As anyone reading this knows, performance is everything in the world of industrial robotics. It often boils down to cycle time – i.e., the time it takes to complete a specific set of tasks. Whether it’s painting, welding, or something else entirely, the goal is always the same: finish as quickly as possible and move on to the next job. But what happens when your cycle time drags? That workcell becomes an inefficient bottleneck, slowing down the entire assembly line.

Naturally, when you’re looking to boost performance, the first idea that comes to mind is adding more robots. It seems straightforward enough; more robots should mean more work gets done and that it gets done faster, right? But the reality is much more complicated. It’s like adding more cooks to a kitchen; sometimes it works, but most of the time, you just end up with too many people tripping over each other.

The fact of the matter is just adding more robots isn’t enough. You need to maximize the benefit of every single one of them.

The Multi-Robot Conundrum

So why is it so hard to achieve better performance in multi-robot workcells?

Picture this: you’re tasked with designing a workcell that has four robots and 50 tasks that need to be completed per cycle. Now, not only do you have to figure out which robot does what, but you also have to decide the order in which they do it. The number of possible designs is mind-boggling. Even if you could test one design per nanosecond, it would still take longer than the age of the universe to go through all the possibilities.

What makes a good design stand out is when every robot is kept busy, with minimal idle time, and when they can move around without getting in each other’s way. It’s all about finding that perfect choreography. The difference between a good design and a bad one is like night and day in terms of performance. Most bad designs mean collisions, resets, and a return to the drawing board.

The Traditional Approach

So, how do engineers typically approach this problem? It’s all about taking a sample design, evaluating it, and then deciding where to go next based on what you’ve learned. It’s a bit like navigating in the dark with only a flashlight to guide you – you take a step, look around, and then decide where to step next.

Choosing that first step, or initial design, is daunting. Even with years of experience, it’s hard to know where to start. You might try to allocate tasks evenly among the robots or assign tasks based on proximity, but those are just educated guesses. The real challenge comes when you start evaluating these designs. You need to create a motion plan for each robot’s movements, which is a complex, time-consuming process – and often the reason why multi-robot workcells don’t work as intended.

And then, once you have the workcell design you feel is best, you need to manually try it out, making adjustments along the way until you reach a point where you feel the design and setup is the most optimal. But you can only evaluate a handful of designs, so you might miss out on the best ones. You’ll never be 100% sure, nor will you have the time or resources to keep trying.

The Needle in the Haystack

This is why we here at Realtime Robotics developed our Optimization solution. In short, it’s a proprietary algorithm that can sift through the enormous number of possible workcell designs and find the one that delivers the best performance. It’s already proven to be successful for many of our clients and partners, such as VW and Valiant TMS.

Our software can analyze hundreds of thousands, even millions, of designs in a matter of hours. To expand upon an analogy, it’s like going from using our flashlight to a floodlight; you can see so much more of the landscape, and you can use that knowledge to make better decisions. This ability to rapidly evaluate so many designs means we can find the highest-performing design – one that would otherwise remain hidden. 

The journey to fully optimize multi-robot workcells is ongoing, and I’m sure there are technological advancements still to come. But there’s no reason to continue with time-consuming, manual methods that deliver less-than-perfect outcomes. Check out our Optimization solution and drop us a line at info@rtr.ai if you’d like to learn more about how it can help you shave seconds off of your cycle times.

About Realtime Robotics

Realtime Robotics is the leader in automatic, collision-free motion planning for industrial robots. Its innovative technology generates optimized motion plans and interlocks to achieve the shortest possible cycle time in single and multi-robot workcells. The company’s solutions expand the potential of automation, empowering multiple robots to work closely together in unstructured and collaborative workspaces, reacting to dynamic obstacles the instant changes are perceived. Learn more about Realtime Robotics here, watch its technology in action here, and connect with us on X and LinkedIn.