Robot Hand Gripping Pose Calculation From 3D Finger Placements

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for calculating gripping poses for robot hands are time-consuming, especially when dealing with multiple hand shape models and varying object orientations, leading to increased processing time and potential collisions in picking randomly placed objects.

Innovation Solution

An information processor that detects candidate finger placement positions based on 3D measurement data and hand shape data, searches for combinations of these positions to calculate a gripping pose, eliminating the need for repeated calculations across multiple hand shape models and ensuring collision avoidance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple hand shape models with different opening widths are used to calculate gripping poses, then the accuracy of gripping pose recognition is improved, but the processing time increases significantly

Engineering Contradiction:

Improvegripping pose recognition accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the gripping pose calculation process into two independent stages: first detecting candidate placement positions for each finger independently using 3D measurement data and hand shape data, then searching for combinations of these positions. This segmentation eliminates the need to create multiple hand shape models with different opening widths, as the candidate position detection is performed once and then combined in various configurations, thereby reducing processing time while maintaining accuracy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary detection of candidate placement positions for each finger before combining them into multi-finger configurations. By pre-detecting valid single-finger positions based on 3D measurement data and hand shape data, the system avoids repeated calculations when evaluating different opening widths, as these candidate positions serve as the foundation for all subsequent combination searches.

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If the largest possible opening width is set for the hand to grip all target objects, then the versatility of the gripping system is improved, but collisions with surrounding objects occur more frequently

Engineering Contradiction:

Improvegripping capability for different objectsVSAvoidcollision with surrounding objects

Core Design Contradiction:

Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The patent applies local quality by detecting candidate placement positions for each finger independently based on the local 3D measurement data and hand shape data at that specific position. Instead of using a uniform largest opening width for all gripping scenarios, the system determines optimal placement positions locally for each finger, allowing the hand configuration to adapt to the specific geometry and constraints of each target object and its surroundings, thereby avoiding collisions while maintaining versatility.

Inventive Principle:

Principle #3Local quality

3Adaptability or versatility

If model-less gripping position recognition is used for objects without 3D CAD models, then the adaptability to various object shapes is improved, but the calculation complexity increases

Engineering Contradiction:

Improvecompatibility with different object typesVSAvoidcalculation complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent replaces the traditional mechanical approach of creating detailed 3D CAD models for each object with a data-driven approach using 3D measurement data. Instead of relying on pre-defined object models, the system directly processes measured geometric data to detect candidate placement positions, substituting complex model-based mechanics with efficient data processing and pattern recognition algorithms that work universally across different object types.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12097615B2Information processing device and information processing method

Publication Date: 2024.09.24 OMRON CORP

AI summary

An information processor calculates, for a robot hand including a plurality of fingers, a gripping pose at which the robot hand grips a target object. The information processor includes a candidate single-finger placement position detector that detects, based on three-dimensional measurement data obtained through three-dimensional measurement of the target object and hand shape data about a shape of the robot hand, candidate placement positions for each of the plurality of fingers of the robot hand, a multi-finger combination searcher that searches for, among the candidate placement positions for each of the plurality of fingers, a combination of candidate placement positions to allow gripping of the target object, and a gripping pose calculator that calculates, based on the combination of candidate placement positions for each of the plurality of fingers, a gripping pose at which the robot hand grips the target object.