Brightpick expands AI in Autopicker robots, Intuition software for more precise fulfillment

a collage of images of the brightpick manipulator.
Clockwise from left: Autopicker, the scanner and what the scanner sees. The bottom right image shows the AI’s outputs, which identify items and the best gripping spots and assign confidence levels to each. | Source: Brightpick

Brightpick, which provides warehouse automation for order fulfillment, has unveiled new details about its Autopicker system, which uses machine vision and artificial intelligence to autonomously pick eaches.

Autopicker uses the Photoneo 3D camera from its parent company, Photoneo, a developer of 3D vision and algorithms for guiding bin-picking robots. In addition to identifying the optimal item to pull from a tote containing multiple items, the technology can also calculate the best location for gripping an item and the approach path for the robot, said Brightpick.

The company added that its Intuition software can orchestrate an entire robot fleet and optimize it for maximum throughput in e-commerce and e-grocery warehouses. The Autopicker autonomous mobile robot (AMR) moves around the facility, retrieves totes from shelving, and robotically picks items to consolidate orders directly in the aisles.

The company described this as a “robot-to-goods” (RTG) workflow. The robot pulls individual SKUs directly from storage totes in the warehouse aisles, without the need to transport bulk inventory to a separate singulation work cell.

Brightpick also said its robot can be used for tasks such as goods-to-person (G2P) picking, pallet picking, stock replenishment, dynamic slotting, order consolidation, buffering, and dispatch.

“Technically, the software is the most highly engineered and valuable part of Brightpick Autopicker,” claimed Jan Zizka, co-founder and CEO of Brightpick, in a release.

brightpick autopicker mobile robot in a warehouse aisle
Autopicker uses an onboard mobile manipulator to pull individual items from bulk inventory. | Credit: Brightpick

Autopicker machine vision trained on 500M picks

Autopicker uses Brightpick’s proprietary machine vision and advanced algorithms, which have been trained on more than 500 million picks, to pick a range of items. The company added that many of its partners have deployed the machine vision system in standalone production applications for robot guidance and bin picking.

The robotic arm on Autopicker uses suction cups, which incorporate sensors to provide feedback to pick items with the right pressure.

To reduce the overall cost, the system has a two-axis SCARA robot. The company asserted that it performs just as well as a six-axis arm because of its combination of machine vision, AI, and suction cups.

Broad-based machine learning and AI enable Autopicker to generalize its perception and pick previously unseen items without additional training, noted Brightpick. The robot can reliably pick ambient and chilled groceries, pharmaceuticals, medical devices, packaged goods, cosmetics, electronics, polybagged apparel, and more, it said.

When consolidating a customer order, Autopicker uses vision to place a picked item at a height of 1 mm (0.03 in.) into the order tote. This is especially useful when picking fragile items, Brightpick explained.

The robot can also make sure that an order tote is not overflowing and place newly picked items in an appropriate empty spot. Not only do AI and machine vision identify gripping points to pull a tote from a shelf, but they also enable Autopicker to compensate for its position in existing facilities with uneven floors.


SITE AD for the 2024 RoboBusiness call for presentations.
Submit your presentation idea now
.


Brightpick Intuition coordinates its fleet

Brightpick said its Intuition software can handle fleet management and fulfillment workflows, including picking, consolidation, and replenishment.

Brightpick Intuition uses “time-space planning” to plan the paths of the Autopicker robots so they do not get stuck in traffic. If a robot must cross the path of another, the software chooses one robot to “give way” to allow the higher-priority robot to pass.

Each Autopicker is designed to avoid obstacles, but if a robot is operating in an aisle and blocking it, other robots will go around it to avoid getting stuck — even if that aisle is the shortest pick path for the other robots. On average, Intuition needs only a few seconds to re-plan the entire robot fleet if something unexpected, such as a human walking into an aisle, disrupts the planned workflows, said Brightpick.

Intuition also dynamically slots bulk inventory in storage locations based on each item’s historical or expected frequency of being ordered by customers. This optimization primarily occurs at night, but the robots can also do this throughout the day as the demand changes, the company said.

It added that Intuition provides standard application programming interfaces (APIs) for integration with warehouse management systems (WMS) or enterprise resource planning (ERP) software.

Zizka recently appeared on The Robot Report Podcast, where he introduced Autopicker and described the opportunity for applications within the warehousing and logistics market.

mike oitzman headshot.
Written by

Mike Oitzman

Mike Oitzman is Senior Editor of WTWH's Robotics Group, cohost of The Robot Report Podcast, and founder of the Mobile Robot Guide. Oitzman is a robotics industry veteran with 25-plus years of experience at various high-tech companies in the roles of marketing, sales and product management. He can be reached at moitzman@wtwhmedia.com.