Actuator Control Parameter Tuning Using Predicted Probability Distributions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing actuator control systems face challenges in optimally adjusting parameters to achieve precise control of actuators, particularly in adapting to different actuators and ensuring reliable operation, especially when parameters are not well adapted.

Innovation Solution

A method for automatically setting actuator control system parameters using a long-term cost function based on predicted temporal evolution of the probability distribution of the control variable, employing a Gaussian process model for prediction and numerical quadrature for approximation, with adaptive model adjustments and episodic training to improve the control strategy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional control parameter adjustment methods are used, then the control system can operate with simple structure, but the adaptation precision to different actuators deteriorates

Engineering Contradiction:
Improvecontrol precisionVSAvoidcontrol system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the parameter representation from fixed control parameters to probability distribution parameters (mean and standard deviation) that can adapt to different actuators. By parameterizing the control approach through distribution characteristics rather than fixed values, the system achieves better adaptation precision while maintaining manageable complexity through statistical modeling.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent introduces a probability distribution as an intermediary between the control system and the actuator. Instead of directly controlling the actuator with fixed parameters, the system uses probability distributions to mediate the control signals, enabling adaptive control that accounts for uncertainties and variations in actuator behavior without requiring complex actuator-specific tuning.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If actuator-specific parameter adaptation is implemented, then the control precision improves, but the adjustment complexity increases

Engineering Contradiction:
Improveactuator operation reliabilityVSAvoidparameter adjustment complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent transforms actuator-specific parameters into probability distribution parameters (mean and standard deviation) that capture the statistical characteristics of actuator behavior. This parameter transformation allows the system to adapt to different actuators by learning their distribution characteristics rather than manually tuning complex actuator-specific parameters, thereby improving reliability while reducing adjustment complexity.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The control system performs self-adaptation by automatically learning the probability distribution parameters from actuator behavior data. Instead of requiring manual adjustment of actuator-specific parameters, the system serves itself by autonomously adapting to different actuators through statistical learning, which improves operational reliability without increasing user-facing complexity.

Inventive Principle:
Principle #25Self-service

3Adaptability or versatility

If fixed control parameters are used, then the system structure remains simple, but the adaptability to different actuators deteriorates

Engineering Contradiction:
Improveactuator adaptabilityVSAvoidcontrol system complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent changes from fixed control parameters to adaptive probability distribution parameters that can be learned and adjusted for different actuators. By representing control knowledge statistically through distributions rather than fixed values, the system gains versatility to work with different actuator types while maintaining structural simplicity through the unified probabilistic framework.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The probability distribution-based control approach serves as a universal interface that can adapt to multiple actuator types and scenarios. The same probabilistic control framework can be applied across different actuators by learning their specific distribution characteristics, providing multi-functionality and broad adaptability without requiring separate control systems for each actuator type.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3646122B1Method and device for setting at least one parameter of an actuator control system and actuator control system
Publication Date: 2023.09.13 ROBERT BOSCH GMBH
  • EP3646122B1 patent drawingFigure 1
  • EP3646122B1 patent drawingFigure 2
  • EP3646122B1 patent drawingFigure 3

AI summary

The invention relates to a method for automatically setting at least one parameter (Θ) of an actuator control system (45) which is designed to control a control variable (x) of an actuator (20) to a predefinable setpoint (xd). The actuator control system (45) is designed to generate a control variable (u) depending on the at least one parameter (Θ), the setpoint (xd) and the control variable (x) and to actuate the actuator (20) depending on this control variable (u). A new value (Θ*) of the at least one parameter (Θ) is selected depending on a long-term cost function (R), wherein this long-term cost function is determined depending on a predicted temporal evolution (F) of a probability distribution (p) of the control variable (x) of the actuator (20) and the parameter (Θ) is then set to this new value (Θ*).