Load maximization of flexible joint mechanical manipulator using nonlinear optimal controller

In this paper, a closed loop nonlinear optimal control approach is investigated for flexible joint manipulators (FJM). The dynamic load carrying capacity (DLCC) of these manipulators is obtained via this approach. The state-dependent Riccati equation (SDRE) technique is used for solving nonlinear optimal control problem. This method uses special parameterization to develop the nonlinear system to a linear structure having state-dependent coefficient matrices. The Taylor series numerical method is addressed by approximating the solution to the SDRE. Simulations for FJM are provided to illustrate the effectiveness of this approach for designing nonlinear feedback controllers and computing DLCC of these manipulators. Also effects of change in spring factors of manipulator and control parameters on resulted trajectory of manipulator are considered. DLCC is calculated subject to the limits in actuators and tracking accuracy. Finally the results are checked by comparison with experimental results.

► The main innovation of this paper is to determine DLCC of FJM using closed loop nonlinear optimal control.
► Dynamic modeling of flexible joint manipulator is presented.
► The implementation of method and simulation results of FJR for point to point and trajectory tracking task are presented and DLCC of manipulator is determined.
► Also different trajectories are obtained subject to different values of control parameters and spring factors.
► Finally, the results are compared with experimental test for a robot named Scout.