Controlling a Humanoid Robot Arm for Grasping and Handling a Moving Object in 3-D Environment without Cameras

Abstract

Lot of researchers have worked on robotic grasping tasks on stationary or fixed objects, others have focused on objects in dynamic motion using cameras to record their images in order to treat them to estimate the position of the capture. This method is very difficult, requiring a lot of computing, image processing. Therefore, we should recall another simple method of handling. In addition, the majority of robotic arms available for humanoid control applications are complex and yet expensive. In this paper, we are going to detail the requirements to manipulating a humanoid robot arm with 7 degree-of-freedom to grasp and handle any moving objects in the 3-D environment in the presence or not of obstacles and without using the cameras. We used the OpenRAVE simulation environment and a robot arm equipped with the Barrett hand. We also describe a randomized planning algorithm capable of planning. This algorithm is an extension of RRT-JT that interleaves exploration using a Rapidly-exploring Random Tree with exploitation using Jacobian-based gradient descent to control a 7-DoF WAM robotic arm to avoid the obstacles, track a moving object, and grasp planning. We present results in which a moving mug is tracked, stably grasped with a maximum rate of success in a reasonable time and picked up by the Barret hand to a desired position.