DESIGN AND DEVELOPMENT OF CABLE-DRIVEN PARALLEL ROBOT FOR ADDITIVE CONSTRUCTION

Abstract

Conventional construction methods are slow, over-budget, hazardous to laborers and environment, lack productivity, and provide limited architectural freedom in design. Automation in construction has a wide potential of overcoming the drawbacks of conventional construction methods. The literature reveals various engineering solutions to automate construction. Cable-driven parallel robots (CDPR’s) are easy to transport, assemble and disassemble with the advantage of being lightweight and relatively inexpensive. Therefore, this thesis focuses on the design and development a CDPR for additive construction. Since the application of additive construction requires high stiffness to external disturbances, the emphasis of this thesis is on the over-constrained redundant CDPR. In addition, the thesis focuses on the design of large scale CDPR for onsite construction. For large-scale CDPR, cable mass and elasticity are non-negligible and hence, cannot be ignored during the analysis of the system behavior. The inverse kinematic problem of CDPR becomes a non-linear kineto-static problem when catenary cables (i.e., cables with mass and elasticity) are taken into account. This kineto-static problem has been extensively studied for minimally constrained CDPRs. However, a limited amount of work has been directed towards redundant CDPRs. This is because, for redundant CDPR, the problem is relatively complex, with the possibility of a large number of multiple kineto-static solutions. Therefore, the initial work in the thesis emphases on investigating various selection criteria to determine a unique kineto-static solution for the redundant over-constrained CDPR. Since kineto-static problem considering catenary cable model is computationally intensive, a rigorous study is done to determine when a massless rigid/elastic cable model can be used as an alternative to the catenary cable model. Furthermore, a neural network-based approach to proposed to rapidly obtain the inverse kineto-static (IKS) as well as the forward kineto-static (FKS) solution of CDPR with significant cable mass and elasticity. Further research work in this thesis focused on the design of CDPR for additive construction. For this, a methodology to determine the printable workspace of a cable robot is proposed by taking cable mass and mobile platform orientation into account. This workspace has a characteristic feature of avoiding the collision of lower cables with the previous layers of the structure during 3D printing while maintaining cable tensions inside the prescribed positive tension bounds. Since redundant robot is taken into consideration, the printable workspace possesses multiple solutions and hence, a unique solution needs to be selected based upon a selection criterion. Therefore, to select a unique solution, a selection criterion to minimize cable sagging is proposed and compared against the conventional selection criterion of minimizing cable tension. Finally, an optimal design approach is proposed using genetic algorithm for the desired printable workspace with an objective to minimize the installation space of the cable robot. After analyzing the kineto-static model, a comprehensive dynamic model of the CDPR is proposed which considers the cable mass, elasticity, flexural stiffness, flexural damping as well as cable-pulley interactions for the dynamic modeling. The modeling of cable is done by considering the cable as a series of spring-mass-damper systems connected by the revolute joints. The bond graph approach is utilized for modeling and simulation of the proposed dynamic model. Subsequent to kinematic and dynamic analysis, the thesis presented a novel concept of self-reconfiguration to improve the pintable workspace of the robot while maintaining an over constrained configuration of CDPR. The reconfiguration is implemented to prevent the interference of the lower cables with the structure being printed. The novel feature of the proposed robot is that the reconfiguration is achieved using cable actuators along with locking/unlocking mechanism/actuators only; thereby avoiding the use of additional heavy actuators to lift the platforms containing lower anchor points and reconfigure the robot. The dynamic model and model based control of the proposed concept is presented. Furthermore, the trajectory planning for concrete printing a structure is discussed, where an optimal trajectory is designed for reconfiguration of the robot. The proposed concept is validated by dynamic simulations. Finally, the detailed steps for the cost-effective design and development of CDPR is presented for additive construction. A screw extruder is fitted on the mobile platform for extruding concrete/clay. The controller design for the whole system is detailed and the measures to obtain the trajectory for printing are discussed. Thereafter, the cost analysis of the whole robot is performed. Lastly, the experimental results for clay/concrete printing are presented to validate the accuracy of the CDPR.