Region-Based Robotic Grasp Generation for Orientation-Constrained Picking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic grasp generation techniques are inefficient, computationally expensive, and lack the ability to automatically generate high-quality grasps that account for the orientation requirements of subsequent part placement, often requiring manual teaching and being unsuitable for machine tending applications.
Innovation Solution
A region-based grasp generation method that uses user-defined target grasp regions on a part, combined with an optimization solver to compute stable grasps, which are stored in a database for efficient robotic grasping, ensuring compatibility with the orientation requirements of the destination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual teaching is used to generate robotic grasps, then the grasps can be customized for specific machine tending requirements, but the process is slow and costly
Solution Approach 1:
The system performs preliminary action by pre-defining target grasp regions on the part model before actual grasp generation. This allows the optimization algorithm to focus computations only on relevant regions, significantly reducing computation time while maintaining high grasp quality suitable for machine tending applications
Solution Approach 2:
The system creates a digital copy of the part model with defined grasp regions, allowing virtual testing and optimization of grasps without physical trial-and-error. This digital twin approach enables automatic grasp generation that matches the quality of manual teaching but at much higher speed
2Productivity
If automated grasp generation is used, then productivity is improved, but computational cost and time increase significantly
Solution Approach 1:
The system segments the part surface into multiple target grasp regions, each independently optimized. This segmentation allows the computation to be distributed and focused on specific areas, reducing the overall computational burden compared to analyzing the entire part surface uniformly
Solution Approach 2:
The system applies local quality by defining specific target grasp regions with unique properties on different parts of the object. Each region can have customized optimization parameters, allowing efficient computation focused on critical grasping areas rather than uniform analysis of the entire part
3Extent of automation
If existing automated grasp generation techniques are used, then manual teaching is eliminated, but the grasps do not account for orientation requirements of subsequent placement
Solution Approach 1:
The system incorporates feedback by using the destination pose requirements as constraints in the grasp optimization process. The orientation requirements from the subsequent placement operation feed back into the grasp generation, ensuring that generated grasps are compatible with the required destination orientation
Solution Approach 2:
The system performs preliminary action by pre-defining target grasp regions based on destination orientation requirements before actual grasp computation. This ensures that the automation process inherently accounts for placement requirements, making the system adaptable to different machine tending scenarios
Data Source
AI summary
A region-based robotic grasp generation technique for machine tending or bin picking applications. Part and gripper geometry are provided as inputs, typically from CAD files, along with gripper kinematics. A human user defines one or more target grasp regions on the part, using a graphical user interface displaying the part geometry. The target grasp regions are identified by the user based on the user's knowledge of how the part may be grasped to ensure that the part can be subsequently placed in a proper destination pose. For each of the target grasp regions, an optimization solver is used to compute a plurality of quality grasps with stable surface contact between the part and the gripper, and no part-gripper interference. The computed grasps for each target grasp region are placed in a grasp database which is used by a robot in actual bin picking operations.