Synthesizing Neural Network Controllers with Closed-Loop Dissipativity Guarantees

ArXi:2404.07373v2 Announce Type: replace-cross This paper presents a method to synthesize neural network controllers to maximize reward subject to the hard constraint that the feedback system of plant and controller be dissipative, certifying requirements such as stability and $L_2$ gain bounds. It considers nonlinear and uncertain plants, modeled as the interconnection of a linear time-invariant (LTI) system and an uncertainty block, which incorporates nonlinearities.