Motion Planning Method of Continuum Manipulator Based on Multi Objective Optimization

This study proposes an innovative algorithm integrating kinematic modeling and intelligent optimization to address the motion planning of multi-segment cable-driven continuum manipulator. Under the piecewise constant curvature (PCC) hypothesis, a forward kinematic framework is formulated to define a structured mapping between joint-space parameters and task-space coordinates. Leveraging the established model, a refined APF strategy is implemented to synthesize collisionavoidance paths for the manipulator, while the Jacobian pseudo-inverse method is applied to obtain corresponding general joint angle solutions. Moreover, to enhance optimization efficiency, the initial population is generated by leveraging the least-squares solution as a baseline, with search boundaries dynamically defined through the null-space properties of the Jacobian pseudoinverse. Then the trajectory optimization is executed under dual constraints of end-effector pose and arm shape via the cheetah optimizer (CO). Numerical simulations validate the algorithm's capability to maintain trajectory tracking accuracy while accounting for end-effector pose, providing an effective solution for motion planning of multi-segment continuum manipulators.