Work

Approximate Optimized Backstepping Control of Uncertain Fractional-Order Nonlinear Systems Based on Reinforcement Learning