Robot Grasp Planning for Human Handovers Without Finger Contact
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic systems face challenges in reliably exchanging objects with humans without interfering with human hands, leading to potential drops and injuries, especially when the robot's grasp interferes with the human hand during object handovers.
Innovation Solution
A vision-based system that uses a depth camera to generate a point cloud of the human hand and object, allowing the robot to classify human hand poses and plan appropriate grasps that avoid contact with the human hand, utilizing a deep network to segment the hand and object and propose safe grasps for the robot.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the robot selects a grasp that can reliably grip an object, then the gripping reliability is improved, but the human hand may be interfered with
Solution Approach 1:
The system segments the object surface into multiple candidate grasp regions and evaluates each region independently. By dividing the grasp planning into discrete regions with different characteristics, the system can identify safe grasp locations that avoid human hand interference while maintaining reliable gripping capability.
Solution Approach 2:
The system applies local quality assessment by evaluating different regions of the object surface with region-specific criteria. Each grasp region is assessed based on its local geometric properties, friction characteristics, and proximity to human hand zones, allowing the robot to select grasps with locally optimized safety and reliability.
2Ease of operation
If the robot uses a standard grasp for object exchange, then the operation simplicity is improved, but the adaptability to different hand poses is reduced
Solution Approach 1:
The system dynamically adapts the grasp selection based on the detected human hand pose. Instead of using a fixed standard grasp, the system adjusts the grasp configuration in real-time according to the human operator's hand orientation and position, maintaining operational simplicity while achieving high adaptability through dynamic reconfiguration.
Solution Approach 2:
The system changes grasp parameters such as contact points, force directions, and grip orientations based on the detected hand pose. By adjusting these parameters dynamically, the system maintains simple operation procedures while adapting to various human hand configurations for reliable object transfer.
Data Source
AI summary
A robotic control system directs a robot to take an object from a human grasp by obtaining an image of a human hand holding an object, estimating the pose of the human hand and the object, and determining a grasp pose for the robot that will not interfere with the human hand. In at least one example, a depth camera is used to obtain a point cloud of the human hand holding the object. The point cloud is provided to a deep network that is trained to generate a grasp pose for a robotic gripper that can take the object from the human's hand without pinching or touching the human's fingers.