An MPC-based Pose Control Framework of Multi-Section Soft Manipulator using Hybrid Jacobian Estimation.

Soft robotic manipulation is currently one of the most active problems in robotics, but simultaneously controlling the position and orientation of the end remains an open research problem for the soft manipulator. In this paper, a novel pose control framework based on model prediction control (MPC) is presented for the multi-section soft manipulator. Firstly, the continuum kinematics is established based on the piecewise constant curvature (PCC) model. Secondly, a hybrid method is proposed to estimate the Jacobian matrix in real time, which combines analytical and numerical methods. Then, an exponentially weighted MPC controller is developed to regulate the end pose in closed loop. Finally, simulations are carried out to verify the feasibility and effectiveness of the Jacobian matrix estimation method and pose control. The simulation results indicate that our controller performs better than peer works.