Robots Atlas>ROBOTS ATLAS
Robotics

Automated order picking and sorting

1995ActivePublished: 9 November 2026Updated: 9 November 2026Published
Key innovation
Shifting order picking and sorting from human "person-to-goods" labour to robotic "goods-to-person" flows and autonomous, vision- and AI-driven single-item grasping.
Category
Robotics
Abstraction level
Technika
Operation level
Robot controlApplicationDeployment
Use cases
E-commerce order fulfilmentDistribution centres and logisticsReturns handling (reverse logistics)Pharmaceutical distribution and pharmaciesGrocery distribution

How it works

In the goods-to-person approach a warehouse management/execution system (WMS/WES) directs mobile robots or a grid storage system to bring a bin or shelf to a workstation, removing operator travel. In robotic piece-picking, cameras and machine vision identify individual items and SKUs, an AI model plans the grasp point and strategy (grasp detection), and an arm with a gripper (suction or jaw) picks the item and places it into an order bin or onto a put wall. Sorting is handled by cross-belt and tilt-tray sorters that route each item to the correct chute or destination based on a scanned code/label (sorter routing). Pick-to-light and put-to-light guide operators via LEDs, and AMRs transport bins between zones.

Problem solved

Manual "person-to-goods" picking is slow, costly and scales poorly across large SKU counts and fragmented e-commerce orders; manual parcel sorting introduces errors and bottlenecks. These systems cut operator travel time, raise accuracy and throughput, and let fulfilment scale.

Components

Goods-to-person systemsBringing inventory to the operator station

Mobile robots carrying shelves, or grid-based cube storage delivering bins to a station, eliminating operator travel. Examples: Kiva/Amazon Robotics, AutoStore.

Robotic piece-pickingAutonomous picking of individual items

A robotic arm with machine vision, a gripper (suction or jaw) and an AI model for SKU recognition and grasp planning (grasp detection).

Cross-belt / tilt-tray sortersAutomated routing of items to destinations

Cross-belt and tilt-tray sorters that route items to the correct chute or shipping destination based on a scanned label.

Put wallsConsolidating items into orders

Cubbyhole walls where picked items are placed (put) into slots corresponding to individual orders, typically guided by put-to-light.

Pick-to-light / put-to-lightGuiding the operator with light signals

LED systems indicating to the operator the location and quantity of items to pick or place, reducing errors and picking time.

Autonomous mobile robots (AMR)Internal transport of bins and items

Mobile robots that navigate autonomously between warehouse zones, transporting bins, shelves or completed orders.

Evolution

1995
AutoStore — grid-based cube AS/RS goods-to-person storage

Founding in Norway of a compact robotic storage system performing goods-to-person picking.

2003
Kiva Systems — mobile robots bringing shelves to the operator
Inflection point

Commercialisation of the robotic goods-to-person model where mobile robots deliver shelves to stationary picking stations.

2012
Amazon acquires Kiva Systems (later Amazon Robotics)

The acquisition scaled robotic goods-to-person picking across global fulfilment centres.

2017
Covariant — AI for robotic piece-picking
Inflection point

Development of AI models for generalised recognition and grasping of diverse items, a breakthrough in autonomous piece-picking.