3D Model Identification of a Soft Robotic Neck

Soft robotics is becoming an emerging solution to many of the problems in robotics, such as weight, cost and human interaction. In order overcome problems, bio-inspired designs have introduced new actuators, links architectures. However, complexity required models for control has increased dramatically geometrical model approaches, widely used rigid dynamics, are not enough these hardware types. this paper, different linear non-linear will be a soft neck consisting central link actuated by three motor-driven tendons. By combining force on tendons, able perform motion similar that neck. simplify modeling, first system input–output redefinition proposed, considering pitch roll angles outputs tendon lengths inputs. Later, two identification strategies selected adapted our case: set membership, data-driven, nonlinear non-parametric strategy which needs no input redefinition; Recursive least-squares (RLS), recognized technique. The method offers possibility modeling complex dynamics without specific knowledge its mathematical representation. selection was done possible extension more fact impact yet studied according current literature. On other hand, RLS shows implication using parametric plant, also helps evaluate degree nonlinearity comparing performances. addition methods, neural network comparison purposes. obtained results validate approaches proposed.