Robot Control Using RF and Vision for Occluded Object Grasping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic systems struggle to locate and retrieve target objects that are partially or fully occluded from view, especially in unstructured environments, leading to inefficiencies and increased costs.
Innovation Solution
A control system that combines radio frequency (RF) signals with visual information from image sensors to determine the location of tagged target objects, even when they are occluded, and generates control signals to guide a robot in grasping the object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If visual sensors are used to locate target objects, then the robot can identify objects in unstructured environments, but the robot cannot retrieve objects that are occluded from view
Solution Approach 1:
The patent combines RFID location determination with visual sensors to create a hybrid system. The RFID tags provide precise location information for occluded objects while visual sensors handle object identification and verification, resolving the contradiction between environment adaptability and retrieval reliability.
Solution Approach 2:
RFID tags serve as intermediaries that bridge the gap between the robot and occluded objects. The tags emit signals that penetrate obstructions, providing location data without requiring direct line-of-sight, thus enabling reliable retrieval of hidden objects.
2Productivity
If the robot operates without human intervention, then efficiency increases, but the robot cannot handle occluded objects requiring human assistance
Solution Approach 1:
The system enables the robot to independently locate and retrieve occluded objects using RFID tags and visual sensors. The robot autonomously determines object locations, plans trajectories, and executes grasping operations without human intervention, maintaining high productivity while handling previously problematic scenarios.
Solution Approach 2:
The system uses feedback from RFID signals and visual sensor data to continuously update the robot's understanding of the environment. This feedback loop enables the robot to adapt to occluded objects and make autonomous decisions, eliminating the need for human assistance while maintaining operational ease.
3Device complexity
If the system is limited to single pile or bin operations, then the system is simpler to control, but it cannot locate or retrieve objects in multiple piles or behind multiple obstacles
Solution Approach 1:
The system divides the environment into multiple discrete piles or bins, each containing RFID-tagged objects. The robot processes each pile independently using the same RFID-visual sensor fusion approach, managing complexity through segmentation while achieving multi-pile operational versatility.
Solution Approach 2:
The control system is designed with universal functionality to handle multiple piles and various obstacle configurations using the same RFID location determination and visual verification approach. This multi-functional design achieves versatility without proportionally increasing control complexity.
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
Enables robots to effectively locate and retrieve occluded target objects, improving efficiency and reducing the need for human intervention by 'seeing through' obstructions using RF-based location determination.
Implementation Method 1
determine a location of a tagged target object in an area of interest based on a radio frequency (RF) signal
Data Source
AI summary
A control system and method locates a partially or fully occluded target object in an area of interest. The location of the occluded object may be determined using visual information from a vision sensor and RF-based location information. Determining the location of the target object in this manner may effectively allow the control system to “see through” obstructions that are occluding the object. Model-based and/or deep-learning techniques may then be employed to move a robot into range relative to the target object to perform a predetermined (e.g., grasping) operation. This operation may be performed while the object is still in the occluded state or after a decluttering operation has been performed to remove one or more obstructions that are occluding light-of-sight vision to the object.