Garrett Place, business development at ifm efector, talks with Manufacturing Matters about making more efficient autonomous mobile robots (AMRs) that deliver a better ROI for customers. One of the biggest challenges companies face when adopting AMRs is the system’s total cost of ownership. But Place describes a “paradigm shift” where end users have to increase cost to decrease cost. To be faster and more efficient, an AMR needs to add perception to attain a greater understanding of its environment — i.e., identifying obstacles so the vehicle plans a new path rather than stops. Adding this technology increases the robot’s cost yet decreases its overall mission time. This reduction in mission time allows the customer to decrease the number of robots required to manage throughput, and fewer robots translates to a significant drop in a company’s capital expenditure.

John Lewis: [00:00:07] Hey everyone. Welcome to another episode of Manufacturing Matters. I’m John Lewis with TechB2B, here in Boston at A3’s the Vision Show. I’m speaking with Garrett Place, business development at ifm effector. Garret, tell me a little bit about what you guys are exhibiting at the show this week.

Garrett Place: [00:00:24] Thanks for having me, John. Yeah. So what ifm is bringing to the show this week is concentrating on 3D imaging for mobile robotics. So the Vision show, co-located with the AMR show, has worked really well for us. And what we’re trying to discuss with our customer base and even other vendors of camera systems is simply how do we almost change the direction or change the discussion with camera systems in trying to make the AMR experience more efficient. So trying to reduce the overall mission times of AMRs or AGVs so that we can bring a better ROI to the end user customer that may be deploying AMRs.

John Lewis: [00:01:09] That’s really interesting, Garrett. Can you tell me a little bit about how decreasing the mission time could impact the adoption of AMR technology?

Garrett Place: [00:01:19] Yeah, and I’ll stumble through this, John, to make sure that I’m getting all of this right. So what we typically hear in the industry is that the challenge for mobile robotic adoption is the total cost of ownership of the system. And one of the first things that we try to do in that is to reduce the BOM cost or hardware cost of our systems to make that total cost a little bit less. But the math doesn’t truly work out because even if we reduce ours by 10% or 15%, we’re such a small part to that robot that we’re actually not making a dramatic increase or decrease to that cost. But what we see is really the cost reduction for AMRs can truly be impacted if you start to decrease the mission time. And if you decrease the mission time, you need less robots to manage the throughput that is required for the customer. And if you have less overall robots, you have a dramatic decrease in the capital expenditure of the company.

John Lewis: [00:02:20] So you’re saying that in order for the robots to go faster, they might actually need more perception?

Garrett Place: [00:02:26] Exactly. It’s almost that paradigm shift where you actually have to increase cost to decrease cost. And so what we’re seeing out in the market space is these AMRS, in order to be more efficient, faster, either faster in their drive time or faster in their pick time, if it’s an automated fork truck, is simply it needs a greater understanding of its environment. It needs to see more. A great example of this simply is, if you have an object in the way, something that might not be seen by the safety ladder, something below, something above, the vehicle might contact that object and stop the vehicle. Well, as long as that vehicle’s stopped, it’s not continuing on its mission. And if we can’t continue on its mission, of course we lose time. We’re less efficient. And so what we’re seeing is that, let’s say the latest trend is to add perception, add cost to the vehicle in order to have greater understanding of the environment. And therefore it decreases overall mission time per robot. And then for the end customer it decreases the number of robots required to manage throughput.

John Lewis: [00:03:30] That’s really interesting because if a robot is not moving because it’s run into something it doesn’t understand, you’re going to need more robots to achieve the throughput required at a facility. Which increases your capital costs.

Garrett Place: [00:03:45] Correct. So, yes, if you need more robots because you have a robot stopping, but the next generation, that distinguish from AGV to AMR, is first you detect the object, so we see where it’s at, and then you can make intelligent decisions on whether you truly have to stop because someone has to come in and take action. Or what most AMRs do today is they actually detect the object and then move around the object. So we don’t actually stop anymore. We dynamically path plan around that object, but that’s only viable or that’s only available if you can actually identify all the objects in the scene. And more importantly, you can identify a proper path that you can see that doesn’t have those obstacles in the way.

John Lewis: [00:04:30] Aside from increasing perception to reduce mission times, average mission times of an AMR fleet, are there other things that are being worked on that can help scale up the adoption of this technology?

Garrett Place: [00:04:45] Yes, and “scale” is a great word because that’s what we’re all trying to get to in the mobile robot industry is getting these robots to mass adoption. So crossing that chasm over from the early adopters into that small to mid-size engineering companies that can handle this. And, yes, you’re correct, there are so many other cost points for these systems. Hardware is one. But what we found in this industry is that most global robots require more cameras, and not just cameras from one supplier but they have cameras from multiple suppliers or different modalities, like 3D cameras, 3D cameras and let’s say lidars. And the problem statement for the OEM at this point is not choosing one camera or the other. It’s how do they put them all together. So they’re spending a lot of time up front in trying to integrate camera systems onto the robot before they even get started in solving problems. And so at ifm, what we’re showing here is simply what is our approach to reducing friction for the integration of not only ifm cameras but all other products that may be used in a mobile robot.

John Lewis: [00:05:48] That’s great. So you’re saying if you can reduce friction in the integration, that will also help address the capital costs of these systems?

Garrett Place: [00:05:55] Yeah, there are two cost points there, right? Two pain points. One is, do I need six weeks to develop a solution or integrate a solution onto the robot or can I do it in six days? So you’re saving all of that engineering and R&D time to manage that part of the process. The second part is, once we have an intelligent system on the robot and you go to deploy that robot out to the field, how can we reduce a three-month deployment into a three-week deployment? Because most companies, when they see the need — let’s say there’s a labor shortage or they can’t get the right help —and they want to apply some automation into that facility, they can’t wait three years or nine months or 12 months. They need that help earlier. And that pain point there, how do we get that time to be smaller, which will enable bigger integration?

John Lewis: [00:06:46] That’s great. Is there anything that you can think of that I may have missed or not asked you about?

Garrett Place: [00:06:51] No, I think this is a great conversation. And it’s interesting because it’s not just an ifm discussion. It’s not just a mobile robot discussion. There has to be really an open discussion on how do we deploy the right technologies onto these robots, so we simplify the R&D, we simplify the deployment of the robots, we make them more efficient. So instead of a 12-minute mission time, we get to an 11-minute mission time, and we make it more approachable for the small to midsized enterprise so they can start adopting these robots like the Fortune 100 and Fortune 500 companies today.

John Lewis: [00:07:28] Excellent. Well, thanks for your time, Garrett. I really appreciate it.

Garrett Place: [00:07:31] Thanks for having me, John.