Trajectory Generation for Numerical Control Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current numerically controlled (NC) systems face suboptimal results due to local modification of input trajectories without considering the structure and possible modifications of other sections, leading to incomplete information for determining outliers, quantization artifacts, or curvature changes.
Innovation Solution
The method globally modifies the entire trajectory by determining local costs of points in various sequences formed by removing unique combinations of points, selecting an optimal sequence based on the sum of local costs, and fitting splines into preserved points to generate a more accurate trajectory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If local modification methods are used to process trajectory data, then computational complexity is reduced and processing speed is improved, but manufacturing precision deteriorates because suboptimal results are obtained due to limited local information
Solution Approach 1:
The trajectory data is segmented into multiple windows of data points, where each window is processed independently to determine local costs. This segmentation allows parallel processing of different trajectory sections while maintaining the ability to make globally optimal decisions by comparing cumulative costs across all windows.
Solution Approach 2:
The invention transitions from purely local 1D processing to a multi-dimensional approach by considering windows of data points that extend across multiple dimensions of trajectory information (spatial coordinates, velocities, accelerations). This dimensional expansion provides sufficient context to accurately identify outliers and artifacts while maintaining computational efficiency.
2Manufacturing precision
If global modification of the entire trajectory is performed, then manufacturing precision is improved by considering all available information, but computational power requirements increase
Solution Approach 1:
The trajectory is divided into manageable windows that can be processed independently. Each window is assigned a cost value based on its local characteristics, and these local costs are then aggregated to determine the global optimal trajectory. This segmentation reduces computational complexity while preserving the benefits of global optimization.
Solution Approach 2:
Instead of processing the entire trajectory uniformly, the method applies partial processing to each window independently, determining local costs only for the points within each window. This partial action approach reduces overall computational complexity while still achieving global optimization through the aggregation of local costs.
3Device complexity
If small windows of data points are used for local processing, then computational complexity is reduced, but measurement precision deteriorates because insufficient information is available to determine outliers and artifacts
Solution Approach 1:
The invention uses windows that are larger than the minimal single-point local processing, providing excessive local information that is sufficient for accurate outlier detection. Each window contains multiple data points that provide contextual information about corners, quantization artifacts, and other trajectory features, enabling accurate measurement while keeping computational complexity manageable through localized processing.
Data Source
AI summary
A method for generating a trajectory based on an input trajectory formed by points representing spatial coordinates for a numerically controlled (NC) process determines a plurality of sequences of the points and determines, for each sequence, local costs of each point in the input trajectory. Each sequence is formed by removing a unique combination of points from the input trajectory and the local cost of a point for a sequence is determined based on a spatial arrangement of the point with respect to at least some of points in the sequence. For each sequence, a sum of corresponding local costs of each point in the input trajectory is determines and the trajectory is determined based on an optimal sequence with an optimal value of the sum of the local costs.