Direct Teaching of Closed Robot Manipulators via Admittance Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot programming methods for industrial manipulators are complex, laborious, and inefficient for small and medium-sized batches, particularly when direct teaching is required for high-speed continuous trajectories, and most controllers are closed systems that cannot overwrite joint positions or speeds in real time.
Innovation Solution
A method for controlling a closed robotised system using a force/torque sensor and admittance control to record and replicate movements directly taught by an operator, allowing high-speed trajectory recording without open control architecture, and incorporating a parameter adaptation unit to ensure stability and safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional point-to-point programming or offline programming methods are used, then programming can be performed with standard equipment, but the programming process becomes complex and laborious requiring specialized knowledge
Solution Approach 1:
The patent replaces traditional mechanical teaching methods (handheld teach pendants requiring point-to-point programming) with a sensor-based detection system that automatically detects operator gestures and movements. This substitution eliminates the need for operators to learn complex programming languages or use specialized teaching equipment, directly resolving the contradiction between programming ease and programming complexity.
Solution Approach 2:
The patent introduces an intermediary detection system with sensors that act as a mediator between the operator's natural gestures and the robot controller. This intermediary layer translates intuitive human movements into robot commands without requiring the operator to directly program the robot, thereby simplifying the programming process while maintaining standard equipment compatibility.
2Ease of operation
If direct teaching programming is implemented to simplify operator tasks, then programming becomes more intuitive, but real-time control of closed robot systems becomes problematic due to inability to overwrite joint positions or speeds
Solution Approach 1:
The detection system serves as an intermediary that captures operator intentions through gestures and translates them into appropriate control commands. This intermediary layer enables direct teaching functionality in closed robot systems by mediating between the operator's intuitive gestures and the system's automated response, allowing both intuitive programming and maintained automation without requiring open control architecture.
Solution Approach 2:
The patent substitutes traditional direct teaching methods (which require open control systems and manual movement of robot joints) with a gesture-based detection system. This substitution enables intuitive programming while preserving the closed control system's automated joint position and speed control, resolving the contradiction between programming intuitiveness and real-time control capability.
3Reliability
If operator manually guides the manipulator at reduced collaborative speeds for safety, then human-robot interaction is safe, but programming time increases significantly and productivity decreases
Solution Approach 1:
The system performs preliminary detection and recording of operator gestures during a brief teaching phase, storing the gesture sequence and corresponding robot states. This preliminary action allows the full production cycle to proceed at high speeds without requiring the operator to manually guide the robot at reduced speeds, thereby maintaining safety during teaching while preserving productivity during execution.
Solution Approach 2:
The detection system captures and stores a copy of the operator's gesture sequence and the corresponding robot states. This copying mechanism allows the programming to be completed quickly by recording the gesture-trajectory relationship, eliminating the need for repeated manual guidance at reduced speeds, thus maintaining safety while significantly improving programming speed and productivity.
4Productivity
If high-speed movement is used during teaching to maintain productivity, then programming efficiency is maintained, but safety risks increase and system stability becomes compromised
Solution Approach 1:
The system performs preliminary detection and recording of gestures at high speeds during the teaching phase, capturing the complete trajectory and gesture relationship before execution. This preliminary action at high speed maintains programming efficiency, while the actual robot execution during production operates under controlled stability conditions, thus resolving the contradiction between programming efficiency and system stability.
Solution Approach 2:
The detection system creates a high-speed copy of the operator's gestures and corresponding robot states during teaching, storing this data for later execution. This copying approach allows high-speed data acquisition during programming (maintaining productivity) while the actual robot operation during production can proceed at optimized speeds with full stability control, separating the speed requirements of teaching from execution.
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 efficient, high-speed direct teaching of robot manipulators by recording and reproducing operator-guided movements, ensuring system stability and safety, enhancing production flexibility and adaptability to evolving products.
Implementation Method 1
a sensor (10), in particular a force/torque sensor, connected to the end effector (4) and designed to detect the force and the torque (Fs) applied to the handling device (9)
Implementation Method 2
a processing system (11), designed to provide Cartesian movement indications for the robot manipulator (5) depending on the data detected by the sensor (10) and following an admittance control (AC)
Data Source
AI summary
A method to control a closed robotised system comprises a learning step and a reproduction step, wherein, during the learning step, an operator exerts a force and/or a torque (Fc) on a driving assembly, whose sensor detects an applied force and/or torque (Fext); and wherein a processing system carries out an admittance control obtaining, depending on the data detected by the sensor, indications (Xref, X*ref) of movement for the robot manipulator in the Cartesian space; the processing system, following the admittance control, delivers the indications (Xref, X*ref) of movement in the Cartesian space to a trajectory interpolation unit of the robotised system so as to generate a desired trajectory through interpolation.