Abstract: Focused on the trajectory tracking problem of continuum manipulator with multisource uncertainties is A linear active disturbance rejection control based on decoupled dualchannel is proposed to suppress the negative impact of uncertainty on tracking performance. Firstly, the decoupling of system is achieved via virtual control variable, and a dualchannel LADRC is designed for both the known and unknown decoupling rate. The linear extended state observer (LESO) is used to compensate for the system uncertainty in real time, and the parameter tuning method of LESO is given. The convergence is proved based on the Lyapunov stability theorem. When designing the simulation, the unknown decoupling rate, unmodeled dynamics, and unknown external interference are considered comprehensively. The results verify the effectiveness of the control method proposed in this article. Further comparing it to computed torque control (CTC), the results show that LADRC can handle a larger range of uncertainties and has stronger robustness, which provides a new idea for highprecision trajectory tracking of continuum manipulator.

Key words: continuum manipulator, strong coupled MIMO system, multisource uncertainties, virtual control variable, linear active disturbance rejection control