Netherlands based Accerion introduces Triton, the world’s first 100% infrastructure-free, sub-millimeter accuracy localization technology for mobile robots. Triton is a new localization sensor designed specifically for autonomous mobile robots (AMR) and automatic guided vehicles (AGV).
The company has tested the Triton in a variety of applications and pilot programs around the world over the last 6 months, and is launching this solution today. With Triton deployed on your AMRs and/or AGVs, you’ll be able to move and position them with unprecedented accuracy. Accerion is quoting “sub-millimeter” accuracy with its solution. This is an important claim, as most AMRs and AGVs can only achieve +/- 10 mm accuracy with existing localization solutions.
Infrastructure-Free? How Does Triton Work?
Consisting of an optical sensor, an integrated processing unit and Accerion’s highly advanced algorithms, Triton scans the floor surface beneath robots and uses it as a reference. Triton uses features of the floor surface as a map to provide coordinates to the navigation system so robots can localize themselves in the environment.
No lasers. No magnetic tape. No induction lines.
“We pushed hard to get maximum performance out of the hardware. We used a superior camera, expanding the optics’ capabilities. Reliability improved dramatically after we put Triton through the ringer to make sure it’s beyond robust for industrial activities,” said Ruben Vandenheede, Accerion System Engineer.
When Do You Need To Use Localization Technology Like Triton?
Every AMR uses some form of localization algorithm or technology onboard the AMR to enable it to remain “localized” within its environment. Localization simply means that the AMR knows where it is on the “global” facility map. When an AMR loses its localization, it becomes “lost” and requires human operator intervention to “re-localize” it.
AMRs typically use an algorithm called Simultaneous Localization and Mapping (SLAM) to remain localized within their workspace. SLAM functions by matching the information coming from either an onboard LIDAR (laser sensor) or an onboard camera to the facility map that the AMR maintains. One potential problem with SLAM is that it can become confused when an AMR crosses through a large empty space (like an empty airplane hanger) or when it’s operating in a space with similar looking aisles (as in a warehouse) or in spaces where there is a continuously changing environment (such as changing pallet locations on a loading dock).
The innovation that Triton brings to the market is that it uses a visual sensor onboard the vehicle to look at the facility floor. When accurately imaged, the facility floor provides sufficient information for the Triton to determine the precise orientation and location of the vehicle.
Not only does this help ensure that the AMR doesn’t get lost (as in the use cases described above), but it also helps to accurately position the vehicle for operations such as interfacing with conveyors or other fixtures in the environment.
Accerion also describes the capability for “virtual line” following in AGVs, which means that an AGV which used to leverage a painted line or magnetic tape on the floor, can now use data from Triton to remain on a virtual path through the facility.
Warehouse fulfillment is Triton’s first market, one of the fastest-growing sectors globally. Accerion responded to market demands, with engineers taking up the challenge to design Triton from the ground up based on client specifications. That meant packing the technology from Accerion’s flagship product Jupiter into a compact localization sensor for smaller AMRs.
Accerion Triton indoor localization solution (image courtesy of Accerion)
“While developing Triton, we listened closely to the market and customers. Optimizing size, performance, cost, reliability, features and Triton’s certification set (CE, FCC, RoHS) was a technical challenge, but our engineering team is one of the most competent in our industry, and now Triton is ready to go global,” said Accerion Founder and Managing Director, Willem-Jan Lamers.
For the autonomous mobile robot sector, easily and accurately determining the robot’s location – independent from the environment – is essential, specifically in warehouse logistics and production automation.