Abstract—One of the most important problems in robot modeling and control of cable robots is the problem of inverse kinematics, that is, using the coordinate data of the end effector to calculate the corresponding joint variables. Especially with regard to the kinematics of the cable robot, the sagging of the driven cable can have a significant effect on the calculation of the cable length, and this is more evident when the cable length is large, for example in construction or agriculture applications, where needs a large workspace. The determination of cable deflection considers modeling the cable as a chain model rather than calculating the cable length as a straight-line model. Furthermore, due to the structure and constraints of cable robots or Cable Driven Parallel robots (CDPRs), the system modeling and simulation becomes complicated, thereby increasing the computation time. In this paper, we propose an algorithm of Adaptive Neural Fuzzy Inference System (ANFIS) that is used to solve the cable sag problem for the 4-cable robots. This model was applied to estimate the cable sag for medium-sized cable robots with low travel speed and does not take into account the impact of cable elasticity. A simulation model was conducted and the results showed the advantages of this method in increasing the probability of convergence with small errors. The results of computation and experiment are analyzed to evaluate the effectiveness of the proposed model