ReACT: Reinforcement Learning for Controller Parametrization using B-Spline Geometries
2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC)(2024)
摘要
Robust and performant controllers are essential for industrial applications.
However, deriving controller parameters for complex and nonlinear systems is
challenging and time-consuming. To facilitate automatic controller
parametrization, this work presents a novel approach using deep reinforcement
learning (DRL) with N-dimensional B-spline geometries (BSGs). We focus on the
control of parameter-variant systems, a class of systems with complex behavior
which depends on the operating conditions. For this system class,
gain-scheduling control structures are widely used in applications across
industries due to well-known design principles. Facilitating the expensive
controller parametrization task regarding these control structures, we deploy
an DRL agent. Based on control system observations, the agent autonomously
decides how to adapt the controller parameters. We make the adaptation process
more efficient by introducing BSGs to map the controller parameters which may
depend on numerous operating conditions. To preprocess time-series data and
extract a fixed-length feature vector, we use a long short-term memory (LSTM)
neural networks. Furthermore, this work contributes actor regularizations that
are relevant to real-world environments which differ from training.
Accordingly, we apply dropout layer normalization to the actor and critic
networks of the truncated quantile critic (TQC) algorithm. To show our
approach's working principle and effectiveness, we train and evaluate the DRL
agent on the parametrization task of an industrial control structure with
parameter lookup tables.
更多查看译文
关键词
Control Parameters,Neural Network,Time Series,Complex Systems,Control System,Industrial Applications,Operating Conditions,Short-term Memory,Nonlinear Systems,Long Short-term Memory,Control Structure,Normalization Layer,Lookup Table,Actor Network,Deep Reinforcement Learning,Industrial Control,Memory Neural Network,Short-term Memory Neural Network,Expensive Task,Deep Reinforcement Learning Agent,Closed-loop System,Parametrized,Closed-loop Control System,System Dynamics,Deep Reinforcement Learning Approach,Discrete-time,B-spline Basis Functions,Reward Function,Parameter Space
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要