Sensorized Robotic Gripper With ML Slip Scoring for Adaptive Grip
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic grippers face challenges in effectively grasping objects without causing slippage, damaging the object, or consuming excessive power, particularly when using grippers with more than two fingers, and existing methods are complex and costly.
Innovation Solution
A robotic gripper equipped with sensors that detect slippage using machine-learning algorithms, adjusting gripping forces incrementally based on slippage detection scores without requiring complex force or friction measurements, allowing for rapid adaptation to slippage and reduced power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If gripping force is increased to prevent slippage, then slippage is reduced, but power consumption increases and object damage risk increases
Solution Approach 1:
The gripping force is made dynamic rather than static. The control system continuously monitors slippage detection scores and adjusts gripping forces in real-time, increasing force only when slippage is detected and decreasing it when stable, thereby optimizing power consumption while preventing slippage
Solution Approach 2:
A feedback loop is implemented where sensors detect slippage conditions and feed this information back to the control system, which then adjusts the gripping force accordingly. This closed-loop control ensures minimal power consumption while maintaining reliable grip
2Reliability
If gripping force is increased to prevent slippage, then slippage is reduced, but object damage risk increases
Solution Approach 1:
The gripping force is dynamically adjusted based on actual slippage conditions rather than applying maximum force continuously. This dynamic adaptation prevents excessive force that could damage the object while still preventing slippage when needed
Solution Approach 2:
The feedback mechanism monitors slippage detection scores and only increases gripping force when slippage is actually detected, avoiding unnecessary force application that could harm the object
3Measurement precision
If complex force and friction measurements are used to optimize grip, then grip precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent extracts only the essential information needed for slippage detection rather than measuring all force components. By focusing on slippage detection scores from simplified sensors rather than comprehensive force measurements, the system achieves adequate grip control with reduced complexity
Solution Approach 2:
Complex mechanical force measurement systems are replaced with simplified sensor-based slippage detection. The system substitutes detailed mechanical sensing with a more straightforward detection approach that provides sufficient information for effective grip control
4Stability of the object's composition
If symmetric finger arrangement is used, then force balance is improved, but adaptability to different objects decreases
Solution Approach 1:
The patent embraces asymmetric finger arrangements to improve adaptability to different object shapes and sizes. Rather than forcing symmetric configurations, the system adapts to asymmetric realities of various objects while maintaining control through sophisticated force management
5Use of energy by moving object
If gripping force is decreased to minimize power consumption, then power consumption is reduced, but slippage increases
Solution Approach 1:
The gripping force is dynamically modulated based on real-time slippage detection. The system maintains lower power consumption during stable periods and automatically increases force when slippage occurs, achieving an optimal balance between energy efficiency and grip reliability
Data Source
AI summary
A robotic gripper including one or more fingers and at least one actuator for acting on the fingers in order to grip an object, the gripper including at least one sensor sensitive to a relative movement between the object manipulated by the gripper and the latter, this sensor delivering an output signal, a control system configured to execute at least one machine-learning algorithm trained to deliver, based on the one or more output signals generated by the one or more sensors, at least one slippage-detection score representative of the confidence of the algorithm in the presence of slippage, and to transmit control data to the actuators controlling the one or more fingers, the control data being generated so as to relate the strength of the forces applied by the one or more fingers to said score.


