Geometric Constraint Solver Segmentation for Predictable CAD Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computed Aided Design (CAD) systems using variational geometric constraint solvers often produce unpredictable results due to two-way geometric constraints, where modifying one entity can transitively affect many others, leading to unintended changes in the design.
Innovation Solution
The method involves classifying geometric entities and constraints into groups based on one-way and two-way constraints, allowing users to specify directionality, and using a variational geometric constraint solver to update affected entities while restricting movements that would violate constraints, thereby preventing unintended change propagation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If two-way geometric constraints are used to link geometric entities, then the system maintains flexibility and automatic updating capability, but modifying one entity causes unintended transitive changes to many other entities
Solution Approach 1:
The patent segments the constraint network by introducing directional constraints (one-way and two-way) that divide the transitive propagation path. By classifying constraints into directional categories, the system breaks the undirected connectivity that causes unintended changes, allowing selective updating of only those entities that should be affected by a modification.
Solution Approach 2:
The patent inverts the traditional symmetric constraint model by implementing asymmetric one-way constraints where the dependency relationship is directional. Instead of allowing bidirectional influence between entities, the constraint enforcement flows in a specified direction, preventing changes from propagating upstream to entities that should remain unaffected.
2Manufacturing precision
If a variational geometric constraint solver is used to satisfy all geometric constraints simultaneously, then mathematical correctness is achieved, but the result may not match user intent due to unexpected change propagation
Solution Approach 1:
The patent applies preliminary action by classifying constraints into one-way and two-way categories before solving the constraint network. This pre-processing step establishes the directional dependency structure that guides the solver, ensuring that when constraints are satisfied, changes propagate only in intended directions rather than causing unexpected transitive effects.
Solution Approach 2:
The patent applies local quality by allowing different constraint relationships to have different propagation characteristics. One-way constraints enforce local dependency in a single direction, while two-way constraints allow bidirectional propagation. This localized control over change propagation ensures that each constraint relationship behaves according to its specific requirements rather than applying a uniform propagation rule throughout the entire network.
3Stability of the object's composition
If geometric entities are highly interconnected through geometric constraints, then the system maintains design consistency, but moving one entity causes the whole design to be reshaped
Solution Approach 1:
The patent segments the interconnected constraint network by introducing directional boundaries through one-way constraints. This segmentation isolates groups of entities that should move together while preventing unwanted influence between independent design groups, thereby maintaining consistency within groups without causing system-wide reshaping.
Solution Approach 2:
The patent extracts the directional dependency information from the symmetric constraint relationships and uses it to control propagation. By separating the constraint enforcement direction from the geometric relationship itself, the system maintains necessary connections for consistency while removing the harmful transitive propagation that causes excessive design reshaping.
Data Source
AI summary
Systems, program products and program products for accepting a request to move a first geometric entity that is constrained to other geometric entities in a drawing that may contain both one-way and two-way geometric constraints. A set of affected geometric entities that need to be repositioned or otherwise changed as the result of moving the first geometric entity is identified. The affected geometric entities and the geometric constraints are classified into groups which are solved using a variational geometric constraint solver. If the geometric constraints cannot be satisfied, the movement of the first geometric entity is restricted. Otherwise the affected geometric entities are updated.