Vision-Guided Soft End Effector for Cluttered Object Picking

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Soft robots face challenges in picking objects from cluttered environments due to difficulties in distinguishing and grasping specific items amidst multiple objects, particularly in bin-picking scenarios where traditional robots excel.

Innovation Solution

A computer-implemented method and system that utilize a vision system to generate a segmentation map of objects, determine their shapes, and adjust a soft robotic end effector to an optimal shape for grasping by predicting the object's shape and pose, allowing the end effector to deform and adapt for successful picking.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional rigid robots are used for bin picking, then grasping accuracy is high, but adaptability to different object shapes is poor

Engineering Contradiction:

Improvegrasping accuracyVSAvoidadaptability to different object shapes

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The end effector transitions from a static rigid structure to a dynamic soft structure that can actively deform and adapt its shape based on the target object's geometry, enabling both high grasping accuracy and versatility across different object types

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The physical parameters of the end effector (shape, stiffness, configuration) are dynamically changed based on vision system feedback about the target object, allowing the same end effector to optimize its geometry for each specific grasping task

Inventive Principle:

Principle #35Parameter changes

2Object-affected harmful factors

If soft end effectors are used to grasp delicate objects, then gentleness is improved, but difficulty in distinguishing specific items in cluttered environments increases

Engineering Contradiction:

Improvegentleness to delicate objectsVSAvoiddifficulty in distinguishing items in clutter

Core Design Contradiction:

Object-affected harmful factorsVSDifficulty of detecting and measuring

Solution Approach 1:

A vision system acts as an intermediary between the soft end effector and the cluttered environment, providing detailed geometric information about target objects to guide the soft gripper's deformation and positioning, enabling accurate identification and gentle grasping simultaneously

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If the end effector is pre-shaped based on vision guidance, then grasping efficiency is improved, but system complexity increases

Engineering Contradiction:

Improvegrasping efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The end effector is pre-shaped based on advance vision analysis of the target object's geometry, allowing the soft robot to approach and grasp the object more efficiently without requiring complex real-time control during the actual grasping moment

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Complex mechanical control systems are replaced with a vision-guided approach where optical sensing and computational geometry analysis drive the end effector's configuration, simplifying the control architecture while maintaining high grasping efficiency

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12175741B2Systems and methods for a vision guided end effector

Publication Date: 2024.12.24 INTRINSIC INNOVATION LLC

AI summary

Systems and method for an object from a plurality of objects are disclosed. An image of a scene containing the plurality of objects is obtained, and a segmentation map is generated for the objects in the scene. The shapes of the objects are determined based on the segmentation map. An end effector is adjusted in response to determining the shapes of the objects. The adjusting the end effector includes shaping the end effector according to at least one of the shapes of the objects. The plurality of objects is approached in response to the shaping of the end effector, and one of the plurality of objects is picked with the end effector.