Autonomous Shelf Robot Identifying Misplaced Products
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing inventory monitoring systems in retail settings require manual creation and updating of planograms, necessitating substantial human intervention for tracking products and identifying misplaced or out-of-stock items.
Innovation Solution
A mobile, autonomous robot equipped with multiple cameras navigates store aisles to collect images of products and fixtures, analyzing panoramic images to detect product presence, identity, and placement, thereby automating inventory tracking and identifying misplaced or out-of-stock items without the need for manual planograms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If manual creation and updating of planograms is used, then human intervention is required for tracking products, but this increases labor cost and time consumption
Solution Approach 1:
The system uses machine vision technology to automatically detect, identify, and track products on shelves without requiring manual intervention. The automated planogram generation compares detected product positions with expected positions to identify misplaced items, eliminating the need for manual planogram updates while maintaining accurate inventory tracking
Solution Approach 2:
The patent replaces manual mechanical processes of planogram creation and updating with automated machine vision systems. Cameras capture images of shelves, computer vision algorithms process the images to detect products and their positions, and software automatically generates and updates planograms, substituting human labor with automated technological systems
2Extent of automation
If machine vision technology with fixed cameras is used, then product monitoring is automated, but manual planogram creation is still required
Solution Approach 1:
The system performs preliminary actions by automatically capturing images of shelves and generating initial planograms using machine vision technology. The automated system processes these images to create detailed product placement maps without requiring manual input, preparing all necessary data in advance for ongoing monitoring operations
Solution Approach 2:
The machine vision system serves itself by automatically detecting products, identifying their positions, generating planograms, and comparing actual versus expected placements. The system performs all these functions autonomously without requiring manual planogram creation or updates, reducing both labor requirements and system complexity
3Measurement precision
If manual monitoring of product inventory is performed, then product tracking is accurate, but this increases labor cost and reduces productivity
Solution Approach 1:
The patent replaces manual visual inspection and monitoring with automated machine vision systems. Cameras capture high-resolution images of shelves, computer vision algorithms automatically detect and identify products, determine their positions, and track inventory levels. This substitution maintains or improves tracking accuracy while dramatically increasing productivity by eliminating manual monitoring labor
Solution Approach 2:
The machine vision system enables continuous monitoring of inventory without interruption. Unlike manual monitoring which occurs periodically when staff are available, the automated system operates continuously, constantly capturing images, detecting products, and updating inventory records, thereby maintaining high accuracy and improving productivity simultaneously
Data Source
AI summary
Disclosed herein is a system and method of identifying misplaced products on a retail shelf using a feature extractor trained to extract features from images of products on the shelf and output identifying information regarding the product in the product image. The extracted features are compared to extracted features in a product library and a best fit is obtained. A misplaced product is identified if the identifying information produced by the feature extractor fails to match the identifying information associated with the best fit features from the product library.