Loop Closure Detection in a Robotic Arm Using a Forward Dynamics Dataset
Authors
John Li, Nikhil Yadav
Share
Annotation
Loop closure detection is significant within the field of robotics due to its role in enhancing accuracy and system efficiency. This study focuses on differentiating between closed-loop and open-loop behaviors in robotic arm motion using a forward dynamics dataset. Closed-loop systems offer heightened accuracy and reliability, finding widespread utility in automotive manufacturing, while open-loop systems, characterized by distinct traits, are extensively employed in entertainment industries. Leveraging a vast dataset encompassing millions of data points covering both closed and open loop movements, this paper employs classical machine and deep learning methodologies to classify such behaviors. Using conventional machine learning models, the discriminatory power is observed to be impressive, with decision trees yielding classification accuracies and F1-scores of up to 90%. Complementing these efforts, a neural network model is employed, achieving a similar accuracy of 91%. This research not only builds upon existing work but also introduces a novel comparative framework that to the best of our knowledge has been unexplored for such a large dataset. By harnessing data generated from a 3-degree-of-freedom robotic arm, the study shows success in discerning the fundamental nature of open-loop or closed-loop configurations. This paper contributes to advancing the understanding of loop closure detection, holding implications for enhancing robotic control and performance across diverse applications.
Keywords
Authors
John Li, Nikhil Yadav
Share
References:
X. Bombois, B. Anderson, and G. Scorletti, "Open-loop vs. closed-loop identification of box-jenkins systems in a ..." ANU Research Publications. March 2015. Available at: [https://openresearch-repository.anu.edu.au/handle/1885/39591](https://openresearch-repository.anu.edu.au/handle/1885/39591)
B. Borovic, A. Liu, D. Popa, H. Cai, and F. L. Lewis, "Open-loop versus closed-loop control of MEMS devices: Choices And Issues," Journal of Micromechanics and Microengineering. October 2005. Available at: [https://www.researchgate.net/publication/228388708_Open-loop_versus_closed-loop_control_of_MEMS_devices_Choices_and_issues](https://www.researchgate.net/publication/228388708_Open-loop_versus_closed-loop_control_of_MEMS_devices_Choices_and_issues)
D. Agudelo-Espana, A. Zadaianchuk, P. Wenk, A. Garg, J. Akpo, J. Felix Grimminger, M. Naveau, L. Righetti, G. Martius, A. Krause, B. Scholkopf, S. Bauer, and M. Wuthrich, "A real-robot dataset for assessing transferability of learned dynamics," 2020 IEEE International Conference on Robotics and Automation (ICRA). 2020. Available at: [https://is.mpg.de/uploads_file/attachment/attachment/589/ICRA20_1157_FI.pdf](https://is.mpg.de/uploads_file/attachment/attachment/589/ICRA20_1157_FI.pdf)
S. Surati, S. Hedaoo, T. Rotti, and V. Ahuja, "Pick and place robotic arm: A Review Paper," IRJET. February 2021. Available at: [https://www.irjet.net/archives/V8/i2/IRJET-V8I2311.pdf](https://www.irjet.net/archives/V8/i2/IRJET-V8I2311.pdf)
M. Soori, B. Arezoo, and R. Dastres, "Artificial Intelligence, Machine Learning and deep learning in advanced robotics, a review," Cognitive Robotics. April 6, 2023. Available at: [https://www.sciencedirect.com/science/article/pii/S2667241323000113](https://www.sciencedirect.com/science/article/pii/S2667241323000113)
B. Vaisi, "A review of optimization models and applications in robotic manufacturing systems: Industry 4.0 and Beyond," Decision Analytics Journal. February 17, 2022. Available at: [https://www.sciencedirect.com/science/article/pii/S2772662222000054](https://www.sciencedirect.com/science/article/pii/S2772662222000054)
V. Trianni and M. López-Ibáñez, "Advantages of task-specific multi-objective optimisation in Evolutionary Robotics," PloS One. August 21, 2015. Available at: [https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4546428/](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4546428/)
T. Gold, A. Volz, and K. raichen, "Model predictive interaction control for industrial robots," IFAC-PapersOnLine. April 14, 2021. Available at: [https://www.sciencedirect.com/science/article/pii/S2405896320334583](https://www.sciencedirect.com/science/article/pii/S2405896320334583)
S. N. Mellatshahi, "Learning Control of Robotic Arm Using Deep Q-Neural Network," Scholarship at UWindsor. March 10, 2021.Available at: [https://scholar.uwindsor.ca/cgi/viewcontent.cgi?article=9573&context=etd](https://scholar.uwindsor.ca/cgi/viewcontent.cgi?article=9573&context=etd)
R. Liu, F. Nageotte, P. Zanne, M. D. Mathelin, and B. Dresp-Langley, "Deep reinforcement learning for the control of robotic manipulation: A ..." Arxiv. 2021. Available at: [https://arxiv.org/pdf/2102.04148](https://arxiv.org/pdf/2102.04148)
Diego. "RR-learning/transferable_dynamics_dataset: Datasets and code to evaluate transferability of dynamics learning methods," rr-learning. 2020. Available at: [https://github.com/rr-learning/transferable_dynamics_dataset](https://github.com/rr-learning/transferable_dynamics_dataset)
R. Kundu, "F1 score in Machine Learning: Intro & Calculation," V7. December 16, 2022. Available at: [https://www.v7labs.com/blog/f1-score-guide](https://www.v7labs.com/blog/f1-score-guide)
Other articles of the issue
