Iterative Learning Control-Informed Reinforcement Learning for Batch Process Control

Researchers introduce IL-CIRL, a framework combining Iterative Learning Control with Deep Reinforcement Learning to address safety risks and stability issues in industrial batch process control. The method uses Kalman filter-based state estimation to guide DRL agents toward safer, constraint-satisfying control policies.

• →Deep Reinforcement Learning faces safety risks due to stochastic uncertainty during exploration-exploitation phases.
• →Industrial practitioners avoid DRL methods due to lack of formal stability and convergence guarantees.
• →IL-CIRL framework combines established Iterative Learning Control with DRL for batch process optimization.
• →The approach incorporates Kalman filter-based state estimation to ensure operational constraints and stability.
• →The method enables systematic design of DRL controllers for batch processes under multiple disturbance conditions.