Parametric Robot Toolpaths for As-Built Object Deviations

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robotic control systems face challenges in performing path-based tasks effectively when the target object deviates from its nominal model, leading to inconsistencies and inaccuracies due to predetermined toolpaths that do not account for real-time object positions and features.

Innovation Solution

A system and method that generate parametric toolpaths defined with respect to object features, allowing for real-time mapping and adjustment of the toolpath to the as-built object, enabling the end-effector to perform tasks accurately even if the object deviates from its model.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If a predetermined toolpath is used for the end-effector, then the task execution is simple and fast, but the accuracy deteriorates when the target object deviates from its model

Engineering Contradiction:
Improvetask execution accuracyVSAvoidtoolpath generation complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies dynamics by transitioning from static predetermined toolpaths to dynamic parametric toolpaths that adapt in real-time. The system continuously updates toolpath parameters based on actual object measurements, allowing the end-effector to adjust its trajectory dynamically during task execution to maintain accuracy despite object deviations

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements parameter changes by using parametric models where toolpath parameters are expressed as functions of object features. When object deviations are detected through scanning, the system modifies the parametric parameters (such as position, orientation, or trajectory points) to compensate for deviations, thereby maintaining task execution accuracy without complete reprogramming

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If the toolpath is updated in real-time to account for object deviations, then the accuracy is improved, but the time consumption increases

Engineering Contradiction:
Improvetask execution accuracyVSAvoidtoolpath update time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-defining the parametric structure of the toolpath based on the nominal object model before actual task execution. This preliminary parametric framework allows for rapid real-time adjustments without requiring complete toolpath regeneration, thus reducing time consumption while maintaining accuracy

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback through a closed-loop system where the object is scanned during task execution, deviations from the nominal model are detected, and the parametric toolpath parameters are automatically adjusted based on this feedback. This real-time feedback mechanism enables accurate adaptation without significant time delays

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3585570B1Automatic generation of toolpaths
Publication Date: 2024.09.04 INTRINSIC INNOVATION LLC
  • EP3585570B1 patent drawingFigure 1
  • EP3585570B1 patent drawingFigure 2
  • EP3585570B1 patent drawingFigure 3

AI summary

Example implementations relate to generating instructions for robotic tasks. A method may involve determining task information of a path-based task by an end-effector on an object, where the task information includes (i) at least one task parameter, and (ii) a nominal representation of the object. The method also involves based on the task information, determining one or more parametric instructions for the end-effector to perform the task, where the one or more parametric instructions indicate a toolpath for the end-effector to follow when performing the task. The method also involves generating, based on sensor data, an observed representation of the object, and comparing the observed and the nominal representations. The method further involves based on the comparison, mapping the parametric instructions to the observed representation of the object.