Teleoperation control of Baxter robot using Kalman filter-based sensor fusion

C. Li, C. Yang, J. Wan, A.S.K. Annamalai, A. Cangelosi

Research output: Contribution to journalArticlepeer-review


Kalman filter has been successfully applied to fuse the motion capture data collected from Kinect sensor and a pair of MYO armbands to teleoperate a robot. A new strategy utilizing the vector approach has been developed to accomplish a specific motion capture task. The arm motion of the operator is captured by a Kinect sensor and programmed with Processing software. Two MYO armbands with the inertial measurement unit embedded are worn on the operator's arm, which is used to detect the upper arm motion of the human operator. This is utilized to recognize and to calculate the precise speed of the physical motion of the operator's arm. User Datagram Protocol is employed to send the human movement to a simulated Baxter robot arm for teleoperation. In order to obtain joint angles for human limb utilizing vector approach, RosPy and Python script programming has been utilized. A series of experiments have been conducted to test the performance of the proposed technique, which provides the basis for the teleoperation of simulated Baxter robot.
Original languageEnglish
Pages (from-to)156-167
JournalSystems Science and Control Engineering
Issue number1
Publication statusPublished - 17 Mar 2017


  • Kalman filter-based sensor fusion
  • vector approach
  • Baxter
  • MYO armband
  • Kinect


Dive into the research topics of 'Teleoperation control of Baxter robot using Kalman filter-based sensor fusion'. Together they form a unique fingerprint.

Cite this