Solving the Forward Kinematics of Cable-Driven Parallel Robots with Neural Networks and Interval Arithmetic

This paper investigates a new approach for solving the forward kinematics of cable-driven parallel robots. This approach combines an interval algorithm with neural networks to provide a fast but accurate initial guess. The neural networks increase the computation speed by a factor of 200 or more, while the interval algorithm provides guaranteed convergence and a definite solution to any chosen degree of accuracy. Iterative techniques are faster still, but the proposed algorithm is considered real-time feasible.