That's our aim! We want to make it incredibly easy for people to set up a lighting or a machine vision system.
One way in which we've done that through creating a technology that we call triniti. Triniti includes a chip that lives in the lighting manufacturers' light so that when you plug the light into one of our controllers, the controller knows exactly how it should be driven and how hard it could be driven. We also store information about the light, such as how many hours it's been on, and you can give the light a name. That brings it into line with manufacturing 4.0 techniques.
The other way that we're trying to make things plug and play is to make it easier for people to create automation using our products. We have a software development kit (SDK), so you can write software for our products in C++, C#, and VB. We've done all the hard work for you, so rather than having to write a script to change the brightness of our lighting controller, you can call a function in one of those languages that will do it for you.
We're trying to make it easy for machine vision programmers to write their code. We're also GenICam compliant. GenICam is a machine vision protocol that machine vision devices use to speak to each other. We're compliant to that standard, and in the future, we want to use technology such as ‘Harvester’ to create multi-platform open source code so that people can drive our controllers more easily.
How will your technology be used in the future?
In the future people will want manufacturing lines that produce more than one item. A good example of that is in car manufacturing. Traditionally you might have one production line producing car bonnets, another line producing body shells and another one doing car doors. Increasingly, you want to have a single production line that can make all those things and be able to adapt inspections to the product being produced at the time.
Using our software tools, it’s possible to create programs to automatically adapt the lighting and other machine vision control sequences to each different component.
How do you view the future of autonomous solutions for machine vision?
Users certainly need things that are easy to set up and will “just work”. Development time is expensive, so anything that we can do to make things plug-and-play the better.
Items don't necessarily need to be in a single box with everything inside them. Devices often need to be in different positions in the factory, so getting them to all talk to each other is key, and this is all about GigE Vision. GigE Vision is an Ethernet standard, which many machine vision components use to communicate. So it's all about applying machine vision standards and being part of the community that drives those standards.
Do your controllers allow customization or do you just have out of the box applications?
No, I'd say the key thing about Gardasoft is our flexibility, we work very closely with our customers, and we're always trying to find solutions to their problems.
We have a huge amount of customer content, and people often come to us so we can help them overcome their problems. Sometimes we can solve it with our existing products because they have a lot of inbuilt versatility. Other times, customers may need a custom feature, and as long as the volumes justify the development work, we're very happy to create custom solutions.