Hybrid multi-objective motion planning of parallel kinematic machines

In this paper we consider the problem of multi-objective trajectory planning to Parallel Kinematic Machines (PKMs). A two stage system is developed. In a first stage is an offline planning based on robot kinematics and dynamics, including actuators, is performed to generate a large dataset of trajectories, these trajectory cover mostly of the robot workspace and minimize time and energy, while avoiding singularities and limits on joint angles, rates, accelerations and torques. An augmented Lagrangian decoupling to solve the resulting non-linear constrained optimal control problem. The offline-planning outcomes are then used to build a data-driven neuro-fuzzy inference system to learn and capture the desired dynamic behavior of the PKM. Once this system is trained, it is used to achieve near-optimal online planning with a reasonable time complexity. Simulations proving the effectiveness of this approach on a 2-degrees of freedom planar PKM are given and discussed.