Model based open-loop and closed-loop control systems make use of the system's inertial parameters. Unfortunately, not all of these values can be determined analytically nor can they be obtained from simple measurements. Established experiments for inertial parameters estimation have been applied to serial and parallel rigid-link manipulators, yet in very few cases to cable-driven parallel robots. Due to their kinematic properties and their unique setup, cable robots are more sensitive to incorrect estimates of the inertial parameters making it important to obtain such quantities through experiments. In this work, we assess the topic of inertial parameter identification of a parallel flexible-link manipulator exemplified by the suspended cable-driven parallel robot CoGIRo. Identification equations are derived from Newton-Euler equations of motion of an arbitrary point fixed to a rigid-body. Laboratory experiments for identification of the inertial parameters are then introduced and results are presented. Within the limitations of the sensors and data acquisition methods, reasonable results have been obtained, thereby validating the procedure for suspended cable-driven parallel robots.

Cable-driven parallel robots (CDPR) are efficient manipulators able to carry heavy payloads across large workspaces. Therefore, the dynamic parameters such as the mobile platform mass and center of mass location may considerably vary. Without any adaption, the erroneous parametric estimate results in mismatch terms added to the closed-loop system, which may decrease the robot performances. In this paper, we introduce an adaptive dual-space motion control scheme for CDPR. The proposed method aims at increasing the robot tracking performances, while keeping all the cable tensed despite uncertainties and changes in the robot dynamic parameters. Reel-time experimental tests, performed on a large redundantly actuated CDPR prototype, validate the efficiency of the proposed control scheme. These results are compared to those obtained with a non-adaptive dual-space feedforward control scheme.