
Publications
2020 |
|
![]() | Donat, Heiko; Lilge, Sven; Burgner-Kahrs, Jessica; Steil, Jochen J Estimating Tip Contact Forces for Concentric Tube Continuum Robots based on Backbone Deflection Journal Article IEEE Transactions on Medical Robotics and Bionics , 2 (4), pp. 619-630, 2020. Abstract | Links | BibTeX | Tags: concentric tube continuum robot, machine learning, sensing @article{Donat2020, title = {Estimating Tip Contact Forces for Concentric Tube Continuum Robots based on Backbone Deflection}, author = {Heiko Donat and Sven Lilge and Jessica Burgner-Kahrs and Jochen J. Steil}, doi = {10.1109/TMRB.2020.3034258}, year = {2020}, date = {2020-10-29}, journal = {IEEE Transactions on Medical Robotics and Bionics }, volume = {2}, number = {4}, pages = {619-630}, abstract = {Concentric Tube Continuum Robots are among the smallest and most flexible instruments in development for minimally invasive surgery, thereby enabling operations in areas within the human body that are difficult to reach. Unfortunately, integrating state-of-the-art force sensors is challenging for these robots due to their small form factor, although contact forces are essential information in surgical procedures. In this work, we propose a novel data-driven approach based on Deep Direct Cascade Learning (DDCL) to create a virtual sensor for computing the tip contact force of Concentric Tube Continuum Robots. By exploiting the robot’s backbone’s inherent elasticity, deflection is used to estimate the respective external tip contact force. We evaluate our approach on different data representations for a single tube and apply it subsequently on a three-segment Concentric Tube Continuum Robot. Furthermore, we devise a novel transfer learning approach through DDCL to improve the estimation accuracy by pre-training a cascaded network with simulated data. Subsequently, we fine-tune the network based on a small real-world data set recorded from the physical robot.}, keywords = {concentric tube continuum robot, machine learning, sensing}, pubstate = {published}, tppubtype = {article} } Concentric Tube Continuum Robots are among the smallest and most flexible instruments in development for minimally invasive surgery, thereby enabling operations in areas within the human body that are difficult to reach. Unfortunately, integrating state-of-the-art force sensors is challenging for these robots due to their small form factor, although contact forces are essential information in surgical procedures. In this work, we propose a novel data-driven approach based on Deep Direct Cascade Learning (DDCL) to create a virtual sensor for computing the tip contact force of Concentric Tube Continuum Robots. By exploiting the robot’s backbone’s inherent elasticity, deflection is used to estimate the respective external tip contact force. We evaluate our approach on different data representations for a single tube and apply it subsequently on a three-segment Concentric Tube Continuum Robot. Furthermore, we devise a novel transfer learning approach through DDCL to improve the estimation accuracy by pre-training a cascaded network with simulated data. Subsequently, we fine-tune the network based on a small real-world data set recorded from the physical robot. |
2018 |
|
![]() | van Roosbroeck, J; Modes, V; Lindner, E; van Hoe, B; Voigtlander, C; Vlekken, J; Burgner-Kahrs, J Curvature and Shape Sensing for Continuum Robotics using Draw Tower Gratings in Multi Core Fiber Inproceedings 26th International Conference on Optical Fiber Sensors (OFS), 2018. Abstract | Links | BibTeX | Tags: concentric tube continuum robot, fbg, sensing @inproceedings{VanRoosbroeck2018, title = {Curvature and Shape Sensing for Continuum Robotics using Draw Tower Gratings in Multi Core Fiber}, author = {J van Roosbroeck and V Modes and E Lindner and B van Hoe and C Voigtlander and J Vlekken and J Burgner-Kahrs}, doi = {10.1364/OFS.2018.ThE70}, year = {2018}, date = {2018-09-28}, booktitle = {26th International Conference on Optical Fiber Sensors (OFS)}, abstract = {Dense arrays of Draw Tower Gratings (DTG®s) have been produced in 7 core multicore fiber. They are measured with a standard spectrometer based readout system using Wavelength Division Multiplexing. We demonstrate that these sensor arrays can be used for curvature and shape sensing in continuum robotics and the accuracy for the measurement of both parameters will be presented.}, keywords = {concentric tube continuum robot, fbg, sensing}, pubstate = {published}, tppubtype = {inproceedings} } Dense arrays of Draw Tower Gratings (DTG®s) have been produced in 7 core multicore fiber. They are measured with a standard spectrometer based readout system using Wavelength Division Multiplexing. We demonstrate that these sensor arrays can be used for curvature and shape sensing in continuum robotics and the accuracy for the measurement of both parameters will be presented. |