End Effector Force Feedback Control for Stable Object Grasping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robots struggle to accurately handle objects due to sensor errors, finger length differences, non-force-based instructions, and occlusions, making it difficult to recognize and manipulate objects without using object information.
Innovation Solution
An operation control method and device that utilize action force information to calculate and control the operation of an end effector with multiple contact portions, setting a center coordinate system, converting contact point positions, and adjusting instructions based on contact forces to match target action forces, allowing object handling without relying on object information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If sensor-based finger angle detection is used for remote control, then operation transmission is enabled, but handling accuracy deteriorates due to sensor errors and finger length differences
Solution Approach 1:
The system continuously detects contact forces at multiple contact portions and feeds back to calculate action forces, which are then used to correct the end effector's operation. This closed-loop feedback mechanism compensates for sensor errors and finger length differences by adjusting the robot's movements based on actual contact force measurements.
Solution Approach 2:
The patent replaces direct mechanical transmission of finger movements with a force-based control system. Instead of directly transmitting mechanical finger angles to robot joint angles, the system uses contact force detection and action force calculation to mediate the control, substituting the mechanical transmission chain with a sensor-based force feedback mechanism.
2Device complexity
If non-force-based instructions are used, then control simplicity is maintained, but handling reliability deteriorates due to inability to detect contact status
Solution Approach 1:
Contact force detection provides real-time feedback about the object's presence and contact status to the control system. This feedback enables the system to reliably determine when the end effector is in contact with the object and adjust its operation accordingly, significantly improving handling reliability while maintaining relatively simple control architecture.
Solution Approach 2:
The contact force detection system acts as an intermediary between the simple non-force-based instructions and the complex force-based execution. It mediates by detecting actual contact conditions and providing corrected instruction values that bridge the gap between simple user input and reliable object handling.
3Manufacturing precision
If object information is used for recognition, then handling accuracy is improved, but system complexity increases due to occlusion problems
Solution Approach 1:
The end effector uses its own contact force sensing capabilities to detect and recognize objects during handling operations. Instead of relying on external vision systems or pre-programmed object databases, the system serves itself by using the contact portions' force sensors to directly perceive the object's presence, shape, and contact characteristics, eliminating the need for complex external recognition systems.
Solution Approach 2:
The patent substitutes optical or sensor-based object recognition systems with a mechanical force-based recognition approach. By using contact force detection at multiple portions, the system can identify objects and their characteristics through mechanical interaction rather than optical sensing, avoiding occlusion problems and reducing system complexity.
4Adaptability or versatility
If multiple contact portions are used, then handling versatility is improved, but control precision deteriorates due to coordinate system transformation errors
Solution Approach 1:
The system calculates action forces based on actual contact force measurements from multiple contact portions and uses this feedback to correct the control instructions. This feedback mechanism compensates for coordinate system transformation errors by continuously adjusting the end effector's movements to match the actual contact conditions, maintaining high control precision despite the complexity of managing multiple contact points.
Data Source
AI summary
An operation control method includes: a step of transmitting action force information when an operator handles an object to an end effector; a calculation step of calculating a target action force for grasping the object using the end effector from an acquired state of the operator and the acquired action force information; a detection step of detecting a contact force at contact points at which the contact portions are in contact with the object; and a control step of controlling an operation of a driver of the end effector such that an action force acting on the object from each contact element matches the target action force.