Optimal Location of Robot Base with Respect to the Application Positions by Using Proper Neural-Network Method

Abstract

Usual the location of robot base with respect to positions (configurations) of an application that the robot must reach is choose in such a way that the application points to be into working space of robot by avoiding the obstacles. Or the application is build from the very first beginning in such a way that all application points to be in the working space of robot because usual the robot base is fixed to the ground. But this is not the optimal solution with respect to an objective function which represents for example minimum time of motion during a cycle, or minimum consumption of energy, or maximum precision, or combination of these. Some objective functions could results from the specificity of the application like is the case of casting of forging where the accumulation of heat for example could be one of the optimization criteria. For example in the automotive industry the owners prefer to replace the whole robotized line when the product is changed instead of reprogramming robots because the prices of robots is decreasing and the price of reprogramming is increasing. For such a situation the placing of robot base in an optimum location from the very first beginning so that the time or/and energy consumption to be minimum is an essential initial task, especially for large series productions. The proposed paper is dealing with the subject of moving the base of robot with respect to the application points so that an objective function representing the minimum time of motion during a cycle to be fulfilled.