This paper presents a novel combined State Dependent Riccati Equation (SDRE) / Function Approximation Technique (FAT)-based control design for nonlinear uncertain systems. The SDRE is employed to construct an optimal controller and the function approximation technique is utilized to estimate time-varying disturbances and uncertainties. Moreover, a robust term in the proposed control law compensates for the truncation error. The closed-loop stability and boundedness of the tracking error and FAT weights approximation error are proved in the sense of Lyapunov, with consideration of truncation error. Due to the great importance of the adequate performance of transient response from practical point of view, performance evaluation has been accomplished. The proposed scheme is computationally simple due to utilizing the FAT to represent uncertainties and disturbances as a function of time. Compared with the SDRE based SMC, the proposed controller is superior in terms of capability to track a fast and highly complicated trajectory and no need to determine time-varying disturbances and uncertainties bounds. The case study is a Selective Compliant Articulated Robot for Assembly (SCARA) flexible joint manipulator as a representative of highly nonlinear, coupled, large robotic systems. Simulation results easily verify the effectiveness and superiority of the proposed controller.

本文提出了一种新的基于状态相关 Riccati 方程 (SDRE) / 函数逼近技术 (FAT) 的非线性不确定系统控制设计。SDRE 用于构建最优控制器，函数逼近技术用于估计时变扰动和不确定性。此外，所提出的控制律中的鲁棒项补偿了截断误差。在考虑截断误差的情况下，证明了Lyapunov意义上的跟踪误差和FAT权重近似误差的闭环稳定性和有界性。由于从实际的角度来看，足够的瞬态响应性能非常重要，因此已经完成了性能评估。由于利用 FAT 将不确定性和干扰表示为时间的函数，因此所提出的方案在计算上很简单。与基于 SDRE 的 SMC 相比，所提出的控制器在跟踪快速且高度复杂的轨迹的能力方面具有优越性，并且无需确定时变扰动和不确定性界限。案例研究是一个选择性兼容的关节式装配机器人 (SCARA) 柔性关节机械手，它是高度非线性、耦合的大型机器人系统的代表。仿真结果很容易验证所提出的控制器的有效性和优越性。所提出的控制器在跟踪快速且高度复杂的轨迹的能力方面具有优越性，并且无需确定时变干扰和不确定性界限。案例研究是一个选择性兼容的关节式装配机器人 (SCARA) 柔性关节机械手，它是高度非线性、耦合的大型机器人系统的代表。仿真结果很容易验证所提出的控制器的有效性和优越性。所提出的控制器在跟踪快速且高度复杂的轨迹的能力方面具有优越性，并且无需确定时变干扰和不确定性界限。案例研究是一个选择性兼容的关节式装配机器人 (SCARA) 柔性关节机械手，它是高度非线性、耦合的大型机器人系统的代表。仿真结果很容易验证所提出的控制器的有效性和优越性。