Robotic Grasp Planning for Unknown Objects in Cluttered Workspaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous robots face challenges in efficiently picking and placing unknown objects from a cluttered workspace to a designated drop-off area, due to difficulties in discerning object boundaries and shapes, leading to potential damage and reduced throughput.
Innovation Solution
A robotic system equipped with a plurality of cameras to visualize the workspace, determining object geometry and potentially graspable features, and selecting appropriate grasp strategies based on scores of successful grasps, allowing for adaptive picking and placing of unknown objects without prior programming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional autonomous robots are used to move objects, then they can work continuously without rest, but they struggle to discern object boundaries and shapes leading to potential damage and reduced throughput
Solution Approach 1:
The patent introduces an intermediary system consisting of multiple cameras and computer vision algorithms that mediate between the robot and the objects. This intermediary captures images from multiple angles, processes them to discern object boundaries and shapes, and provides guidance information to the robot, thereby enabling safe and reliable object manipulation without direct physical contact or trial-and-error approaches
Solution Approach 2:
The system performs preliminary actions by capturing images of objects from multiple angles before the robot attempts to manipulate them. The computer vision system processes these images in advance to identify object boundaries, shapes, and characteristics, creating a digital model that guides the robot's subsequent actions. This preliminary visualization and analysis prevents damage by ensuring the robot understands what it is about to manipulate
2Adaptability or versatility
If robots attempt to grasp unknown objects without prior identification, then they can handle diverse object shapes, but they face difficulties in discerning object boundaries and selecting appropriate grasp strategies
Solution Approach 1:
The patent applies segmentation by dividing the object recognition task into multiple components: capturing images from multiple angles, processing each image to identify features, combining information to discern complete object boundaries, and separating different object characteristics (shape, size, texture) for independent analysis. This segmented approach makes the complex task of identifying unknown objects manageable and accurate
Solution Approach 2:
The system implements universality through a multi-functional camera system that captures images serving multiple purposes: boundary detection, shape identification, texture analysis, and grasp point selection. The same image processing pipeline handles diverse object types (rigid, flexible, transparent, reflective) by adapting algorithms based on detected object characteristics, enabling the robot to grasp various unknown objects with a single unified system
Data Source
AI summary
A set of one or more potentially graspable features for one or more objects present in a workspace area are determined based on visual data received from a plurality of cameras. For each of at least a subset of the one or more potentially graspable features one or more corresponding grasp strategies are determined to grasp the feature with a robotic arm and end effector. A score associated with a probability of a successful grasp of a corresponding feature is determined with respect to each of a least a subset of said grasp strategies. A first feature of the one or more potentially graspable features is selected to be grasped using a selected grasp strategy based at least in part on a corresponding score associated with the selected grasp strategy with respect to the first feature. The robotic arm and the end effector are controlled to attempt to grasp the first feature using the selected grasp strategy.