MASTERLY_niMble Artificial intelligence driven robotic SoluTions for Efficient and self-deteRmined handLing and assemblY operations (G.A.101091800)
Published September 24, 2024 | Version v1
Book chapter Open

Rapid and Precise Method for Object Detection and Localization Based on Primitive Geometrical Forms

Authors/Creators

  • 1. ROR icon RWTH Aachen University
  • 2. Institute of Mechanism Theory, Machine Dynamics and Robotics
  • 3. Chair of Individualized Production in Architecture

Description

In an era where industries are becoming increasingly automated, the role of robotic manipulators is vital, particularly in tasks such as various pick-and-place operations. To enhance the capabilities of robotic manipulators, this study develops a rapid detection and localization method for simple objects using a depth camera. The method is implemented on an Kawasaki Astorino 3D-printed low-cost robot connected to a low-performance personal computer running ROS2. The method-ology involves the development of a detection method that uses a combination of a Haar cascade classifier and a Support Vector Machine (SVM) with Histogram of Oriented Gradient (HOG). Utilizing the information from the depth camera, the system determines the pose of detected objects in the robot’s base frame. The models are subsequently trained, tested, and evaluated with two types of objects: boxes and cylinders. The results are promising. Both models achieve a precision score of 1 and a recall score of approximately 0.85. Remarkably, both operate at a speed exceeding 60 FPS. Furthermore, grasp tests yield a success rate of at least 80%. The finding of the research was conducted as part of the MASTERLY project and not only solves the need for fast and efficient object detection and localization in the context of friendly-budget systems but also paves the way for more advanced applications not only in manufacturing but also in other industries and adaptive pick-and-place tasks.

Files

ISRM_Ergebnis_Chapter27.pdf

Files (4.6 MB)

Name Size Download all
md5:4c339ff5c14e8f0c1e9db86e1e10281c
4.6 MB Preview Download

Additional details

Identifiers

ISBN
978-3-031-59888-3

Funding

European Commission
MASTERLY - Nimble Artificial Intelligence driven robotic solutions for efficient and self-determined handling and assembly operations 101091800

References

  • Nof, S.Y.: Handbook of industrial robotics. John Wiley & Sons Inc., Hoboken, NJ, USA (1999)
  • Barbosa, W.S. et al.: Industry 4.0: examples of the use of the robotic arm for digital manufac-turing processes. Int. J. Interact. Des. Manuf. (IJIDeM) 4(14), 1569–1575 (2020). https://doi. org/10.1007/s12008-020-00714-4
  • O'Shea, K., Nash, R.: An introduction to convolutional neural networks, (2015). https://doi. org/10.48550/arXiv.1511.08458
  • Redmon, J., et al.: You only look once: unified. R.-Time Object Detect. (2015). https://doi.org/ 10.48550/arXiv.1506.02640
  • Girshick, R. et al.: Rich feature hierarchies for accurate object detection and semantic segmentation, (2013). https://doi.org/10.48550/arXiv.1311.2524
  • Redmon, J., Farhadi, A.: YOLO9000: Better, faster. Stronger (2016). https://doi.org/10.48550/ arXiv.1612.08242
  • Soltan, S. et al.: Deep learning-based object classification and position estimation pipeline for potential use in robotized pick-and-place operations. Robotics 3(9), 63 (2020). https://doi.org/ 10.3390/robotics9030063
  • Yu, Y., et al.: Fruit detection for strawberry harvesting robot in non-structural environment based on Mask-RCNN. Comput. Electron. Agric. 163, 104846 (2019). https://doi.org/10.1016/ j.compag.2019.06.001
  • Chen, Z. et al.: An object detection and localization method based on improved YOLOv5 for the teleoperated robot. Appl. Sci. 22(12), 11441 (2022). https://doi.org/10.3390/app122211441
  • Kawasaki Robotics.: Astorino|Industrial robots by Kawasaki Robotics. https://kawasakirobo tics.com/eu-africa/astorino/. Accessed 9 Aug 2023
  • Macenski, S. et al.: Robot operating system 2: Design, architecture, and uses in the wild. Sci. Robot. 66/7, eabm6074 (2022). https://doi.org/10.1126/scirobotics.abm6074
  • MASTERLY Projekt; supported by the European Union under the Horizon Europe programme under grant agreement N. 101091800 https://www.masterly-project.eu/partners/
  • Chandran, S., Desai, U.B.: Sixth international conference on computer vision. Sponsored by the IEE Computer Society, January 4–7, 1998, Bombay, India/[editorial production: Sharat Chandran and Uday Desai]. Narosa, New Dehli, London (1998)
  • Viola, P., Jones, M.: Rapid object detection using a boosted cascade of simple features. In: Proceedings of the 2001 IEEE computer society conference on computer vision and pattern recognition, pp. I–I. CVPR 2001, Kauai, HI, USA (2001). https://doi.org/10.1109/CVPR.2001. 990517
  • Freund, Y., Schapire, R.E.: A decision-theoretic generalization of on-line learning and an application to boosting. J. Comput. Syst. Sci. 1(55), 119–139 (1997). https://doi.org/10.1006/ jcss.1997.1504
  • Freeman, W., Roth, M.: Orientation histograms for hand gesture recognition, (1995)
  • Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Schmid, C., Soatto, S., Tomasi, C. (eds.) CVPR 2005. Proceedings 2005 IEEE computer society confer-ence on computer vision and pattern recognition, pp. 886–893. IEEE Computer Society, Los Alamitos Calif., (2005). https://doi.org/10.1109/CVPR.2005.177
  • Cortes, C., Vapnik, V.: Support-vector networks. Mach. Learn. 3(20), 273–297 (1995)
  • Sturm, P.: Pinhole camera model: computer vision, pp. 610–613. Springer, Boston, MA (2014)
  • Daniilidis, K.: Hand-eye calibration using dual quaternions. Int. J. Robot. Res. 3(18), 286–298 (1999). https://doi.org/10.1177/027836499220662
  • Belsare, K., et al.: Micro-ROS: Robot Operating System (ROS), pp. 3–55. Springer, Cham (2023)
  • Samer Khshiboun.: Realsense-ros: Intel(R) RealSense(TM) ROS wrapper for depth camera. https://github.com/IntelRealSense/realsense-ros. Accessed 26 Oct 2023
  • Coleman, D.T., et al.: Reducing the barrier to entry of complex robotic software: a MoveIt! case study. Universit degli studi di Bergamo (2014). https://doi.org/10.48550/arXiv.1404.3785
  • Ullah, M.B., Ullah, M.B.: CPU based YOLO: A real time object detection algorithm. In: 2020 IEEE region 10 Symposium (TENSYMP), [S.l.], pp. 552–555. IEEE, (2020). https://doi.org/ 10.1109/TENSYMP50017.2020.9230778