Motor Control Device Dynamic IIR Filter Coefficient Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motor control devices face challenges in effectively managing vibrations in industrial robots due to changing natural vibration frequencies with robot posture, leading to discrepancies in target positions during filter coefficient adjustments, which increase computational load and are difficult to implement for multi-axis control.
Innovation Solution
A motor control device employing a second-order infinite impulse response (IIR) filter with dynamically changing filter coefficients, ensuring the sum of input data equals the sum of output data without increasing CPU computation, by adjusting coefficients a1, a2, b0, b1, and b2 across sample periods to maintain accurate position control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If filter coefficients are dynamically adjusted according to robot posture to suppress vibration, then vibration suppression effectiveness is improved, but computational load increases
Solution Approach 1:
The patent dynamically changes filter coefficients (a1, a2, b0, b1, b2) based on robot posture parameters to adapt the cutoff frequency to the current natural vibration frequency. This allows effective vibration suppression across different postures without requiring complex real-time frequency analysis, thus improving reliability while controlling computational load.
Solution Approach 2:
The filter transitions from a static coefficient design to a dynamic one where coefficients are adjusted according to robot posture. This dynamic adaptation enables the filter to track changing natural vibration frequencies while maintaining a computationally efficient structure that doesn't significantly increase CPU load.
2Adaptability or versatility
If filter coefficients are changed during data processing, then adaptability to changing conditions is improved, but position accuracy deteriorates due to sum discrepancy
Solution Approach 1:
The patent incorporates a feedback mechanism that monitors the sum of input data and adjusts the filtered output to maintain sum equality. This feedback ensures that even when coefficients change dynamically, the integrated value (target position) remains accurate, preventing position drift and maintaining manufacturing precision.
Solution Approach 2:
The patent introduces an intermediary correction mechanism that acts between the filter processing and the final position control. This intermediary ensures that the sum of inputs equals the sum of outputs by adjusting for any discrepancies introduced by coefficient changes, thus maintaining position accuracy while allowing coefficient adaptability.
3Manufacturing precision
If complex filter processing is used to maintain position accuracy during coefficient changes, then position control precision is improved, but device complexity increases
Solution Approach 1:
The patent extracts the essential requirement for position accuracy (sum equality) from the complex filter processing and implements it as a separate, simple correction mechanism. This separates the adaptive coefficient adjustment from the position accuracy maintenance, allowing each to be implemented with minimal complexity while achieving both goals.
Solution Approach 2:
The patent segments the filter processing into two independent parts: coefficient adjustment based on posture and sum equality maintenance. This segmentation allows each part to be implemented with simple, dedicated logic rather than a single complex processing block, reducing overall device complexity while maintaining position accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A motor control device includes a movement command generator that outputs a movement command of a motor, a filter that performs filter processing of the movement command and outputs the filtered movement command, and a position controller that performs a position control of the motor. The filter is an infinite impulse response digital filter that is represented by a transfer function H(z) : H(z)=(b0+b1•z-1+b2•z-2)/{1-(a1•z-1+a2•z-2)} with filter coefficients a1, a2, b0, b1, and b2. The filter is configured to change the filter coefficient b0 in a first sample period, to change the filter coefficients a1 and b1 in a second sample period which follows the first sample period, and to change the filter coefficients a2 and b2 in a third sample period which follows the second sample period.