AMRs can safely work around people and forklifts in dynamic warehouse environments
After over 4 years of hard work and input from hundreds of industry experts, the Robotics Industry Association (RIA) has published the new American National Standard for safety requirements for industrial mobile robots, R15.08, an important step toward common guidelines in the growing sector of mobile robotics.
To ensure a facility is using the safest equipment, you should choose AMRs that meet all aspects of the new R15.08 safety standard, meaning that the complete AMR, base together with modular top, conform. While many AMR manufacturers comply with the safety standard for the base robot, their integrators that assemble a modular top together with the AMR base have yet to conform. The entire AMR (base plus top together) needs to conform to the standard to truly address safety with the facility and allow full collaboration with humans.
At Fetch Robotics, we’ve worked diligently to ensure our entire commercial AMR product line not only conforms with the new R15.08 standard, but with the all the requirements for CE marking as well, so that you can confidently deploy on-demand automation knowing your workforce and facility will be safe and in compliance with the latest regulations
All AMRs are equipped with a sophisticated sensor package consisting of a 2D LiDAR scanner, as well as multiple 3D depth cameras. These sensors work seamlessly together to facilitate robust robot vision and smart identification of both static and dynamic objects within the robot’s field of view.
From the robot’s perspective, it can not only see the associate pulling the pallet jack, but also the boxes on the jack and the surrounding racks and walls. This is facilitated by upward and downward facing 3D cameras. Our proprietary safety technology enables the robot to identify moving objects, as identified by the red voxels overlaid on the human and pallet jack.
These technologies combine to provide a safe and efficient autonomous transport system, since the robot can now safely plan and move around objects in a dynamic environment.
AMRs differ front AGVs in that AMRs can re-plan and avoid obstacles, while AGVs are programmed to stop and wait. This is an important distinction, since AMRs not only avoid collisions, but can actually plan a path around a moving obstacle, increasing efficiency.
If you look closely at the robot perspective on the bottom right, you will notice that Fetch’s obstacle avoidance technology has calculated the direction of an object’s movement, which is notated with green arrows. Along with the sensor data, these inputs are indicating to the robot that it will need to stop in order to avoid a moving obstacle.
Importantly, because of the upward facing 3D camera, the robot is also smart enough to not attempt to travel between the cart legs. The robot only proceeds once it deems the object has been cleared, and it is now safe to proceed along its navigated path.
Proper sensing is mission critical to mitigate safety risks with autonomous travel in a varied environment such as a warehouse or manufacturing facility. Here, the AMR is demonstrating overhang detection, enabling the robot to avoid a collision with a hanging object in the robot’s path.
Robust sensor integration and system safety protocols prevent mobile robots from traveling under structures that they cannot clear (with or without payloads), hitting overhanging objects, or getting stuck in a dangerous location.
AMRs are able to see a pallet jack on the ground (and other low-to-ground objects such as forklift tines) and instantly plan a safe path around them. This feature is especially important in warehouse environments, since it is common to have debris, boxes, pallets, and other potential safety hazards in the path of an AMR. Our AMRs have been designed to detect ground obstacles reliably.
AMR workflows can be initiated in a variety of methods, by both the warehouse operator and WMS, WES, or MES.
FetchCore™ is cloud-based robotics software that gives you complete control of your robots and automation.