In this paper, an online tracking system has been developed to control the arm and head of a Nao robot using Kinect sensor. The main goal of this work is to achieve that the robot is able to follow the motion of a human user in real time to track. This objective has been achieved using a RGB-D camera (Kinect v2) and a Nao robot, which is a humanoid robot with 5 degree of freedom (DOF) for each arm. The joint motions of the operator's head and arm in the real world captured by a Kinect camera can be transferred into the workspace mathematically via forward and inverse kinematics, realitically through data based UDP connection between the robot and Kinect sensor. The satisfactory performance of the proposed approaches have been achieved, which is shown in experimental results.
|Title of host publication||2016 International Conference on Advanced Robotics and Mechatronics (ICARM)|
|Publication status||Published - 27 Oct 2016|
|Event||2016 International Conference on Advanced Robotics and Mechatronics - Macau, China|
Duration: 18 Aug 2016 → 20 Aug 2016
|Conference||2016 International Conference on Advanced Robotics and Mechatronics|
|Period||18/08/16 → 20/08/16|