DESIGN AND IMPELEMENTATION OF ROBOTIC MANIPULATOR FOR HUMAN MOVEMENTS

number: 
3358
إنجليزية
Degree: 
Author: 
AHMED FADHIL SHANTA
Supervisor: 
Dr. Mohammed Zeki Al-Faiz
year: 
2014
Abstract:

Most human interactions with the environment depend on our ability to navigate freely and to use our hands and arms to manipulate objects. An ideal terface for Humanoid Robotic Operation (HRO) will be person-independent,inexpensive, easy to use, requires no wearable equipment, and requiring little or no user training at all. Humanoid Robotic Control (HRC) and interaction interface that uses depth images and skeletal tracking software to control Humanoid Robotic Arms (HRA).Solution for 4 Degree of Freedom (DOF) human arm inverse kinematic is proposed that tracks the end effecter in real time update. eveloping environment that detects, tracks, follows the user’s body and extracts 3D co-ordinates of the user’s both arms joints in real-time. Arduino microcontroller then transfers the data between both of computer and the HRA and gathers the feedback signal from the arms. Designing complete prototype of a humanoid robotic arms with 4DOF three joints in shoulder and one elbow joint to the wrist that looks like the human arm structure, appearance and action that represent the human arm movement performed by the humanoid robotic arm.Humanoid robotic arm is controlled by using real-time human pose recognition skeletal tracking methods, implemented by the use of a Microsoft Kinect sensor.It has been found that 3D HRC does not necessarily require the user toutilize any specialized devices attached to their bodies so as to perform 3D tracking of their movements. In addition, it is proven that a low cost system can be built using Kinect as input device to control the HRA. The designed HRA system was tested by user stands in front of a depth camera to initiate skeleton tracking, the initial location of the user automatically becomes the origin of the control coordinate system. The user can then use arm gestures to turn the robot's motors and control the behavior of the robot and the results of both percentage error (less than 4.6%) and time response (105 ms) generated between the user and HRA movements.