Kinematic modeling of PAM-based 3-UPS parallel robot module using string encoders

Abstract

This paper presents a kinematic modeling of a 3-UPS (universal–prismatic–spherical) parallel robot module actuated by three PAMs (pneumatic artificial muscles). The soft robots using the PAMs have benefits in terms of the safe interaction between humans and robots due to their intrinsic safety, but they are not often utilized in practice. This is because soft robots using the PAMs are not as controllable as rigid robots when it comes to performing some tasks. While the existing modeling approaches concentrate on perceiving the shapes and motions of the soft robot itself, the approach presented in this paper focuses on finding the task position of the robot regardless of its shape. This method computes the task position of the soft robot with forward kinematics and obtains the lengths of the three PAM actuators equipped inside the soft robot with inverse kinematics. A small string encoder sensor is developed and calibrated to measure the actuators’ lengths required during the kinematic modeling process. Experimental results verify that the task position of the soft robot can be obtained accurately from forward kinematics, and ultimately, we show that precise position control is possible using inverse kinematics. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.