Smart Cart Hand-Motion Recognition for Lightweight Item Tracking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing shopping cart robots fail to accurately detect the addition or removal of products, especially lightweight items, and do not automatically update the shopping list when a user changes their mind about a product.

Innovation Solution

A cart robot equipped with sensors to detect hand motions and a controller using a trained artificial neural network to classify these motions as product addition or removal operations, allowing for accurate updating of the shopping list without relying on weight changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If weight sensors are used to detect product addition/removal, then the detection mechanism is simple, but lightweight items cannot be detected properly

Engineering Contradiction:
Improveproduct detection accuracyVSAvoidlightweight item undetection
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent replaces the mechanical weight-based detection system with an optical sensing system using depth sensors and cameras to detect hand motions. This substitution allows detection of lightweight items that would be imperceptible to weight sensors, as optical detection is not limited by item mass.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces hand motion as an intermediary indicator of product addition/removal. Instead of directly detecting the product itself (which may be too light), the system detects the user's hand motions that precede and accompany product placement, using these motions as a proxy signal for item addition.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If only weight change is used to determine product loading, then the system is simple to operate, but products removed during shopping are not automatically removed from the list

Engineering Contradiction:
Improveautomatic shopping list updatingVSAvoidproduct removal detection
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system continuously monitors hand motions throughout the shopping process and provides real-time feedback by updating the shopping list dynamically. When removal motions are detected, the system immediately reflects these changes, ensuring the shopping list remains synchronized with actual cart contents without requiring manual intervention.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If hand motion sensors and AI classification are added, then product addition/removal detection accuracy improves, but device complexity increases

Engineering Contradiction:
Improvehand motion recognition accuracyVSAvoidsensor and processing system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The depth sensor and camera system serve multiple functions: detecting hand motions for product addition, detecting removal motions, tracking user interactions, and potentially other shopping assistance tasks. This multi-functionality justifies the added complexity by providing comprehensive monitoring capabilities beyond simple product detection.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS11526871B2Cart robot
Publication Date: 2022.12.13 LG ELECTRONICS INC
  • US11526871B2 patent drawing
  • US11526871B2 patent drawing
  • US11526871B2 patent drawing

AI summary

A cart robot analyzing a hand motion of a user is disclosed. The cart robot comprises a code input interface obtaining a product identification code, a sensor detecting a hand motion of a user, a weight measuring device measuring a weight of one or more products contained in a product loading space, and a controller. When there is no change in weight of the loaded products, the cart robot updates a shopping list with respect to the products loaded in the product loading space in real time by analyzing the hand motion of the user. Therefore, the convenience of the user is enhanced.