Robot Control Device Vibration Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot control devices face challenges in robustly suppressing vibration in multi-link robots due to the use of approximative observers based on linear models, which result in non-robust vibration suppression and high engineering costs, especially when the end effector load is small or has low inertia.
Innovation Solution
A robot control device that employs a nonlinear dynamic model to estimate angular velocities and accelerations, using a feedback system with proportional-integration control and a feedback constant calculating unit to compensate for low inertia, allowing for precise control and vibration suppression without the need for extensive engineering adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If an approximative observer based on a one-input/output linear model is used to reduce calculation amount, then the calculation complexity is reduced, but the vibration suppression effect becomes non-robust and engineering costs increase
Solution Approach 1:
The patent replaces the complex approximative observer based on linear models with a simplified observer that directly utilizes angular velocity information from sensors. This substitution maintains vibration suppression effectiveness while reducing calculation complexity and eliminating the need for complex dynamic model-based observers.
Solution Approach 2:
The system utilizes the angular velocity sensor mounted on each link to directly provide the necessary feedback signal for vibration suppression. The sensor data is used self-sufficiently to calculate axis torsion angular velocities without requiring complex intermediate observers or approximations, thereby simplifying the control system.
2Reliability
If angular velocity sensors are mounted on each link to directly measure axis torsion angular velocities, then vibration suppression performance is improved, but the number of wirings increases and cost increases
Solution Approach 1:
The patent combines the angular velocity sensor functionality with the existing motor encoder system. By utilizing the angular velocity information already available from the motor side and processing it through the observer, the system achieves direct measurement capability without requiring separate sensor installations and additional wiring for each link.
3Reliability
If a nonlinear observer based on nonlinear dynamic model is used to achieve robust vibration suppression, then vibration suppression robustness is improved, but the amount of calculation increases and engineering costs increase
Solution Approach 1:
The patent replaces the computationally intensive nonlinear observer based on dynamic models with a simpler observer that directly processes angular velocity sensor data. This substitution maintains the robust vibration suppression performance while significantly reducing the calculation burden during control operations.
Solution Approach 2:
The patent extracts only the essential angular velocity information needed for vibration suppression from the available sensor data, rather than using a comprehensive nonlinear observer that processes all dynamic model parameters. This extraction approach maintains effectiveness while reducing computational complexity.
Data Source
AI summary
A robot control device according to an embodiment includes: an observer receiving the angular velocity of the motor and the current command value, and estimating an angular acceleration of the link, and angular velocities of the link and the motor from a simulation model of an angular velocity control system of the motor; a first feedback unit calculating an axis torsion angular velocity from a difference between the angular velocities of the link and the motor estimated by the observer, and giving feedback to the angular velocity control system; a second feedback unit feeding back the angular acceleration of the link estimated by the observer to the angular velocity control system; and a first feedback constant calculating unit compensating an end effector load mass and increases inertia at the second feedback unit when an end effector load in the nonlinear dynamic model has low inertia.