Graph-Based Simultaneous Localization and Calibration for Planar Cable-Driven Parallel Robots

Planar Cable-Driven Parallel Robots (CDPRs) offer significant advantages in various applications, but their deployment has been limited due to challenges in setup and calibration. In this paper, we address these challenges by fusing visual state estimates with joint measurements, resulting in a unified framework for both calibration and localization. This is achieved through a unified graph-based formulation that integrates these processes, allowing for the simultaneous identification of kinematic parameters and end-effector state estimations. This method not only simplifies the calibration process but also enhances localization accuracy. The practical effectiveness and precision of the proposed approach are demonstrated through experiments on the Kamal-ol-Molk Planar CDPR.