22.09.2020
OptoFence II

Projektziele

• Entwicklung eines schnellen Teleskopsystems für die Erkennung und Verfolgung von Drohnen
• Deep Learning-Objekterkennung und -Verfolgung in Echtzeit
• Kamerabasierte Regelung und Nachführung

Beschreibung

Drohnen und andere Arten von unbemannten Flugobjekten (UAVs) gewannen in den letzten Jahren nicht nur im professionellen, sondern auch im privaten Sektor massiv an Popularität. Vorfälle, wie die Schließung des Londoner Flughafens Gatwick aufgrund der Sichtung einer Drohne, zeigen, dass die Fortschritte in der UAV-Technologie eine Bedrohung der öffentlichen Sicherheit darstellen. Die frühzeitige Identifizierung ankommender UAVs ist für die Lagebeurteilung von höchster Priorität.
Kommerzielle Drohnenerkennungssysteme nutzen multispektrale Erkennungssysteme zur Objekterkennung und -identifizierung. Hierfür wird das auf das Zusammenspiel verschiedener Sensoren gesetzt, um Objekte erkennen und identifizieren zu können. Die folgende Abbildung zeigt ein Beispiel, bei dem ein Objekt in einer Entfernung von 5 bis 10 km mit Hilfe von Radar erkannt wird. Problematisch ist jedoch, dass nur schwer eine Unterscheidung zwischen UAV und beispielsweise einem Vogel getroffen werden kann. Hierfür werden optische Sensoren verwendet, welche basierend auf einem aufgenommenen Kamerabild das Objekt eindeutig klassifizieren können. Die Einsatzdistanz dieser optischen Komponente ist zur Zeit auf ein bis zwei Kilometer limitiert, was im Falle einer Bedrohung nur kurze Reaktionszeiten zulässt.

OptoFence II zielt darauf ab, eine teleskopbasierte optische Pattform zu entwickeln, um ein größeres Identifikationsgebiet zu ermöglichen, was die Zeitspanne für die Lagebeurteilung erheblich verlängert. Durch die Kombination einer präzisen und schnellen Montierung, eines hochwertigen Teleskops, eines Kamerasystems und fortschrittlicher Methoden von Regelungssystemen und Computer Vision entsteht eine vielseitige optische Plattform für die optische Erkennung, Verfolgung und Identifizierung von UAVs.

Das Grundkonzept ist dargestellt in Abbildung 2. Ein geeignetes paar aus Teleskopen und Kameras liefern hochauflösende Bilder. Diese werden im nächsten Schritt mit Hilfe von modernen Deep Learning Algorithmen analysiert, um die Position des UAVs in den einzelnen Bildern zu extrahieren. Hierfür wurde im Projekt eine effiziente Software Architektur implementiert, welche das Detektieren und Verfolgen von Drohnen mit bis zu 100 Bildern pro Sekunde ermöglicht. Die gewonnenen Daten dienen fortschrittlichen regelungstechnischen Kontrollern als Eingabe, um das Teleskop präzise der Flugbahn des Objektes folgen zu lassen. Zusätzlich ermöglicht eine eigens implementierte automatische Fokusnachführung die Drohne scharf im Fokus zu halten. Das implementierte System wurde während Feldtests unter diversen Szenarien intensiv getestet. Es wurde demonstriert, dass kleine Drohnen, wie beispielsweise die DJI Mavic 3, bis hin zu einer Distanz von 5 km detektiert und verfolgt werden können.

 

Anwendungen

• Drohnenaufklärung

Relevante Publikationen

• D. Ojdanić, A. Sinn, C. Naverschnigg, and G. Schitter, Feasibility Analysis of Optical UAV Detection Over Long Distances Using Robotic Telescopes, IEEE Transactions on Aerospace and Electronic Systems, vol. 59, iss. 5, pp. 5148-5157, 2023.
  [BibTex] [Download]

  @ARTICLE{2023_ojdanic_taes,
  author={Ojdani{\'c}, Denis and Sinn, Andreas and Naverschnigg, Christopher and Schitter, Georg},
  journal={IEEE Transactions on Aerospace and Electronic Systems},
  title={Feasibility Analysis of Optical UAV Detection Over Long Distances Using Robotic Telescopes},
  year={2023},
  volume={59},
  number={5},
  pages={5148-5157},
  doi={10.1109/TAES.2023.3248560},
  }

• D. Ojdanić, C. Naverschnigg, A. Sinn, D. Zelinskyi, and G. Schitter, Parallel Architecture for Low Latency UAV Detection and Tracking Using Robotic Telescopes, IEEE Transactions on Aerospace and Electronic Systems, vol. 60, iss. 4, pp. 5515-5524, 2024.
  [BibTex] [Download]

  @Article{2024_ojdanic_taes,
  author={Ojdanić, Denis and Naverschnigg, Christopher and Sinn, Andreas and Zelinskyi, Daniil and Schitter, Georg},
  journal={IEEE Transactions on Aerospace and Electronic Systems},
  title={Parallel Architecture for Low Latency UAV Detection and Tracking Using Robotic Telescopes},
  year={2024},
  volume={60},
  number={4},
  pages={5515-5524},
  doi={10.1109/TAES.2024.3396418},
  }

• D. Ojdanić, D. Zelinskyi, C. Naverschnigg, A. Sinn, and G. Schitter, High-speed telescope autofocus for UAV detection and tracking, Optics Express, vol. 32, iss. 5, p. 7147–7157, 2024.
  [BibTex] [Download]

  @Article{2024_ojdanic_oe,
  author = {Ojdani{\'c}, Denis and Zelinskyi, Daniil and Naverschnigg, Christopher and Sinn, Andreas and Schitter, Georg},
  journal = {Optics Express},
  title = {High-speed telescope autofocus for UAV detection and tracking},
  year = {2024},
  issn = {1094-4087},
  month = feb,
  number = {5},
  pages = {7147--7157},
  volume = {32},
  doi = {10.1364/oe.514859},
  publisher = {Optica Publishing Group},
  }

• D. Ojdanić, N. Paternoster, C. Naverschnigg, A. Sinn, and G. Schitter, Evaluation of the required optical resolution for deep learning-based long-range UAV detection, in Pattern Recognition and Tracking XXXV, 2024.
  [BibTex] [Download]

  @InProceedings{2024_ojdanic_spie,
  author = {Ojdani{\'c}, Denis and Paternoster, Niklas and Naverschnigg, Christopher and Sinn, Andreas and Schitter, Georg},
  booktitle = {Pattern Recognition and Tracking XXXV},
  title = {Evaluation of the required optical resolution for deep learning-based long-range UAV detection},
  year = {2024},
  month = jun,
  publisher = {SPIE},
  doi = {10.1117/12.3013251},
  }

• D. Ojdanić, C. Naverschnigg, A. Sinn, and G. Schitter, Algorithm evaluation for parallel detection and tracking of UAVs, in Optics, Photonics, and Digital Technologies for Imaging Applications VIII, 2024.
  [BibTex] [Download]

  @InProceedings{2024_ojdanic_spie1,
  author = {Ojdani{\'c}, Denis and Naverschnigg, Christopher and Sinn, Andreas and Schitter, Georg},
  booktitle = {Optics, Photonics, and Digital Technologies for Imaging Applications VIII},
  title = {Algorithm evaluation for parallel detection and tracking of UAVs},
  year = {2024},
  month = jun,
  publisher = {SPIE},
  doi = {10.1117/12.3017037},
  }

• D. Ojdanić, C. Naverschnigg, A. Sinn, and G. Schitter, Deep learning-based long-distance optical UAV detection: color versus grayscale, in Pattern Recognition and Tracking XXXIV, 2023.
  [BibTex] [Download]

  @InProceedings{2023_ojdanic_spie,
  author = {Ojdani{\'c}, Denis and Naverschnigg, Christopher and Sinn, Andreas and Schitter, Georg},
  booktitle = {Pattern Recognition and Tracking XXXIV},
  title = {Deep learning-based long-distance optical UAV detection: color versus grayscale},
  year = {2023},
  month = jun,
  publisher = {SPIE},
  doi = {10.1117/12.2663318},
  }

• D. Ojdanić, B. Gräf, A. Sinn, H. W. Yoo, and G. Schitter, Camera-guided real-time laser ranging for multi-UAV distance measurement, Appl. Opt., vol. 61, iss. 31, p. 9233–9240, 2022.
  [BibTex] [Download]

  @article{ojdanic2022AppliedOptics,
  author = {Denis Ojdani\'{c} and Benjamin Gr\"{a}f and Andreas Sinn and Han Woong Yoo and Georg Schitter},
  journal = {Appl. Opt.},
  keywords = {Laser beams; Laser light; Laser ranging; Optical components; Optical testing; Remote sensing},
  number = {31},
  pages = {9233--9240},
  publisher = {Optica Publishing Group},
  title = {Camera-guided real-time laser ranging for multi-UAV distance measurement},
  volume = {61},
  month = {Nov},
  year = {2022},
  doi = {10.1364/AO.470361},
  }

• D. Ojdanic, A. Sinn, C. Schwaer, and G. Schitter, UAV Detection and Tracking with a Robotic Telescope System, in Proceedings of the Advanced Intelligent Mechatronics Conference 2021, 2021.
  [BibTex]

  @inproceedings{TUW-299533,
  author = {Ojdanic, Denis and Sinn, Andreas and Schwaer, Christian and Schitter, Georg},
  title = {UAV Detection and Tracking with a Robotic Telescope System},
  booktitle = {Proceedings of the Advanced Intelligent Mechatronics Conference 2021},
  year = {2021},
  note = {Posterpr{\"a}sentation: 2021 IEEE International Conference on Advanced Intelligent Mechatronics (AIM), Delft, Netherlands; 2021-07-12 -- 2021-07-16}
  }

Projektpartner

• ASA Astrosysteme GmbH
• Bundesministerium für Landesverteidigung

Funding

Das Projekt wird innerhalb des Verteidigungsforschungs–Förderprogramm FORTE durch das Bundesministerium für Landwirtschaft, Regionen und Tourismus (BMLRT) gefördert.

 

 

 

 

  27.03.2020
Precision robotic inline metrology for freeform surfaces

Project focus

• 6DoF measurement platform with arbitrary operation orientation
• Integrated control strategies for dual sage-actuated robotic positioning and automated measurement procedures
• Conduct high precision 3D measurements on a moving object directly in a production line

Description

High precision in-line measurements are considered as a key factor for future industrial production.  Particularly for surface measurements of free forms directly in a production line, flexible and versatile measurement systems are required. Robot-based measurement systems enable the required flexibility and are thus considered as a key factor. High-precision measurement systems, which operate at the micrometer level, are typically highly sensitive to vibrations and, therefore, usually applied in vibration-free laboratory environments. This is, however in complete contrary to the vibration-prone environment of an industrial manufacturing plant. Relative motion between the measurement tool and the sample in 6 DOFs is caused by environmental floor vibrations, disturbing measurements at the micrometer level.

Within the scope of this project, a vibration compensating measurement module is developed. This module is used as an end effector and coarsely aligned by the robot arm. The actuated platform, which carries the high precision measurement tool, compensates for disturbing vibrations in 6 DoFs by means of feedback control. A stiff link between the 3D measurement tool and the sample is established, maintaining a constant alignment of the tool with respect to the sample surface. Using dual stage control strategies, the robot arm is repositioned, such that the measurement platform is maintained within its limited actuation range. In this way, the long-range tracking of moving samples is enabled.

The feasibility of the system concept has been shown for horizontal operation orientation in the project aim4np. The goal of this project is to enable contact-less high precision robot-based measurements of free form surfaces by a measurement module with orientation-independent system performance.

 

3D Measurement Module Design and Integration

To mechanically decouple the measurement platform from the robot arm, (quasi-)zero stiffness actuation is desired. Therefore, eight voice coil actuators are utilized to levitate and actuated the platform in 6 DoFs within the airgap between coils and magnet housings. By use of FEM analysis tools, the structural stiffness of the platform is analyzed, aiming for a high position control bandwidth and vibration compensation capability. Good decoupling of the DoFs is ensured by a balanced design, avoiding unintended torsional motion through actuation and keeping the subsequent control effort at a low level. The system is deliberately over-actuated, providing high peak forces, and the arbitrary operation orientation of the entire system. An integrated sensor system determines the actual position of the moving platform with respect to the supporting frame, while tracking sensors are used to measure the distance and orientation between platform and sample.

By manipulating the optical path of a 1-D confocal chromatic sensor with a high performance 2-D fast steering mirror, 3-D imaging is enabled. Using a CMOS camera, calibration data is acquired for a precise image reconstruction. The developed 3-D tool is capable of performing fast overview images and high resolution scans of arbitrary region of interests. The lateral scan range is about 350µm x 250µm, with scan speeds of up to 1 frame per second. The achieved lateral and axial resolution is 2.5µm and 7nm, respectively.

 

 

Precision Motion Control

Model-based control techniques are applied to ensure a robust and high-performance positioning of the electro-magnetically actuated platform. A system model is built to decouple the MIMO plant, resulting in the implementation of a SISO controller for each DoF. A transition scheme is researched to operate the system in two different control modes, the stabilization of the platform with respect to the supporting frame and the tracking control of a sample surface.

To enable the long-range tracking of moving samples, dual stage control strategies are applied. The robot arm is repositioned, such that the measurement platform is maintained within its limited actuation range.

 

High-precision 3D measurements on moving samples

In case of disabled sample-tracking, relative motion corrupts the targeted 3D measurement. By enabling the sample-tracking controller, the contactless stiff link between the scanning confocal chromatic sensor and the sample, in this case a 20µm pitch grating, compensates the disturbing relative motion. In this way, 3D images of the moving sample are achieved with sub-µm precision.

 

 

Applications

• Inline measurement systems
• Robot-based precision measurement and manufacturing systems

Project Partners

• Micro-Epsilon Messtechnik GmbH & Co. KG
• ATENSOR Engineering and Technology Systems GmbH

Funding

• Christian Doppler Research Association
• The Austrian Federal Ministry for Digital and Economic Affairs
• The National Foundation for Research, Technology and Development

  26.03.2020
Hybrid reluctance actuators for high precision motion

Project focus

• Hybrid reluctance actuators (HRAs) realize a motor constant higher than comparable voice coil actuators for motion control with nanometer resolution.
• HRAs’ magnetic force cancels the stiffness of guiding flexures for large motion and energy efficiency.
• Integrated mechatronic system design enables a robust PID controller to realize a high bandwidth.

Description

Next-generation high-quality motion systems require high-precision actuators with higher energy efficiency and larger force to improve the system throughput. Particularly, actuators with a motor constant higher than comparable voice coil actuators are highly desired. This project investigates hybrid reluctance actuators (HRAs) with guiding flexures as a promising candidate of next-generation systems. The following table and video show the currently ahcieved performance:

Table 1: Evaluated performance of a HRA

Actuation principle

A hybrid reluctance actuator (HRA) uses a permanent magnet and coils to generate magnetic fluxes through the ferromagnetic stator and mover. As Fig. 1(a) illustrates, the direction of the coil’s flux is the opposite of the magnet’s flux in the variable right gap. These fluxes cancel each other. Consequently, the total flux in the variable right gap is weaker than that in the variable left gap. The unbalanced flux results in the lateral actuation force.

HRAs generate a bidirectional force. It is relatively proportional to the coil current. The motor constant of HRAs can be a few times larger than that of a comparable voice coil actuators. However, unlike voice coil actuators, HRAs exhibit position-dependent magnetic force, which is modeled by a nonlinear negative stiffness.

Stiffness cancellation for long-range motion and high energy efficiency

A challenge of flexure-guided systems in general is that positioning requires a constant force against the flexure stiffness, which is significant around the stroke end. For energy efficiency, a HRA is designed such that its negative stiffness cancels the flexure stiffness. The simulation in Fig. 1(b) shows that the negative stiffness decreases the force required for a quasi-static positioning from more than +/-80 N to less than +/-15 N. The experiment in Fig. 1(c) reveals that the power required for the positioning around the stroke end at 1 mm is less than that around 0.7mm, demonstrating the effectiveness of the stiffness cancellation.

For high precision motion, the HRA is designed such that a robust PID controller with a position sensor compensates for the nonlinear negative stiffness for closed-loop stability within the entire actuation rage of 2 mm. The closed-loop system achieves a positioning resolution of 2.5 nm and a bandwidth of 640 Hz. Furthermore, data-driven learning control is combined for accurate scanning motion. The experimental results in Fig. 2 show triangular scanning motions with an amplitude of 1.6 mm at 1 Hz and with an amplitude of 10 um at 100 Hz, where the tracking error is 10 nm (rms). Overall, the HRA realizes a large dynamic range with high energy efficiency.

Magnetic flux feedback control for system linearity

HRAs exhibit nonlinearities, which include not only the negative stiffness, but also the magnetic hysteresis seen in a B-H curve. They impair scanning motions and can make HRAs open-loop unstable with a commonly used current amplifiers. Such nonlinearities are compensated by replacing conventional current amplifiers by a proposed flux amplifier where feedback control regulates a magnetic flux within the frequency band of interest.

As illustrated in Fig. 3, the flux loop includes the B-H hysteresis between the coil current and the resulting flux. Furthermore, the position-dependent flux that results in the negative stiffness is regarded as a disturbance. Consequently, the flux feedback controller rejects the nonlinearities. In fact, experimental results in Fig. 4 demonstrate that the proposed flux amplifier significantly decreases the motion error of a triangular scanning in comparison with a conventional current amplifier, even without real-time motion control.

Applications

• Nanopositioner / Micropositioner
• Dual stage actuator
• Scanner
• Vibration isolation

Related publications

Journal articles

• S. Ito, F. Cigarini, and G. Schitter, Flux-controlled Hybrid Reluctance Actuator for High-precision Scanning Motion, IEEE Transactions on Industrial Electronics, vol. 67, iss. 11, pp. 9593-9600, 2020.
  [BibTex] [Download]

  @Article{TUW-282490,
  author = {Ito, Shingo and Cigarini, Francesco and Schitter, Georg},
  title = {Flux-controlled Hybrid Reluctance Actuator for High-precision Scanning Motion},
  journal = {IEEE Transactions on Industrial Electronics},
  year={2020},
  volume={67},
  number={11},
  pages={9593-9600},
  doi = {10.1109/TIE.2019.2952829},
  keywords = {Actuators, Nanopositioning, Magnetic variables control, Motion control},
  }

• S. Ito, S. Troppmair, B. Lindner, F. Cigarini, and G. Schitter, Long-range Fast Nanopositioner Using Nonlinearities of Hybrid Reluctance Actuator for Energy Efficiency, IEEE Transactions on Industrial Electronics, vol. 66, iss. 4, 2019.
  [BibTex] [Download]

  @Article{TUW-271193,
  author = {Ito, Shingo and Troppmair, Stefan and Lindner, Bernhard and Cigarini, Francesco and Schitter, Georg},
  title = {Long-range Fast Nanopositioner Using Nonlinearities of Hybrid Reluctance Actuator for Energy Efficiency},
  journal = {IEEE Transactions on Industrial Electronics},
  year = {2019},
  volume = {66},
  number = {4},
  doi = {10.1109/TIE.2018.2842735},
  keywords = {Actuators, motion control, nanoposition- ing, magnetic circuits},
  numpages = {9},
  }

• F. Cigarini, S. Ito, S. Troppmair, and G. Schitter, Comparative Finite Element Analysis of a Voice Coil Actuator and a Hybrid Reluctance Actuator, IEEJ Journal of Industry Applications, vol. 8, iss. 2, 2019.
  [BibTex]

  @Article{TUW-276121,
  author = {Cigarini, Francesco and Ito, Shingo and Troppmair, Stefan and Schitter, Georg},
  title = {Comparative Finite Element Analysis of a Voice Coil Actuator and a Hybrid Reluctance Actuator},
  journal = {IEEJ Journal of Industry Applications},
  year = {2019},
  volume = {8},
  number = {2},
  doi = {10.1541/ieejjia.8.192},
  keywords = {Electromagnetic actuators, high-precision motion, finite element analysis},
  numpages = {8}
  }

• S. Ito, S. Troppmair, F. Cigarini, and G. Schitter, High-speed Scanner with Nanometer Resolution Using a Hybrid Reluctance Force Actuator, IEEJ Journal of Industry Applications, vol. 8, iss. 2, 2019.
  [BibTex]

  @Article{TUW-272944,
  author = {Ito, Shingo and Troppmair, Stefan and Cigarini, Francesco and Schitter, Georg},
  title = {High-speed Scanner with Nanometer Resolution Using a Hybrid Reluctance Force Actuator},
  journal = {IEEJ Journal of Industry Applications},
  year = {2019},
  volume = {8},
  number = {2},
  doi = {10.1541/ieejjia.8.170},
  keywords = {High-precision actuators, hybrid reluctance actuator, nanopositioning, motion control.},
  numpages = {7}
  }

Conference papers

• F. Cigarini, S. Ito, J. Konig, A. Sinn, and G. Schitter, Compensation of Hysteresis in Hybrid Reluctance Actuator for High-Precision Motion, in Proceedings of the Joint Conference 8th IFAC Symposium on Mechatronic Systems (MECHATRONICS 2019), and 11th IFAC Symposium on Nonlinear Control Systems (NOLCOS 2019), 2019.
  [BibTex]

  @InProceedings{TUW-281659,
  author = {Cigarini, Francesco and Ito, Shingo and Konig, Julian and Sinn, Andreas and Schitter, Georg},
  title = {Compensation of Hysteresis in Hybrid Reluctance Actuator for High-Precision Motion},
  booktitle = {Proceedings of the Joint Conference 8th IFAC Symposium on Mechatronic Systems (MECHATRONICS 2019), and 11th IFAC Symposium on Nonlinear Control Systems (NOLCOS 2019)},
  year = {2019},
  volume = {52/15},
  note = {Vortrag: Joint Conference 8th IFAC Symposium on Mechatronic Systems (MECHATRONICS 2019), and 11th IFAC Symposium on Nonlinear Control Systems (NOLCOS 2019), Wien; 2019-09-04 -- 2019-09-06},
  doi = {10.1016/j.ifacol.2019.11.721},
  journal = {IFAC-PapersOnLine/Elsevier},
  keywords = {Hybrid reluctance actuator, magnetic hysteresis, magnetic flux control},
  numpages = {6}
  }

• S. Ito, B. Lindner, and G. Schitter, Sample-tracking Vibration Isolation with Hybrid Reluctance Actuators for Inline Metrology, in Proceedings of the Joint Conference 8th IFAC Symposium on Mechatronic Systems (MECHATRONICS 2019), and 11th IFAC Symposium on Nonlinear Control Systems (NOLCOS 2019), 2019.
  [BibTex]

  @InProceedings{TUW-281655,
  author = {Ito, Shingo and Lindner, Bernhard and Schitter, Georg},
  title = {Sample-tracking Vibration Isolation with Hybrid Reluctance Actuators for Inline Metrology},
  booktitle = {Proceedings of the Joint Conference 8th IFAC Symposium on Mechatronic Systems (MECHATRONICS 2019), and 11th IFAC Symposium on Nonlinear Control Systems (NOLCOS 2019)},
  year = {2019},
  volume = {52/15},
  note = {Vortrag: Joint Conference 8th IFAC Symposium on Mechatronic Systems (MECHATRONICS 2019), and 11th IFAC Symposium on Nonlinear Control Systems (NOLCOS 2019), Wien; 2019-09-04 -- 2019-09-06},
  doi = {10.1016/j.ifacol.2019.11.731},
  journal = {IFAC-PapersOnLine/Elsevier},
  keywords = {Vibration control, Micro and Nano Mechatronic Systems, Motion Control Systems},
  numpages = {6}
  }

• F. Cigarini, S. Ito, S. Troppmair, and G. Schitter, Comparison of a Voice Coil and a Hybrid Reluctance Actuator via FEM, in Proceedings of the IEEJ International Workshop on Sensing, Actuation, Motion Control and Optimization (SAMCON), 2018.
  [BibTex]

  @InProceedings{TUW-268854,
  author = {Cigarini, Francesco and Ito, Shingo and Troppmair, Stefan and Schitter, Georg},
  title = {Comparison of a Voice Coil and a Hybrid Reluctance Actuator via FEM},
  booktitle = {Proceedings of the IEEJ International Workshop on Sensing, Actuation, Motion Control and Optimization (SAMCON)},
  year = {2018},
  note = {Vortrag: IEEJ International Workshop on Sensing, Actuation, Motion Control and Optimization (SAMCON), Tokyo (Japan); 2018-03-06 -- 2018-03-09},
  keywords = {Electromagnetic actuators, high-precision applications, finite element method},
  numpages = {6}
  }

• S. Ito, S. Troppmair, F. Cigarini, and G. Schitter, Hybrid Reluctance Actuator for High-speed Scanning with Nanometer Resolution, in Proceedings of the IEEJ International Workshop on Sensing, Actuation, Motion Control and Optimization (SAMCON), 2018.
  [BibTex]

  @InProceedings{TUW-268844,
  author = {Ito, Shingo and Troppmair, Stefan and Cigarini, Francesco and Schitter, Georg},
  title = {Hybrid Reluctance Actuator for High-speed Scanning with Nanometer Resolution},
  booktitle = {Proceedings of the IEEJ International Workshop on Sensing, Actuation, Motion Control and Optimization (SAMCON)},
  year = {2018},
  note = {Vortrag: IEEJ International Workshop on Sensing, Actuation, Motion Control and Optimization (SAMCON), Tokyo (Japan); 2018-03-06 -- 2018-03-09},
  numpages = {6}
  }

 

Project Partners

• Micro-Epsilon Messtechnik GmbH & Co. KG
• ATENSOR Engineering and Technology Systems GmbH

Funding

• Christian Doppler Research Association
• The Austrian Federal Ministry for Digital and Economic Affairs
• The National Foundation for Research, Technology and Development

  21.02.2020
Robotisches Bekleben von 3D-Objekten

Projektfokus

• Aufkleben von zugeschnittenen Klebebändern auf Freiform-3D-Oberflächen
• Drapieren mit einer definierten Normalkraft
• Roboterwerkzeug entlang eines Oberflächenpfades auf einer Freiform-3D-Oberfläche führen

Beschreibung

Die robotische Handhabung von formlabilen Materialen, wie z.B. Textilien, Karbon- und Glasfaserstoffe, Folien und Papier, ist seit langem eine herausfordernde Aufgabe in der Industrie und das wissenschaftliche Interesse

Robotische Handhabung von formlabilen Materialien wie z. B. Textilien, Karbon- und Glasfasergewebe, Folien und Papier ist seit langem eine herausfordernde Aufgabe in der Industrie und haben in jüngster Zeit große Aufmerksamkeit in der Forschung erhalten. Eine flexible Automatisierung auf menschlichem Niveau wurde für viele Handhabungsaufgaben noch nicht erreicht, wie z.B. das Greifen, Abstapeln und Drapieren. Die Hauptherausforderungen für automatisierte Handhabung von formlabilen Materialien sind ihre Eigenschaften, insbesondere das nichtlineare Materialverhalten, die Anisotropie, die vielseitigen Außenkonturen, die hohe Dimension des Objektzustands und die Vielzahl von Materialparametern wie Porosität und Dichte. In diesem Zusammenhang wurden viele spezialisierte mechatronische Lösungen zum Greifen und Transportieren implementiert. Die Industrie benötigt jedoch für die hohe Vielfalt und geringe Stückzahl flexiblere Lösungen.

In diesem Projekt wird ein flexibler Ansatz für das impedanzgeregelte faltenfreie Aufkleben von zugeschnittenen Klebebändern auf 3D-Freiformflächen vorgestellt. Für diese Anwendung wird ein Konzept für eine oberflächenbasierte Pfadfolgeregelung entwickelt, das auf einem lokalen Koordinatensystem aufbaut, basierend auf der Oberflächennormalen und der Pfadtangente. Darüber hinaus stellen die Algorithmen und Lösungen, die aus dieser Anwendung hervorgehen, auch eine Lösung für ein allgemeineres Problem in der Robotik dar: das Durchlaufen eines Werkzeugs entlang eines gekrümmten Pfades auf einer 3D-Freiformoberfläche mit definierten kinematischen Einschränkungen.

Der Aufklebeprozess besteht aus den folgenden Schritten und ist in Abb. 1 dargestellt:

• (a) 2D-Pfad für das Klebeband: Ein planarer Pfad wird auf der Form des planaren Klebebandes erstellt.
• (b) Übertragen auf das 3D-Objekt: Die Geometrie und Krümmung des planaren Pfades wird auf das CAD-Modell des 3D-Objekts abgebildet, wodurch ein 3D-Pfad erstellt wird.
• (c) Startposition des Roboters: Der Aufklebeprozess wird simuliert, um eine geeignete Startgelenkkonfiguration für den Roboter zu bestimmen, die zu einer kollisionsfreien Ausführung führt und eine angemessene Reserve in Bezug auf die Gelenkgrenzen aufweist.
• (d) Vorbereitung und Ausführung: Das vorgeschnittene Klebeband wird in der erforderlichen Startposition dem Applikationswerkzeug platziert und der Roboter führt den impedanzgesteuerten Aufklebeprozess durch.

Das oberflächenbasierte Konzept der Pfadfolgeregelung wird an einem 3D-gedruckten Objekt demonstriert, wie in Fig. 2 gezeigt: Die Pfade der zugeschnittenen Klebebänder in (a) werden auf die Oberfläche des 3D-Objekts in (b) übertragen. Das Ergebnis des Aufklebeprozesses ist in (c) dargestellt.

Videos

Ausgewählte Publikationen

• T. Weingartshofer, M. Schwegel, C. Hartl-Nesic, T. Glück, and A. Kugi, Collaborative Synchronization of a 7-Axis Robot, in Proceedings of the 8th IFAC Symposium on Mechatronic Systems MECHATRONICS 2019, Vienna, Austria, 2019, pp. 507-512.
  [BibTex]

  @InProceedings{Weingartshofer2019,
  author = {Weingartshofer, T. and Schwegel, M. and Hartl-Nesic, C. and Gl\"uck, T. and Kugi, A.},
  title = {Collaborative Synchronization of a 7-Axis Robot},
  booktitle = {Proceedings of the 8th IFAC Symposium on Mechatronic Systems MECHATRONICS 2019},
  year = {2019},
  volume = {52},
  number = {15},
  month = {9},
  pages = {507-512},
  doi = {10.1016/j.ifacol.2019.11.726},
  address = {Vienna, Austria},
  issn = {2405-8963},
  }

• C. Hartl-Nesic, B. Bischof, T. Glück, and A. Kugi, Pfadfolgeregelung mit Konzepten für den Pfadfortschritt: Ein Assemblierungsszenario, at – Automatisierungstechnik, vol. 68, iss. 1, p. 44–57, 2020.
  [BibTex]

  @Article{Hartl-Nesic2020,
  author = {Hartl-Nesic, Christian and Bischof, Bernhard and Gl{\"u}ck, Tobias and Kugi, Andreas},
  title = {Pfadfolgeregelung mit Konzepten für den Pfadfortschritt: Ein Assemblierungsszenario},
  journal = {at -- Automatisierungstechnik},
  year = {2020},
  volume = {68},
  number = {1},
  pages = {44--57},
  issn = {2196-677X},
  doi = {10.1515/auto-2019-0114},
  }

• C. Hartl-Nesic, Surface-Based Path Following Control on Freeform 3D Objects, A. Kugi, K. Schlacher, and W. Kemmetmüller, Eds., Düren: Shaker Verlag, 2020, vol. 49.
  [BibTex]

  @Book{HartlNesic2020,
  author = {Hartl-Nesic, C.},
  title = {Surface-Based Path Following Control on Freeform 3D Objects},
  editor = {A. Kugi and K. Schlacher and W. Kemmetm\"uller},
  isbn = {978-3-8440-7637-0},
  publisher = {Shaker Verlag},
  series = {Modellierung und Regelung komplexer dynamischer Systeme},
  volume = {49},
  address = {D\"uren},
  organization = {Institute f{\"u}r Automatisierungs- und Regelungstechnik (TU Wien) und Regelungstechnik und Prozessautomatisierung (JKU Linz)},
  year = {2020},
  }

• C. Hartl-Nesic, T. Glück, and A. Kugi, Surface-Based Path Following Control: Application of Curved Tapes on 3-D Objects, IEEE Transactions on Robotics, vol. 37, iss. 2, p. 615–626, 2021.
  [BibTex] [Download]

  @Article{HartlNesic2021,
  author = {Hartl-Nesic, Ch. and Gl{\"u}ck, T. and Kugi, A.},
  title = {Surface-Based Path Following Control: Application of Curved Tapes on 3-D Objects},
  doi = {10.1109/TRO.2020.3033721},
  issn = {1552-3098},
  number = {2},
  pages = {615--626},
  volume = {37},
  journal = {IEEE Transactions on Robotics},
  year = {2021},
  }

Anwendungen

• Textil- und Bekleidungsindustrie
• Faserverstärkte Bauteile
• Lebensmittelindustrie

  22.01.2020
Modellierung, Optimierung und Qualitätsmonitoring für das Besäumen von Grobblechen

Projektschwerpunkte

• Untersuchung und Modellierung des Besäumprozesses von Grobblechen
• Optimierung des Prozesses und der Anlagenkonstruktion
• Ursachenforschung und Minimierung von Qualitätsmängeln
• Entwicklung eines kamera- und laserbasierten Systems zum Qualitätsmonitoring

Beschreibung

In der Metallindustrie bezeichnet Besäumen das möglichst rechteckige Zuschneiden von Grobblechen nach dem Walzvorgang. Rollschnittscheren (siehe Abb. 1) bewerkstelligen diese Aufgabe schrittweise und rein mechanisch mit rollenden Messern. Aufgrund der Blechdicken von bis zu 45mm entstehen dabei Prozesskräfte bis zu 5MN. Trotz der weiten Verbreitung von Besäumscheren treten häufig Qualitätsmängel an den Schnittkanten wie Ausbrüche, ungerade Kanten oder Schnittgrate auf. Zudem kann das Besäumen zu einer Abweichung von der idealen rechteckigen Walztafelkontur führen.

Im Rahmen dieses Projekts wurde die seitliche Abdrängkraft, welche auf die Besäummesser wirkt, als wesentliche Einflussgröße auf die erzielte Qualität der Schnittkanten identifiziert. Ein Schwerpunkt dieser Arbeit liegt darin, diese Abdrängkraft durch Optimierung der Anlagengeometrie und der Antriebskinematik möglichst konstant zu halten und somit die Qualitätsmängel an der Schnittkante deutlich zu reduzieren. Zudem kann damit der Messerverschleiß verringert werden.

Ein weiterer Schwerpunkt dieser Arbeit liegt auf der Entwicklung einer automatisierten Qualitätskontrolle. Die Qualität und Maßgenauigkeit der Schnittkante wird mittels Bildverarbeitung und Laser-Sensorik automatisiert ausgewertet und in einer Datenbank abgelegt. Auf Basis dieser Daten werden Modelle und Algorithmen entwickelt, um eine gleichbleibend hohe Qualität der Schnittkanten zu gewährleisten, optimale Maschineneinstellungen zu berechnen und den optimalen Messerwechselzeitpunkt vorherzusagen.

Ausgewählte Veröffentlichungen

• A. Zeiler, A. Steinboeck, A. Kugi, and M. Jochum, Lateral Forces in Rolling-Cut Shearing and Their Consequences on Common Edge Defects, Journal of Manufacturing Science and Engineering, vol. 141, iss. 4, p. 41001-1–41001-9, 2019.
  [BibTex] [Download]

  @Article{Zeiler2019,
  author = {Zeiler, A. and Steinboeck, A. and Kugi, A. and Jochum, M.},
  title = {Lateral Forces in Rolling-Cut Shearing and Their Consequences on Common Edge Defects},
  journal = {Journal of Manufacturing Science and Engineering},
  year = {2019},
  volume = {141},
  number = {4},
  pages = {41001-1--41001-9},
  doi = {10.1115/1.4042578},
  }

• A. Zeiler, A. Steinboeck, M. Vincze, M. Jochum, and A. Kugi, Vision-based inspection and segmentation of trimmed steel edges, in Proceedings of the 18th IFAC Symposium on Control, Optimization and Automation in Mining, Mineral and Metal Processing, Stellenbosch, South Africa, 2019, p. 165–170.
  [BibTex]

  @InProceedings{Zeiler2019a,
  author = {Zeiler, A. and Steinboeck, A. and Vincze, M. and Jochum, M. and Kugi, A.},
  title = {Vision-based inspection and segmentation of trimmed steel edges},
  booktitle = {Proceedings of the 18th IFAC Symposium on Control, Optimization and Automation in Mining, Mineral and Metal Processing},
  year = {2019},
  month = {8},
  pages = {165--170},
  doi = {10.1016/j.ifacol.2019.09.182},
  address = {Stellenbosch, South Africa},
  }

Anwendungsbereiche

• Scherprozesse
• Walzwerksautomatisierung
• Kontinuierliche Produktionsprozesse

  23.10.2019
Omron

The purpose of this project is to develop a software for estimating 6D pose and grasping points of non-rigid objects in bin-picking scenarios. This software would be used with industrial robots for picking individually different shaped objects like toys, vegetables, and fruits those are randomly piled in boxes or on belt conveyors. Multiple objects of one type are in a box or on a conveyor.

 

FUNDING:

Company Funding

  23.10.2019
Sasha

Sasha ist ein Assistenz-Roboter und ist von der Firma Toyota mit den Typennamen HSR „Human Support Robot“. Er kommt im privaten und industriellen Bereich zum Einsatz. Er soll Ordnung halten und muss dazu unterschiedliche Objekte und Gegenstände erkennen und zuordnen können. Dazu helfen ihm zwei Weitwinkelkameras und eine Stereokamera. Zur Orientierung besitzt er einen Laser Range Sensor. Das V4R-Team arbeitet an der Entwicklung der Bildverarbeitung, Objekterkennung und dem gezielten Greifen von Objekten. Ein Roboter mit ähnlichen Aufgaben wurde schon vom V4R Team im SQUIRREL Projekt entwickelt und heißt Kenny.

 

FUNDING:

Company Funding

  23.10.2019
Aeolus

The goal of the project with Aeolus is to advance current robotic capabilities to pick up unknown objects in conditions of clutter and fetching known objects, with emphasis on the diversity of objects addressed, and on the reliability of grasping.  During object pick-up, the robot may not need to maintain a constraint on the pose of grasped objects, while for fetching known objects includes delivering them to a human user while maintaining constraints on object pose.  Part of the project is also to acquire object models through interaction with a person, finding, remembering and updating the location of objects in a home-like environment, and physically executing the fetch and object transfer.

FUNDING:

Company Funding

  09.10.2019
InDex

Robot In-hand Dexterous manipulation by extracting data from human manipulation of objects to improve robotic autonomy and dexterity

The InDex project aims to understand how humans perform in-hand object manipulation and to replicate the observed skilled movements with dexterous artificial hands, merging the concepts of deep reinforcement and transfer learning to generalise in-hand skills for multiple objects and tasks. In addition, an abstraction and representation of previous knowledge will be fundamental for the reproducibility of learned skills to different hardware. Learning will use data across multiple modalities that will be collected, annotated, and assembled into a large dataset. The data and our methods will be shared with the wider research community to allow testing against benchmarks and reproduction of results. The core objectives are: (i) to build a multi-modal artificial perception architecture that extracts data of object manipulation by humans; (ii) the creation of a multimodal dataset of in-hand manipulation tasks such as regrasping, reorienting and finely repositioning; (iii) the development of an advanced object modelling and recognition system, including the characterisation of object affordances and grasping properties, in order to encapsulate both explicit information and possible implicit object usages; (iv) to autonomously learn and precisely imitate human strategies in handling tasks; and (v) to build a bridge between observation and execution, allowing deployment that is independent of the robot.

 

 

FUNDING:

The project is funded by FWF – Austrian Science Foundation & CHIST-ERA.

 

  09.10.2019
HEAP

Human-Guided Learning and Benchmarking of Robotic Heap Sorting

In HEAP we focus on advancing the state-of-the-art for sorting a heap of unknown, irregular objects and provide appropriate benchmarks. In our scenarios, we deal with unknown, broken or deformed object instances such as concrete, metal pipes and other plastic/metal parts of possibly complex shape. A major goal of this project is to make challenging manipulation tasks easily accessible and reproducible, and to allow for a comparative evaluation of different approaches using a standardized robotic platform and an open source simulation framework. Our intention in providing such a benchmark framework is to (i) evaluate state-of-the-art grasping and manipulation algorithms in these complex heap sorting setups and (ii) to define new challenges in terms of object recognition and manipulation that need to be solved by the community.

FUNDING:

The project is funded by FWF – Austrian Science Foundation & CHIST-ERA.