Design, Modeling, Self-Calibration and Grasping Method for Modular Cable-Driven Parallel Robots

Abstract Compared with traditional rigid serial robots, cable-driven parallel robots (CDPRs) have the advantages of simple structure, lightweight, large workspace, strong flexibility, fast reconstruction speed, and low construction cost. However, there are still many problems in the optimal design of the mechanism, system calibration, and trajectory planning of the CDPR. Based on these, a modular cable-driven parallel robot (MCDPR) is designed and the nonlinear kinematic modeling, system self-calibration, and grasping planning methods are proposed. Firstly, according to the application requirements of MCDPRs in large space scenarios, the design indicators and performance requirements of MCDPRs are analyzed and an MCDPR is designed. Secondly, the "motor-cable-end" multi-layer kinematic equations of the MCDPR are derived. Further, the system is calibrated using a vision sensor before the motion of the mobile platform. A vision-based self-calibration method for mechanism parameters of the MCDPR is proposed to reduce the number of calibrations during robot operation. Thirdly, a grasping planning method based on visual measurement is proposed for autonomous object grasping. Finally, a software/hardware combination MCDPR experimental prototype with encoders and tension sensors is built to verify the above-designed prototype and the proposed methods. Kinematic model calibration experiments, system self-calibration experiments, and object grasping control experiments are carried out. The simulation based on the CopperliaSim software and practical experiments prove the effectiveness of the designed structure and the proposed method.