Robot Learning Control Using Gyro Sensors for Vibration Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional learning control systems for robots, particularly in spot welding applications, face limitations such as narrow application range, usability issues, and difficulties in calculating trajectory/position errors on each axis directly from sensor data, leading to reduced productivity and increased system costs due to vibration and trajectory errors.
Innovation Solution
A robot system with a learning control unit that calculates a learning correction amount based on data from sensors like acceleration sensors, using a H-infinity norm design for the learning control filter to ensure convergence and monotonic decrease, allowing for increased operation speed and reduced vibration, and incorporating a zero-phase filter to address offset issues without performance deterioration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If learning control is executed using conventional methods with acceleration sensors, then trajectory and position errors can be measured, but the application range is narrow and usability is poor due to difficulties in calculating errors on each axis
Solution Approach 1:
The patent replaces conventional acceleration sensors with gyro sensors to detect angular velocity. This substitution enables direct calculation of trajectory and position errors on each axis through integration of angular velocity data, eliminating the complex mathematical operations required by acceleration sensors and improving both applicability and usability.
Solution Approach 2:
The patent changes the detection parameter from acceleration to angular velocity. By using gyro sensors that detect angular velocity and integrating this data over time, the system directly obtains position information on each axis, simplifying the calculation process and expanding the application range to include more robot types and operations.
2Productivity
If robot operates at high speed, then productivity increases, but trajectory error and position vibration are produced
Solution Approach 1:
The patent implements feedback control by continuously detecting actual position and trajectory using gyro sensors during high-speed operation, calculating errors on each axis, and applying correction amounts to the control commands. This closed-loop feedback enables the robot to maintain high speed while correcting trajectory deviations and reducing vibration in real-time.
Solution Approach 2:
The learning control function enables the robot to automatically learn and store correction amounts for trajectory and position errors that occur during high-speed operation. The system self-corrects by applying these learned correction amounts in subsequent operations, allowing sustained high-speed operation with maintained precision without continuous manual intervention.
3Measurement precision
If sensor is mounted directly on the arm to measure trajectory error, then measurement accuracy improves, but system cost and complexity increase
Solution Approach 1:
The patent replaces complex acceleration sensor systems with simpler gyro sensor systems. The gyro sensors directly measure angular velocity, which can be integrated to obtain position information, eliminating the need for complex mathematical operations and additional processing components required by acceleration sensors, thus reducing system complexity while maintaining measurement precision.
Data Source
AI summary
A robot (100) has a robot mechanism unit (1) having a sensor (10) and a control unit (2), and the control unit (2) includes a normal control unit (4) that controls the operation of the robot mechanism unit, and a learning control unit (3) that, when the robot mechanism unit (1) is operated by a speed command that is given by multiplying a teaching speed designated in a task program by a speed change ratio, performs learning to calculate, from a detection result by the sensor (10), a learning correction amount for making the trajectory or position of the control target in the robot mechanism unit (1) approach the target trajectory or target position, or for reducing the vibration of the control target, and performs processes so that the control target position of the robot mechanism unit (1) moves along a fixed trajectory regardless of the speed change ratio.