Robotic Grasp Planning With Physics-Based Candidate Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robotic devices face challenges in securely grasping objects due to varying object sizes and environmental obstacles, leading to inefficiencies and potential damage during object movement.
Innovation Solution
A method using a physics-based model to evaluate and select high-quality grasp candidates for robotic grippers, considering multiple feasible placements and suction cup activations to ensure secure object grasping, even in constrained environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the gripper is positioned to grasp the object, then the object can be moved, but the gripper may hang off the object face or collide with obstacles
Solution Approach 1:
The system performs preliminary collision checks and grasp quality evaluations before the gripper attempts to grasp the object. Multiple candidate grasps are pre-computed and ranked by quality metrics, allowing the system to select the best feasible grasp while avoiding collisions with obstacles.
Solution Approach 2:
The gripper is divided into multiple suction cups that can be independently controlled. This segmentation allows the system to activate only the necessary suction cups for a secure grasp, improving adaptability to different object geometries and reducing the risk of the gripper hanging off the object face.
2Reliability
If multiple grasp candidates are evaluated using physics-based models, then grasp quality improves, but computational time increases
Solution Approach 1:
The system pre-computes multiple grasp candidates and evaluates them using physics-based models before the actual grasping operation. This preliminary evaluation ensures high grasp quality by selecting from pre-ranked candidates, while the computational overhead is performed in advance rather than during critical execution.
Solution Approach 2:
The system generates and evaluates a limited number of top-ranked grasp candidates rather than exhaustively searching all possible grasps. By focusing computational resources on the most promising candidates (excessive action on a subset), the system achieves high grasp quality without excessive computational time.
3Adaptability or versatility
If the gripper is designed to accommodate various object sizes, then versatility improves, but the gripper may not achieve secure contact on small object faces
Solution Approach 1:
The gripper is segmented into multiple independently controllable suction cups. This allows the system to activate only the suction cups that will make contact with the object, adapting to various object sizes and geometries. For small objects, only the necessary subset of suction cups is activated, ensuring secure contact without the gripper hanging off the object face.
Solution Approach 2:
The system dynamically selects and activates specific suction cups based on the detected object geometry and the planned grasp pose. This dynamic configuration allows the same gripper hardware to securely grasp objects of varying sizes by adapting which actuators are engaged, maintaining grasp security across different object scales.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the robotic device's ability to efficiently and securely grasp objects, reducing the need for human intervention and minimizing object damage by prioritizing higher quality grasps and adapting to environmental constraints.
Implementation Method 1
An on-board vacuum system may then be activated to use suction to adhere the object to the gripper.
Data Source
AI summary
Methods and apparatus for determining a grasp strategy to grasp an object with a gripper of a robotic device are described. The method comprises generating a set of grasp candidates to grasp a target object, wherein each of the grasp candidates includes information about a gripper placement relative to the target object, determining, for each of the grasp candidates in the set, a grasp quality, wherein the grasp quality is determined using a physical-interaction model including one or more forces between the target object and the gripper located at the gripper placement for the respective grasp candidate, selecting, based at least in part on the determined grasp qualities, one of the grasp candidates, and controlling the robotic device to attempt to grasp the target object using the selected grasp candidate.