Last September, Clearpath Robotics introduced the OTTO 1500, a robotic mover-of-stuff designed to autonomously haul pallets of things weighing up to 1500 kilograms. That’s a lot of kilograms, and frankly, it’s more kilograms than most people need delivered to them personally at any given time. Today, Clearpath is announcing a bite-sized follow-up to the OTTO 1500, the OTTO 100. It’s smaller, more agile, slightly cuter, and with a 100 kg payload, is just the right size to be your new best friend. If you work in a warehouse, that is.
Factories and warehouses are often already set up with autonomous vehicles that can move stuff from place to place, but these robots that I’m just going to go ahead and call “primitive’ must follow predefined routes and react to obstacles by coming to a panicky halt, taking the rest of the system with them. OTTO 100 is smart enough and flexible enough to avoid obstacles on its own, plan new routes when necessary, and make sure to keep from smushing the humans that it works with. You can equip it with a lift, bin carrier, or a cart, and as you add more units, it’ll easily scale within Clearpath’s existing fleet management system.
OTTO 100 and OTTO 1500 are designed to work together: pallets are great, and the OTTO 1500 can move them around for you, but at some point, all of that palletized whatnot has to get depalletized and distributed, and this is where the OTTO 100 fits in.
For more details on OTTO 100 and how things are going at Clearpath, we spoke with CEO Matt Rendall:
IEEE Spectrum: What kinds of things is OTTO 100 designed to do?
Matt Rendall: We’re very focused on line-side parts delivery for manufacturing. In manufacturing plants, you need to bring inventory from the warehouse to the assembly line, and the OTTO 100 specializes in that. If you look at how materials are transported through factories today, a lot of carts are used. We want to be able to plug into that existing concept of operation; the difference is that we’re able to use self-driving vehicles.
How does OTTO navigate safely?
Our focus is building a very robust navigation and mapping capability that relies primarily on a single 2D LIDAR, and being able to accomplish that, it turns out, is quite complex. We also fuse in odometry and inertial data, and those are the three primary systems we use for navigation and mapping. There are other companies that are doing a lot of stuff with cameras, and I think vision systems have come a long way over the years and will play an important role in our future, but today, in order to accomplish what we require, we only rely on LIDAR as our primary navigation system.
We’ve done some very extensive experimentation on the impact on mapping capability, localization score, and anomalies that result from varying LIDAR height. The lower you are to the ground, the less likely you are to run over something that’s below your LIDAR field, but the more likely you are to encounter something overhanging that you don’t see. There’s a delicate balance that needs to be achieved there, and it’s a combination of making an informed decision about the engineering tradeoffs, and providing reasonable external requirements for your customer: to put it simply, just keep the roads clear.
In terms of how we maintain safety, we’ve done a lot of analysis on what motion profiles are acceptable in what circumstances to insure that if there is a motion being executed that’s outside of the field of view of our LIDAR, it’s at a speed where you’ve reduced the energy in the system.
What distinguishes the OTTO 100 from similar robots in this space?
I think the important thing to acknowledge is that we need to put egos aside, and at the end of the day, a mobile robot, by itself, is not nearly as powerful as an end-to-end solution to a very specific problem. What we’re offering is much more than just a mobile robot that you can just apples-to-apples compare to the competition. There’s system-level thinking that needs to be applied here.
To put this into a sound bite, the factory-side integration through our fleet management system I think is a very different aspect of what we’re offering. It’s how we think about charging and maximizing uptime. It’s the mating infrastructure, so carts for instance, that enables a specific use case. It’s our understanding of the domain of line-side delivery. And it’s our experience over the last six or seven years designing highly rugged field robotics systems, which we’re bringing into the factory to fulfill the promise of 24/7/365 productivity.
You’re trying to hire about 75 new people this year. With all of the competition out there for roboticists, how are you going to make that happen?
What we have working in our favor is that we’re an established company. There’s a lot of spotlight that’s applied to startups because that’s an exciting area, but there’s only a subset of people who want jobs that come with that amount of risk. And because we’re more established, and we have very strong ties to the global academic community, our brand is a dominant brand in education and research, so roboticists already know us, and they’re trained on our products. To put this in context, last year, we had 5,000 people apply to work at Clearpath.
How are the PR2s doing?
The PR2 community is still very active and engaged. The way the research world works, they go through grant cycles, and they will stretch their equipment for decades. That’s part of the reason why, if you tour around robotics labs, you’ll see some labs are still using the iRobot ATRV, from back when they were still doing robotics research platforms. So I think that even though there aren’t new PR2s being released in the world, there’s a very vibrant community of top robotics research labs that are building the future of mobile manipulation and humanoid robotics research on that platform. There are newer systems emerging on the market, but the PR2 is still a very capable platform.
Anyway, back to the news: Clearpath won’t tell us what the price is for the OTTO 100, since like many robots of this type, you can’t really just buy one of them. Clearpath has to help map your site, integrate OTTO’s software, provide support, and so on. The cost of the hardware itself is all wrapped up in there as well, but separating it out isn’t really meaningful, they tell us. And besides, these little guys work best in teams, so you wouldn’t want just one anyway.
OTTO will make its first public appearance at MODEX, a big supply chain show (they have those) that takes place April 4-7 in Atlanta, Ga.
Evan Ackerman is a senior editor at IEEE Spectrum. Since 2007, he has written over 6,000 articles on robotics and technology. He has a degree in Martian geology and is excellent at playing bagpipes.