Waste Crane Control Variables Using Bayesian Predictive Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing techniques for automating crane control in waste disposal plants fail to account for the variable characteristics of waste, leading to inconsistent and ineffective mixing, lifting, and dropping operations due to unstable graspable waste amounts and varying waste properties.

Innovation Solution

An information processing device using Bayesian optimization and robust Gaussian process regression to determine optimal control variables for cranes, adjusting operations based on predictive distributions and task similarity to achieve desired control results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If conventional automatic control techniques are used to control crane operations, then automation is achieved, but control precision deteriorates due to variable waste characteristics

Engineering Contradiction:
Improveautomation of crane controlVSAvoidcontrol precision
Core Design Contradiction:
Extent of automationVSManufacturing precision

Solution Approach 1:

The system implements feedback by using cameras to capture images of waste, calculating graspable amounts and characteristics, and adjusting control variables based on this information. This closed-loop feedback mechanism allows the automated crane to adapt to variable waste characteristics while maintaining control precision.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system dynamically changes control parameters (grasp position, lifting position, dropping position, graspable amount) based on real-time waste characteristics analysis. By adjusting these parameters according to actual waste conditions, the system maintains precision despite automation and waste variability.

Inventive Principle:
Principle #35Parameter changes

2Ease of operation

If fixed control variables are used for crane operations, then control simplicity is maintained, but adaptability deteriorates due to varying waste characteristics

Engineering Contradiction:
Improvecontrol simplicityVSAvoidadaptability to waste variability
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The system performs self-service by automatically analyzing waste characteristics and determining appropriate control variables without human intervention. The crane control system adapts to different waste conditions autonomously, maintaining simplicity of operation while achieving high adaptability through self-adjustment based on real-time waste assessment.

Inventive Principle:
Principle #25Self-service

3Productivity

If graspable waste amount is not stabilized, then operational flexibility is maintained, but reliability deteriorates due to cumulative deviation in waste transfer

Engineering Contradiction:
Improveoperational flexibilityVSAvoidreliability of waste transfer
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system performs preliminary action by calculating and determining the optimal graspable amount before each crane operation. By pre-determining the appropriate graspable quantity based on waste characteristics, the system ensures reliable waste transfer while maintaining operational flexibility for different waste types and conditions.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4112498B1Information processing device, control system, control variable determination method, and control variable determination program
Publication Date: 2026.04.01 KANADEVIA CORP
  • EP4112498B1 patent drawingFigure 1~2
  • EP4112498B1 patent drawingFigure 3
  • EP4112498B1 patent drawingFigure 4

AI summary

A control variable that is likely to yield a desired control result is determined. An information processing device (1A) includes a predictive distribution calculating section (102) configured to calculate or update a predictive distribution of an evaluation function by using control result data (201) of a crane that transports waste; a control variable searching section (103) configured to search for a candidate for an optimum value of a control variable in accordance with the predictive distribution; and a control variable determining section (104) configured to determine the control variable by using the evaluation function based on the predictive distribution to which the update has been made.