Robotic Grasping Adaptation via Feature Extrapolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robotic manipulators face challenges in grasping objects when their orientation differs from the predetermined fixed orientation, limiting their ability to perform tasks efficiently and impacting related systems.
Innovation Solution
A grasp management system that generates and validates grasp sets by identifying contact points, expanding feasible contact points based on the geometry of the end-of-arm tool, and extrapolating successful grasps to similar features on different items, allowing robotic manipulators to adapt to various orientations and tools.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the robotic manipulator is instructed to identify and grasp a particular object with a fixed orientation, then the grasping task can be completed successfully, but the system cannot handle objects with different orientations
Solution Approach 1:
The system changes the parameter of orientation by identifying the object's actual orientation through image processing and adjusting the grasp plan accordingly. The grasp management system modifies grasp parameters (contact points, approach vectors, end effector orientation) based on the detected object orientation, allowing successful grasping regardless of initial object placement.
Solution Approach 2:
The system transitions from a static, fixed-orientation grasp plan to a dynamic, adaptive grasp planning process. The grasp management system continuously evaluates object orientation and generates appropriate grasp plans in real-time, enabling the robotic manipulator to adapt to varying object orientations during operation.
2Ease of operation
If the robotic manipulator uses a specific end-of-arm tool for grasping, then the grasping action can be performed, but the system cannot adapt to use different tools
Solution Approach 1:
The grasp management system implements universality by generating tool-agnostic grasp plans that can be executed by different end effector tools. The system identifies grasping features and contact points that are independent of the specific tool, allowing the same grasp plan to be transferred across multiple tool types while maintaining grasping effectiveness.
Solution Approach 2:
The system segments the grasping task into independent components: object feature identification, grasp pose determination, and tool-specific execution. This segmentation allows the core grasp planning to be tool-independent while enabling different tools to execute the same grasp plan through their respective capabilities.
3Device complexity
If the system requires objects to be in a predetermined fixed orientation, then the grasping process is simplified, but the system cannot handle objects in various orientations
Solution Approach 1:
The system performs preliminary action by pre-identifying multiple candidate grasp poses and contact points on the object before execution. The grasp management system generates a set of feasible grasp plans in advance, allowing the robotic manipulator to select the most appropriate grasp based on the actual object orientation without requiring complex real-time decision-making.
Data Source
AI summary
A grasp management system and corresponding methods are described. In some examples, information about a set of grasps of a robotic manipulator is accessed. The information is used by the robotic manipulator to attempt to grasp an item using the set of grasps associated with a first grasping orientation. An orientation of the robotic manipulator can be adjusted into a second grasping orientation and the robotic manipulator can attempt to grasp the item using the set of grasps associated with the second grasping orientation. Information about the attempts can be recorded and used to determine a richness measure that may represent a richness of the set of grasps for the item.