Servo Control Architecture With Split MPC and Feedback Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high operation load of model predictive control in servo control systems makes it difficult to achieve precise and stable control due to the processor's limited processing capability, leading to deviations from the command trajectory and instability in feedback systems.
Innovation Solution
A control system is designed with separate processors for model predictive control and feedback control, allowing for distinct operation periods and processes to distribute the load and improve processing capability, thereby enhancing control accuracy and stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If model predictive control is executed in a single processor with high operation load, then control accuracy can be maintained, but processing capability becomes insufficient leading to trajectory deviation
Solution Approach 1:
The control system is segmented into two separate processors: a first processor dedicated to model predictive control operations and a second processor dedicated to feedback control operations. This segmentation distributes the high operation load across multiple processing units, preventing any single processor from becoming a bottleneck while maintaining both control accuracy and processing capability.
Solution Approach 2:
The first processor acts as an intermediary that performs model predictive control calculations and generates reference trajectories, which then serve as input for the second processor's feedback control. This intermediary structure allows complex MPC operations to be separated from real-time feedback operations, enabling each processor to be optimized for its specific function.
2Reliability
If the operation period is shortened to improve control stability, then control accuracy improves, but operation load increases making it difficult to execute model predictive control
Solution Approach 1:
The control operations are segmented into different time scales: the first processor executes model predictive control at a longer operation period suitable for computation-intensive tasks, while the second processor executes feedback control at a shorter operation period for rapid response. This segmentation allows the system to maintain stability without overwhelming any single processor.
Solution Approach 2:
The system dynamically adjusts operation periods based on the type of control operation. Model predictive control uses a longer, computationally manageable operation period, while feedback control uses a shorter period for rapid response. This dynamic approach allows the system to optimize between computational load and control responsiveness.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A control system for causing an output of a control target to follow a command includes: a first processing device which is a processing device having a first processor and a prediction model that defines a correlation between a state variable with respect to the predetermined control target and a control input to the predetermined control target in the form of a state equation, performs model predictive control using the first processor, and outputs a servo command corresponding to the control input at an initial time point of the prediction interval; and a second processing device which is a processing device having a second processor different from the first processor and a feedback system including controllers to which a feedback signal related to an operation of the predetermined control target is input and receiving the servo command from the first processing device, and performs feedback control using the second processor.