Pallet Sorting Conveyor Using Sensor Fusion and ML Grading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pallet sorting systems lack efficient methods for grading and sorting pallets based on physical properties, leading to inefficiencies in pallet handling and management.
Innovation Solution
A pallet sorting system utilizing a conveyor system, loading assembly, and machine learning algorithms to collect and analyze data from sensors, determining optimal conveyance paths for pallets, and incorporating a lifting mechanism, overhead conveyor, and release assemblies to sort and stack pallets based on their condition and type.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual pallet sorting is used, then simplicity of system is maintained, but productivity and sorting precision deteriorate
Solution Approach 1:
The patent replaces manual mechanical sorting with an automated system using sensors, machine learning algorithms, and robotic actuators. The sensor array captures pallet images and data, the machine learning module analyzes this data to determine pallet grades and optimal stacking positions, and robotic push rails automatically move pallets to designated locations, eliminating manual labor while improving productivity
Solution Approach 2:
The system enables pallets to be automatically graded and sorted without human intervention. The sensor array autonomously captures pallet characteristics, the machine learning module independently determines pallet grades and stacking positions based on captured data, and the robotic system self-executes the sorting process by moving pallets to appropriate locations, creating a self-service automated sorting system
2Productivity
If automated sorting system is implemented, then productivity is improved, but measurement precision requirements increase
Solution Approach 1:
The patent divides the measurement task into multiple specialized sensor components, each optimized for detecting specific pallet properties. The sensor array includes cameras for visual inspection, LIDAR for dimensional measurement, and other specialized sensors for detecting different physical characteristics. This segmentation allows each sensor to focus on specific measurements, improving overall measurement precision while maintaining high processing speed
Solution Approach 2:
The patent combines multiple types of sensors and data sources into a unified measurement system. The sensor array integrates various sensing technologies to capture comprehensive pallet data simultaneously, and the machine learning module merges this multi-source data to make accurate grading decisions, achieving both high precision and high speed through data fusion
3Manufacturing precision
If pallets are sorted by multiple criteria, then sorting precision is improved, but device complexity increases
Solution Approach 1:
The patent implements a universal machine learning module that handles multiple sorting criteria through a single integrated system. This module can evaluate pallets based on various factors including visual condition, dimensional specifications, weight, and other physical properties, adapting to different sorting requirements without requiring separate dedicated systems for each criterion, thus achieving high sorting precision while controlling system complexity
Data Source
AI summary
A method for conveying a pallet through a pallet sorting system having a conveyor system and a loading assembly. The method includes receiving a pallet at a first position of the conveyor system or the loading assembly. A first sensor senses a first set of pallet data and a second sensor senses a second set of pallet data. A module receives the first set of pallet data and the second set of pallet data. The module selectively communicates at least one subset of pallet data from the first set of pallet data, the second set of pallet data, or a combination thereof to a machine learning module, which uses the subset of pallet data to determine a third set of pallet data. The third set of pallet data is used to determine a conveyance path for the pallet.