LogiMAT 2023: AMR from ABB Robotics for Intelligent Navigation Decision Making – Robotics, Other Material Handling Equipment and AGV, LogiMAT | News | Logistics today

Visual SLAM uses AMR-mounted cameras to create a real-time 3D map of all objects in the environment. The system distinguishes between fixed reference points for navigation, such as floors, ceilings, and walls that need to be added to the map, and objects such as people or vehicles that move or change their locations. The cameras detect and track the natural features of the environment, allowing the AMR to dynamically adapt to its surroundings and determine the safest and most efficient path to its destination. Unlike 2D SLAM, Visual SLAM does not require additional reference points such as reflectors or markers. This saves space and costs and enables positioning with an accuracy of up to three millimeters, according to ABB Robotics.

With no need to change the environment, stop production, or add infrastructure, Visual SLAM reduces startup time by up to 20 percent compared to 2D SLAM. This can also reduce the time it takes to introduce a new AMR into an existing fleet. With remote updates, the technology can be widely used for AMR fleets. According to ABB Robotics, the technology is safe because only raw data is analyzed and no images are stored on AMR or any server.

ABB developed Visual-SLAM-AMR in collaboration with its partner Sevensense Robotics, a provider of artificial intelligence and 3D vision technology. The technology will be integrated into ABB’s latest generation AMR and will be available in the third quarter of 2023 with the AMR T702V and in the fourth quarter of 2023 with the AMR P604V. More AMR solutions with Visual-SLAM will follow by 2025. Visual SLAM technology is already being used on a number of projects for customers in the automotive and retail industries. There it offers the possibility of replacing traditional production lines with intelligent modular production cells fed by AMR.

The Robotic Item Picker uses machine vision and artificial intelligence to determine the best grab points for each item before a vacuum gripper picks up the item and places it in the intended container. According to ABB Robotics, the system requires no human supervision or information about the physical properties of the items to be picked. With a picking rate of up to 1,400 items per hour, companies will be able to process more orders without having to increase staff or time, according to ABB Robotics.

Equipped with a robot, vacuum handles and special image processing software, the Robotic Item Picker automatically takes on complex picking and placing tasks for a large number of items such as cubes, cylinders, bags, boxes, poly bags and blister packs, handling which would otherwise require skill and flexibility from humans.

The Robotic Item Picker is suitable for a wide range of payload requirements and applications and is available with one of the following three ABB industrial robots: the IRB 1200, IRB 1300 and IRB 2600. With a payload of up to three kilograms and a range of up to 1.65 metres, the item picker can meet different requirements processing and sorting orders.

“With its ability to learn and adapt to ever-changing warehouse environments, the AI-powered robotic item picker identifies items and decides how to pick them,” said Daniel Navarro, president of customer and service robots at ABB Robotics for the global line of business.