Session: MSNDC-01-02 Computational Methods and Software Tools in Multibody Systems and Nonlinear Dynamics

Paper Number: 74607

Start Time: August 18, 01:00 PM

74607 - Multi-Objetive Cycle Optimisation of a Three Degrees of Freedom Robotic System 

Production improvements and cost reductions in modern manufacturing relies in a massive deployment of robotic systems. The interaction of these manufacturing systems with continuously adaptive manufacturing technologies is a key factor for ensuring competitive improvements and the optimisation of using the power resources. One of the persistent drawbacks in newest manufacturing options using robots are the limits of adapting programmed sequences to different products and work sequences [1-3]. The development of flexible robotic manufacturing can be considered the core improvement of the next-step in robotic massive deployment keeping efficiency as the base line. The efficiency, in a more general scenario, is also obtained by improving the basic parameters such as the electrical consumption and work cycle time. This is one key component when adapting each cycle to the individual product/case in the manufacturing line.
This work analyses the kinematic behaviour of a three-DoF robotic arm, evaluating the power consumption to optimize that value considering the trajectory and cycle case. The arm is modelled in MATLAB, considering its foward kinematics using Denavit-Hartenberg equations. In order to solve this optimisation problem, a multi-objective genetic algorithm is used, which optimise the trajectories, comparing and discarding those farthest to our objectives. The cases are defined after defining the starting and end points of the trajectory to be analysed. Subsequently, several intermediate points are freely proposed to define and analyse the trajectory in one plane. The trajectory is then approximated by some mathematical functions, polynomial or splines (Figure-1) and the analysis parameters are evaluated: deviation from the target points, energy consumption and cycle-time. The genetic algorithm, also defined in MATLAB and coupled to the arm model, evaluates the optimization function considering the three goals of minimizing the three parameters according to specific weighting functions to evaluate the arm efficiency.
This type of study shows the capabilities of an optimization analysis to define the best choice for defining an individual cycle considering several aspects of the process optimization. The optimization is fast and robust. It can provide, on the one hand, the analysis of multiple cases to fast adapt to multiple manufacturing line situations. On the other hand, the method can be applied to any other 'arms' with different number of DoF and any other complex motions.

Presenting Author: Rodrigo Randulfe López Universidade de Vigo

Authors:

Rodrigo Randulfe López Universidade de Vigo
Marcos López Lago Universidade de Vigo
Enrique Paz Domonte Universidade de Vigo
Jacobo González Baldonedo Universidade de Vigo
José Ángel López Campos Universidade de Vigo
Abraham Segade Robleda Universidade de Vigo
Enrique Casarejos Ruiz Universidade de Vigo