Modeling of Forward/Inverse Kinematics and Closed-Loop Control for Soft Robotic Arms

Soft robotic arms offer a novel and secure technical approach for assisting astronauts in space stations, owing to their unique advantages such as flexibility and safety in human–machine interaction. During astronaut operational assistance, the forward and inverse kinematics models serve as the theoretical foundation for achieving interactive control. Typically, forward kinematics calculates the end position based on control inputs, while inverse kinematics determines the necessary inputs to achieve a desired position. Existing methods encounter challenges due to the large deformations and redundancies of the arms, which can lead to numerical instability or high data requirements. This study presents an innovative kinematic approach specifically designed for pneumatic soft arms. The forward model employs a fitting function, while the inverse kinematics is formulated as a constrained optimization problem that minimizes the deviation of the end position from the target. Experiments demonstrate that the forward fitting method is effective for both single and multichamber actuators. The inverse optimization substantially enhances control robustness and ensures motion convergence. This algorithm serves as a foundation for closed-loop feedback control, facilitating precise positioning. Demonstrations indicate that the proposed method enables rapid positioning through a single optimization calculation and achieves precise positioning with closed-loop control based on the optimization, offering promising extensions to other types of pneumatic robotic arms. The forward and inverse kinematics algorithm proposed in this paper provides a theoretical basis for the application of soft robotic arms in space operation assistive missions.