All news

Safe reinforcement learning combined with MPC and control barrier functions

The PECTS framework learns stochastic system dynamics with probabilistic neural networks and control barrier functions with Lipschitz-bounded networks. Learned barrier constraints are embedded in a model predictive control formulation that accounts for model uncertainty. This provides probabilistic safety when learning under unknown and stochastic dynamics.

Why it matters

Combining learned models with safety guarantees is a key precondition for using adaptive control on real assets.

Relevant to DT products

Original headline
A Control Barrier Function-Constrained Model Predictive Control Framework for Safe Reinforcement Learning
Read the original: arXiv — Systems and Control (eess.SY)

Translation and summary are machine-generated from the source. The full article is not republished; its rights belong to the source.