TY - JOUR
T1 - Neural Network Augmented Sensor Fusion for Pose Estimation of Tensegrity Manipulators
AU - Kuzdeuov, Askat
AU - Rubagotti, Matteo
AU - Varol, Huseyin Atakan
N1 - Funding Information:
Manuscript received November 14, 2019; accepted December 10, 2019. Date of publication December 13, 2019; date of current version March 5, 2020. This work was supported in part by the Ministry of Education and Science of the Republic of Kazakhstan Grant “Methods for Safe Human Robot Interaction with Variable Impedance Actuated Robots.” The associate editor coordinating the review of this article and approving it for publication was Dr. Amitava Chatterjee. (Corresponding author: Matteo Rubagotti.) A. Kuzdeuov is with the Institute of Smart Systems and Artificial Intelligence, Nazarbayev University, 010000 Nur-Sultan, Kazakhstan (e-mail: [email protected]).
Publisher Copyright:
© 2001-2012 IEEE.
PY - 2020/4/1
Y1 - 2020/4/1
N2 - In this paper, we present a pose estimation strategy for the end effector of a tensegrity manipulator, based on the use of an extended Kalman filter and a deep feedforward neural network with three hidden layers. Our scheme is based on the fusion of sensor data obtained from an inertial measurement unit and ArUco fiducial markers. The method was implemented on a six bar tensegrity prism manipulator, tested using ground truth acquired from an external vision-based motion capture system, and compared with other estimation methods. The experimental results show the ability of our method to provide reliable pose estimates, also dealing with the problems caused by the tensegrity structure, including marker occlusions due to the presence of bars and strings.
AB - In this paper, we present a pose estimation strategy for the end effector of a tensegrity manipulator, based on the use of an extended Kalman filter and a deep feedforward neural network with three hidden layers. Our scheme is based on the fusion of sensor data obtained from an inertial measurement unit and ArUco fiducial markers. The method was implemented on a six bar tensegrity prism manipulator, tested using ground truth acquired from an external vision-based motion capture system, and compared with other estimation methods. The experimental results show the ability of our method to provide reliable pose estimates, also dealing with the problems caused by the tensegrity structure, including marker occlusions due to the presence of bars and strings.
KW - extended Kalman filters
KW - fiducial markers
KW - inertial measurement sensors
KW - neural networks
KW - Pose estimation
KW - sensor fusion
KW - tensegrity robots
UR - http://www.scopus.com/inward/record.url?scp=85081718854&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85081718854&partnerID=8YFLogxK
U2 - 10.1109/JSEN.2019.2959574
DO - 10.1109/JSEN.2019.2959574
M3 - Article
AN - SCOPUS:85081718854
SN - 1530-437X
VL - 20
SP - 3655
EP - 3666
JO - IEEE Sensors Journal
JF - IEEE Sensors Journal
IS - 7
M1 - 8932581
ER -