Initial-Pose Self-calibration for Deployable Overconstrained Cable-Driven Parallel Robots

Deployable Cable-Driven Parallel Robots (CDPRs) are often designed to be lightweight so that they can be easily transported and mounted in the field. They often rely on simplified cable transmission systems with non-constant and uncertain transmission ratios. These uncertainties complicate the kinematic model, making pose estimation and calibration challenging, as motor angles are not easily linked to cable lengths. This paper presents a comparative study of three self-calibration algorithms for deployable CDPRs, analyzing the impact of different sensor configurations on initial-pose estimation accuracy. Simulations evaluate each sensor set using position and orientation errors as metrics. Recommendations are proposed on optimal sensor configurations for deployable CDPRs, accounting for measurement and modeling errors and the practical limitations of simplified transmission systems.