Complementary Model for Load Distribution in Transport Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods cannot specify the three-dimensional position of a transport object in a drop target, limiting the accuracy of load distribution and balance in transport vehicles.
Innovation Solution
A method for producing a complementary model that acquires distribution information and incomplete distribution information as a dataset for learning, training the model to estimate the complete distribution information from incomplete data, thereby specifying the three-dimensional position of the transport object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the center of gravity position is determined using weighting sensor output, then the center of gravity position can be calculated, but the three-dimensional position of the transport object in the drop target cannot be specified
Solution Approach 1:
The patent uses a complementary model that learns from paired data (complete distribution information and corresponding incomplete distribution information) to generate complete distribution information from incomplete sensor data. This copying approach allows the system to infer missing spatial distribution details while maintaining the core measurement capability from the weighting sensor.
Solution Approach 2:
The system performs preliminary learning by training the complementary model on paired datasets before actual measurement. During operation, the trained model then automatically complements incomplete distribution information from sensor inputs, enabling three-dimensional position specification without requiring additional complex sensors during the actual measurement process.
2Measurement precision
If complete distribution information is obtained through additional sensors, then three-dimensional position can be specified, but the device complexity increases
Solution Approach 1:
The patent replaces additional mechanical sensors with an information-processing approach. Instead of physically adding sensors to capture complete three-dimensional distribution data, the system uses a complementary model that processes incomplete sensor data through machine learning to infer the complete distribution, thereby avoiding the complexity of additional hardware.
Solution Approach 2:
The complementary model acts as an intermediary between the simple weighting sensor data and the required complete three-dimensional distribution information. This intermediary component translates incomplete sensor readings into complete distribution maps, eliminating the need for direct complex sensor systems while maintaining measurement precision.
Data Source
AI summary
A method for producing a complementary model includes acquiring distribution information and incomplete distribution information as a dataset for learning and training the complementary model using the dataset. The distribution information indicates a distribution of an amount of a transport object in a drop target of a work machine, and the incomplete distribution information is information from which some values of the distribution information are missing. The complementary model is trained using the dataset such that when the incomplete distribution information is used as an input value, the complementary model complements the missing values and outputs the distribution information an output value.