Reinforcement Learning Augmented Optimization for Smart Mobility
Roman Overko, Rodrigo Ordonez-Hurtado, et al.
CDC 2019
The technical note studies the problem of sliding mode control design for linear systems with incomplete and noisy measurements of the output and additive/multiplicative exogenous disturbances. First, we construct a linear minimax observer to have an estimate of the system's state with minimal worst-case error. Second, we establish the optimality of the constructed observer in the class of all observers represented by measurable functionals of the output. Finally, we propose an algorithm, generating continuous and discontinuous feedbacks, which steers the observer as close as possible to a given sliding hyperplane in finite time. The optimality (sub-optimality) of the designed feedbacks is proven for the case of bounded noises and additive (multiplicative) disturbances of L2-class. The efficacy of the proposed algorithm is illustrated by a numerical example.
Roman Overko, Rodrigo Ordonez-Hurtado, et al.
CDC 2019
Jonathan P. Epperlein, Orest V. Iftime, et al.
CDC 2018
Tigran Tchrakian, Sergiy Zhuk
SIAM Journal on Scientific Computing
Sergiy Zhuk, Tigran Tchrakian, et al.
SIAM Journal on Scientific Computing