The solution of the inverse kinematics problem of soft manipulators is essential to generate paths in the task space. The inverse kinematics problem of constant curvature or piecewise constant curvature manipulators has already been solved by using different methods, which include closed-form analytical approaches and iterative methods based on the Jacobian method. On the other hand, the inverse kinematics problem of nonconstant curvature manipulators remains unsolved. This study represents one of the first attempts in this direction. It presents both a modelbased method and a supervised learning method to solve the inverse statics of nonconstant curvature soft manipulators. In particular, a Jacobian-based method and a feedforward neural network are chosen and tested experimentally. A comparative analysis has been conducted in terms of accuracy and computational time.

Control and modelling of continuum robots are challenging tasks for robotic researchers. Most works on modelling are limited to piecewise constant curvature. In many cases they neglect to model the actuators or avoid a continuum approach. In particular, in the latter case this leads to a complex model hardly implemented. In this work, a geometrically exact steady-state model of a tendon-driven manipulator inspired by the octopus arm is presented. It takes a continuum approach, fast enough to be implemented in the control law, and includes a model of the actuation system. The model was experimentally validated and the results are reported. In conclusion, the model presented can be used as a tool for mechanical design of continuum tendon-driven manipulators, for planning control strategies or as internal model in an embedded system.

No abstract