Eventos Anais de eventos
COBEM 2021
26th International Congress of Mechanical Engineering
Grasp synthesis for a spatial 6 DoF gripper based on its wrench capability.
Submission Author:
Leonardo Mejia Rincon , SC
Co-Authors:
Leonardo Mejia Rincon, Daniel Ponce
Presenter: Leonardo Mejia Rincon
doi://10.26678/ABCM.COBEM2021.COB2021-1193
Abstract
In the last decades, the use of robots has migrated from the traditional industry to highly complex areas of knowledge such as the nuclear industry, the aerospace industry and ocean exploration. The insertion of robotics in these areas of knowledge has increasingly demanded the development of autonomous robots that can act in unpredictable environments and that can grasp objects with different physical characteristics, becoming one of the main challenges of robotics, artificial intelligence and cognitive sciences. In this work, a method of grasping synthesis is proposed for a robotic hand with two fingers and six degrees of freedom. Grasping synthesis basically consists of discovering where and how to place the fingers on the surface of an object that a robot needs to manipulate. This process seeks to find the best contact position between the fingers of the robotic hand and an object to be manipulated at the same time that external loads to the object are balanced and the best posture of the robotic hand is guaranteed in order to perform a previously defined task. The proposed method in this work presents an algorithmic solution to the grasping problem by optimizing two intrinsic characteristics to the process. The first characteristic and its corresponding algorithm, seek to guarantee the static stability criteria by forcing the sum of forces and moments found in the contact interface to be equal to zero. The optimization of this characteristic together with a set of restrictions using a differential evolution algorithm (DE), results in contact points containing optimized forces and moments to be applied to the object through the fingers of the robotic hand. The second characteristic and its corresponding algorithm, minimize the energy consumption of the actuated joints of the robotic hand, resulting in a more adequate position and orientation in the space according to the handle requirements. These requirements are nothing more than the points of contact, forces and moments calculated as a result of the optimization algorithm of the first characteristic. In comparison with the existing approaches in the literature, the method proposed here differs in the fact that it deals with the gasping synthesis from the definition of the tasks to be performed in conjunction with the achievement of ideal closing forces. The proposed method was validated using specific grasping synthesis algorithms developed for a robotic hand with 6 degrees of freedom and which were implemented and improved using MATLAB software.
Keywords
Grasping Synthesis., optimization., Wrench Capability., Davies' Method
