Warehouse picking robot

This invention

This invention is an autonomous warehouse picking robot that brings several capabilities together in one mobile system. Stereo cameras perceive shelving in 3D. A suction gripper lifts items. A learned model estimates where and how to grasp an object from depth images. SLAM-based navigation lets the robot map and move through aisles while holding mixed types of products (SKUs). It belongs to the field of warehouse robotics and automated material handling — specifically robotic perception, grasping, and autonomous navigation working as one.

Where it fits

Your idea sits in a well-developed corner of Robotics (B25J) and Conveyors & Material Handling (B65G). These areas appear in this result set at roughly 128× and 113× the overall corpus baseline — a sign of a tightly clustered, actively pursued direction. The work also draws on Image Processing (G06T), Computer Vision (G06V), and AI & Machine Learning (G06N) for grasp estimation, plus Autonomous Control (G05D) for navigation. Filings here run steadily from 2007 through 2024, with consistent recent activity (5 in 2023, 5 in 2024). Active groups include AMAZON TECHNOLOGIES, SYMBOTIC, INDUSTRIAL PERCEPTION, and EVOLUTION ROBOTICS, among others — established players in a lively field. The landscape also touches Traffic Control (G08G) and Vehicle Control Systems (B60W), which appear less often in these results.

Closest related work

US-11602857-B2 — Autonomous mobile robotic systems and methods for picking and put-away (IAM ROBOTICS · 1 citation · 5-member family)

Filed and granted in 2023, this describes a logistics-facility system that pairs a remote server with manipulation robots — and optionally transport robots — to autonomously pick and put away items, totes, or cases. It shows how IAM Robotics coordinated mobile manipulation with central task assignment, which overlaps directly with your robot's role of picking across mixed-SKU shelving. Its companion patent US-9785911-B2 lays out the earlier piece-picking foundation.

US-10646999-B2 — Systems and methods for detecting grasp poses for handling target objects (TATA CONSULTANCY SERVICES · 1 citation · 5-member family)

This focuses on the grasp-pose problem in cluttered warehouse bins. It uses a neural network for object recognition, then builds surface segments to find graspable affordances. It shows how Tata approached learned grasp estimation in a confined, messy space — close to your depth-image-based grasp-pose model, and a useful comparison for structuring perception and graspability scoring.

US-12159192-B2 — Robotic systems and methods for identifying and processing a variety of objects (BERKSHIRE GREY OPERATING COMPANY · 0 citations · 24-member family)

Filed in 2024, this uses an articulated arm with multiple perception systems aimed at a common area to capture several views of an object being processed. It shows how Berkshire Grey approached multi-view perception for handling a wide variety of objects — relevant to your stereo-vision and mixed-SKU goals, and an example of how richer sensing improves grasp reliability.

US-7162338-B2 — Computing a relative pose for global localization in a visual SLAM system (EVOLUTION ROBOTICS · 339 citations · 40-member family)

A foundational, highly-cited patent on visual SLAM. It combines a visual sensor with dead-reckoning to localize and build maps economically. It shows how Evolution Robotics approached the navigation half of your invention — the SLAM-based mapping that lets a robot move autonomously through shelving — and it's a heavily referenced touchstone for the field.

What you can do next

• Explore & build on it. Browse the related work above — new, differentiated ideas often come from combining or improving on existing approaches (a specific grasp-scoring method, a suction-gripper geometry, a sensor-fusion technique, or a navigation strategy for mixed-SKU shelving that others haven't pinned down).
• If you'd like to protect it. Filing a provisional application (usually with a patent attorney) is a common first step. Most inventions can be protected in some form — what matters is how broad and defensible that protection is, which is where a patent attorney adds value (a very narrow claim may be granted but protect very little