A model-based deep reinforcement learning method applied to finite-horizon optimal control of nonlinear control-affine system
Jong Woo Kim, Byung Jun Park, Haeun Yoo, Tae Hoon Oh, Jay H. Lee, Jong Min Lee
Topics & Concepts
Hamilton–Jacobi–Bellman equationOptimal controlReinforcement learningDynamic programmingComputer scienceMathematical optimizationBellman equationNonlinear systemArtificial neural networkAffine transformationMathematicsControl theory (sociology)Artificial intelligenceControl (management)PhysicsQuantum mechanicsPure mathematicsAdaptive Dynamic Programming ControlMechanical Circulatory Support DevicesReinforcement Learning in Robotics