Trajectory-Based Mechanism Generation Without Manual Geometry Iteration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mechanism design methods are time-consuming, difficult to modify, and fail to account for factors like motion and actuation, leading to inefficient and labor-intensive processes.
Innovation Solution
A method based on motion trajectory generation, involving augmented trajectories, screening criteria, and deep learning models to efficiently construct three-dimensional mechanisms with integrated physical and dynamic characteristics, reducing the need for geometric definitions and manual adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional geometric-based mechanism design methods are used, then design precision can be achieved through detailed calculations, but the design process becomes time-consuming and labor-intensive
Solution Approach 1:
The patent replaces traditional geometric calculation methods with a data-driven approach using machine learning models. The system learns from training data containing mechanism configurations and their corresponding motion trajectories, enabling the model to predict mechanism parameters directly from trajectory requirements without time-consuming geometric calculations and manual optimization iterations.
Solution Approach 2:
The patent creates a digital model (machine learning model) that copies and learns from existing mechanism design data and performance characteristics. By training on comprehensive datasets including geometric parameters, motion trajectories, and actuation requirements, the model captures design patterns and can generate new mechanism designs that replicate successful design principles without requiring repeated manual analysis.
2Reliability
If traditional mechanism design processes are followed, then comprehensive analysis can be performed, but design modification becomes difficult and time-consuming
Solution Approach 1:
The patent creates a dynamic design system where the machine learning model can be retrained and updated with new data. When design modifications are needed, the system can incorporate new requirements into the training data and retrain the model, allowing flexible adaptation without restarting the entire design process. The model continuously learns from new information, making the design process adaptable to changing requirements.
Solution Approach 2:
The patent implements a feedback mechanism where the machine learning model's predictions can be evaluated against actual performance data, and the model is continuously improved through retraining with new data. This feedback loop allows the system to learn from design modifications and performance outcomes, enabling easy adaptation while maintaining comprehensive analysis through the model's learned knowledge of design-performance relationships.
3Manufacturing precision
If conventional design methods are used, then detailed geometric parameters can be optimized, but factors like motion and actuation are not adequately considered
Solution Approach 1:
The patent creates a universal machine learning model that handles multiple design aspects simultaneously - geometric parameters, motion trajectories, actuation requirements, and performance constraints. The model is trained on comprehensive datasets that include all these factors, enabling it to generate mechanism designs that optimize geometric parameters while simultaneously satisfying motion and actuation requirements, rather than treating them as separate optimization problems.
Solution Approach 2:
The patent merges previously separate design considerations (geometry, motion, actuation) into a single integrated machine learning model. The training data combines all these factors, and the model learns the interrelationships between them, allowing simultaneous optimization of geometric parameters while ensuring motion accuracy and actuation feasibility are met through the unified prediction framework.
4Manufacturing precision
If genetic algorithms are used for data processing, then optimization can be achieved, but the iterative process becomes time-consuming and inefficient
Solution Approach 1:
The patent performs preliminary action by pre-training the machine learning model on comprehensive optimization data before actual design tasks. During training, the model learns from extensively optimized examples and performance data, capturing optimal design patterns and relationships. When deployed, the model can directly predict optimized mechanism parameters without requiring time-consuming iterative optimization runs, as the optimization knowledge has already been embedded during the preliminary training phase.
Data Source
AI summary
A method of mechanism generation based on trajectory includes the following steps: step A: generating plural augmented trajectories based on an initial trajectory, and collecting plural pieces of motion trajectory data from the initial trajectory and the augmented trajectories; step B: screening each piece of motion trajectory data according to a screening criterion in order to eliminate motion trajectory data falling short of the screening criterion; step C: forming a diagram by blending motion trajectory data conforming to the screening criterion, wherein the diagram includes at least two nodes and at least one connecting line connecting the two nodes, each node includes a physical feature, and the connecting line includes a dynamic characteristic between the two nodes; and step D: constructing a three-dimensional mechanism based on the physical features and the dynamic characteristic.