Energy-Efficient Operation of Linear Delta Robots with an Artificial Intelligence Model

A new artificial intelligence model of the linear Delta robot is developed and two problems of energy-efficient pick-and-place operations are solved in this work. The first one refers to the minimization of power consumed by robot drives in different tool positions upon various joint configurations. This problem is actual for such industrial processes as welding, packaging, quality control, or wrapping, in which the steady state placing and holding phases of the pick-and-place cycle continue much longer time than picking. The second problem concentrates on searching for the shortest path through all nodes considering all possible robot joint configurations at which the minimal overall energy is consumed. This solution is essential for processes that require the fastest picking and placing cycles, such as assembling, loading, or winding. The feasibility and applicability of the offered neural network tandem are validated experimentally on the Festo EXPT-45-E1 robot.