5 questions to ask yourself before developing your own AMR
As an automation solutions provider, you’re always looking for the best way to help your clients boost their efficiency. Robot technology seems like an innovative way to provide more flexible and efficient automation solutions so you’re looking to add autonomous mobile robots to your product offerings, and you believe you should be the one to develop them. This is a costly investment though, and you’re not the only one in the race, so you should carefully consider your options. In this post, we’ll go through 5 questions to ask yourself to make sure you’re in the right position to start development.
Why a robot?
Robots are awesome. Mobile robots especially. There is nothing like seeing a machine you built perform difficult tasks autonomously. But they are also a little hyped. It’s easy to think a robot is the solution to your customer’s problem, because they can do whatever you want them to. However, at RUVU we believe that a simpler solution is a better solution. For example, if you want to move products from one place to another repetitively, you could design a robot that picks them up at A, and places them at B, but a simple conveyor belt might be a better solution. Of course, this depends completely on the requirements of your customer and his process. So make sure you know those, so you can determine the optimal solution.
What’s special about your robot?
When you’re sure a mobile robot is the best solution, you’re ready for the second question. Many AMRs and AGVs already exist. These all have their own specialties. What will be your robot’s specialty?
Some existing vehicles are designed to carry specific loads, such as the Lowpad (Danish trolleys), or Movigo Robotics’ Sharko 10 (Euro pallets). Some are designed for harsh conditions, like ExRobotics’ ExR-1. All of them have something special to make them a good fit for their specific application, but also allows them to scale in that niche. Before bringing your own robot to market, you should figure out what will make your robot stand out from the crowd, and allow it to be a great success.
How is this robot going to make you money?
In order for your robot product to be a business success, obviously it will have to make you money. Of course, you can simply sell units. In many industries, customers expect to buy a machine that simply works for 10, 20 or 30 years. This may be a feasible business model for simple machines, or even robots in very controlled environments. However, generally, robots are complex, software driven utilities. They require service and maintenance. It’s good to have a plan for this before you start development, as it will influence the “returns” part of your return on investment (ROI).
Robots as a Service
In recent years, the phrase Robot as a Service has gained popularity. The philosophy is that the customer does not care about machines or robots, but just wants to be sure that his boxes are moved, or his facility is clean. If you notice this in your customer segment, this can be a very interesting model. Although it is not taking off like a rocket just yet, as robots are moving to smaller businesses that have less capacity to invest in automation solutions, we can see it growing over the next few years.
What’s your budget?
The other side of the ROI is, of course, the investment. Developing your own robot from scratch is a lot of work. Leveraging open source software such as ROS (like we do at RUVU) allows you to decrease that time significantly, but you still need experienced robotics software engineers to distinguish the good from the bad, and to fill in the gaps. It’s possible to accumulate this knowledge and experience in house. This would give you a powerful position, but a time to market of multiple years quickly renders the endeavor useless.
At RUVU, our setup projects with new customers typically range from €50.000 and €100.000, corresponding to lead times of 5-10 months, depending on the complexity of the robot and the desired behavior. This results in a pilot version that you can use for launching customers. (A time to market of less than a year!). As RUVU’s customers develop their own robots, they take care of the mechanical design and production. This can easily cost as much, or even a multiple of the software development. It’s good to be aware of these numbers before starting the development of a robot, so make a good estimation of these costs and your budget before you start.
What can you do yourself? (And what partners do you need?)
Now we’re getting somewhere. You know it’s going to be a robot, what’s special about it, and how you’re going to make a return on the investment. As an automation specialist, you’re familiar with designing complex machines and controlling them. With so many different mobile robots already on the market, most generic problems are solved by now. Still, the complexity of mobile robots, and especially their control software, can be pretty intimidating (here’s a short intro).
It’s worth looking around what’s readily available, as for some use cases control software is available as commercial off the shelf products, just like sensors, motors and wheels. If it looks like such a product solves the challenges in your use case, it’s generally better to buy such a product for its scale benefits, than to develop the component yourself. This will lead you to a few partners or suppliers in the development of your AGV. If you consider these well, they can take you a long way in managing the complexity of the final system.
Now it’s up to you (But you’re not alone!)
If you’ve come this far, you probably have answers to all of the above questions. You’re sure a robot is the right solution for your customers, you’re going to provide them with something unique and useful, all while earning you a revenue with a controlled budget, because you have the right partners. Now it’s up to you!
Of course, you’re free to keep every part of the development of your AMR in house. At RUVU, we’ve seen people underestimate the difference between programming a PLC and building control software for mobile robots, the complexity of fusing sensor data, or the expertise it takes to design and implement an efficient multi-robot planning algorithm. We believe it is essential for any business to specialize, and we decided to specialize in the control software for mobile robots. We’re always happy to chat about robots, even if we’re not doing business. So if you have any questions before or while developing your robot, just call, or send us a message.