Operation Support System for Working Machines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing operation support systems for working machines like hydraulic shovels face challenges in adapting to changing work environments and efficiently evaluating operator skills, leading to suboptimal work quality due to manual data selection costs and difficulty in determining skill levels from input signals.
Innovation Solution
An operation support system that includes an operation data detection unit, operator identification, data accumulation, work quality evaluation, and learning units to automatically select and learn from high-quality data, generating an operation model to support operators based on environment conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual selection of operation data is performed to ensure high work quality, then data quality is improved, but labor cost and time consumption increase significantly
Solution Approach 1:
The system performs automatic data selection and quality evaluation without human intervention. The work quality evaluation unit automatically assesses operation data quality based on predefined criteria, and the learning unit automatically selects high-quality data for model training, eliminating the need for manual data curation while maintaining high data quality standards
Solution Approach 2:
The patent replaces the manual mechanical process of data selection with an automated information processing system. The work quality evaluation unit and learning unit use computational algorithms to evaluate and select data based on objective criteria, substituting human judgment with automated evaluation metrics that consistently assess data quality without time constraints
2Adaptability or versatility
If learning data is collected from multiple operators to improve model generalization, then model adaptability is improved, but difficulty in evaluating operator skill levels increases
Solution Approach 1:
The system evaluates each operator's data quality locally based on their individual skill level and work conditions. The work quality evaluation unit assigns different evaluation criteria and weights to different operators, allowing the system to handle heterogeneous data from multiple operators while maintaining accurate skill level differentiation through customized evaluation metrics for each operator
3Measurement precision
If operation model is updated frequently to adapt to changing work environments, then model accuracy is improved, but computational cost and system complexity increase
Solution Approach 1:
The system implements dynamic model updating where the operation model is adaptively adjusted based on changing work environments. The learning unit continuously learns from new high-quality operation data and updates the model parameters dynamically, allowing the system to maintain high accuracy in varying conditions without requiring complete model reconstruction, thus balancing adaptability with computational efficiency
Data Source
AI summary
An operation support system includes an operation data detection unit that detects operation data of a working machine; an operator identification unit that identifies a plurality of operators who operate the working machine; a data accumulation unit that accumulates the operation data and identification information of the plurality of identified operators; a work quality evaluation unit that evaluates work qualities of the plurality of operators and selects the best operator, based on the accumulated data; a learning unit that learns parameters of an operation model of the working machine, based on operation data corresponding to the best operator; and an operation support unit that supports the operators, based on the operation model.