Information Processing System for Task Manual Creation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for creating task manuals struggle to accurately detect human movements from moving images, making it difficult to correlate operator movements with pre-created task procedures due to the complexity of human movements.
Innovation Solution
An information processing system that detects and correlates changes in the state of a target object with pre-created task procedures without requiring the detection of operator movements, using image processing to identify and track the target object's state changes within captured images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If operator's movement is detected from moving images to create task manuals, then the task manual can be correlated with task procedures, but the detection accuracy is low due to the complexity of human movements
Solution Approach 1:
The patent introduces an intermediary approach by detecting target objects and their state changes as a mediator between the operator's complex movements and the task procedure correlation. Instead of directly analyzing complex human movements, the system detects changes in target objects (e.g., position, state) that result from operator actions, thereby simplifying the detection task while maintaining accuracy in correlating with task procedures
Solution Approach 2:
The patent replaces the mechanical system of direct movement detection with an information processing approach. Instead of using complex computer vision algorithms to track human body parts and movements, the system substitutes this with a simpler detection method that identifies target objects and their state changes, reducing computational complexity while achieving the same goal of creating accurate task manuals
2Loss of information
If every movement of the operator is detected to create comprehensive task manuals, then the completeness of the task manual improves, but the detection difficulty increases significantly
Solution Approach 1:
The patent extracts only the essential information needed for task manual creation by focusing on target object detection and state changes rather than attempting to detect all operator movements. This extraction approach captures the critical information (what objects are affected and how their states change) while filtering out unnecessary data about the operator's body movements, thereby improving completeness without proportionally increasing detection difficulty
3Productivity
If traditional movement detection methods are used to correlate operator actions with task procedures, then the task manual creation is possible, but the efficiency is reduced due to detection inaccuracies
Solution Approach 1:
The patent inverts the traditional approach by not detecting operator movements directly to correlate with task procedures. Instead, it detects target objects and their state changes, then correlates these with task procedures. This inversion simplifies the detection process, improves accuracy, and ultimately enhances the efficiency of task documentation while maintaining precise correlation with task procedures
Data Source
AI summary
An information processing system includes one or more processors configured to: from an image captured of a target object of a task and a worker performing the task, detect the target object and a change in a state of the target object; and correlate information about the detected change in the state of the target object with information about a pre-created procedure of the task.