Abstract
In this paper, we propose a real-time multi-class detection system for the NAO V6 robot in the context of RoboCup SPL (Standard Platform League) using state-of-the-art structural pruning techniques on neural networks derived from YOLOv7-tiny. Our approach combines structural pruning and fine-tuning, to obtain a pruned network that maintains high accuracy while reducing the number of parameters and the computational complexity of the network. The system is capable of detecting various objects, including the ball, goalposts, and other robots, using the cameras of the robot. The goal has been to guarantee high speed and accuracy trade-offs suitable for the limited computational resources of the NAO robot. Moreover, we demonstrate that our system can run in real-time on the NAO robot with a frame rate of 32 frames per second on \(224\times 224\) input images, which is sufficient for soccer competitions. Our results show that our pruned networks achieve comparable accuracy to the original network while significantly reducing the computational complexity and memory requirements. We release our annotated dataset, which consists of over 4000 images of various objects in the RoboCup SPL soccer field.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Albani, D., Youssef, A., Suriani, V., Nardi, D., Bloisi, D.D.: A deep learning approach for object recognition with NAO soccer robots. In: Behnke, S., Sheh, R., Sarıel, S., Lee, D. (eds.) RoboCup 2016: Robot World Cup XX. LNCS, vol. 9776, pp. 392–403. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-68792-6_33
Bloisi, D., Duchetto, F.D., Manoni, T., Suriani, V.: Machine learning for realisticball detection in RoboCup SPL (2017)
Fang, G., Ma, X., Song, M., Mi, M.B., Wang, X.: Depgraph: towards any structural pruning (2023)
Leiva, F., Cruz, N., Bugueño, I., Ruiz-del Solar, J.: Playing soccer without colors in the SPL: a convolutional neural network approach. In: Holz, D., Genter, K., Saad, M., von Stryk, O. (eds.) RoboCup 2018. LNCS, pp. 122–134. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-27544-0_10
Narayanaswami, S.K., et al.: Towards a real-time, low-resource, end-to-end object detection pipeline for robot soccer. In: Eguchi, A., Lau, N., Paetzel-Prüsmann, M., Wanichanon, T. (eds.) RoboCup 2022. LNCS, vol. 13561, pp. 62–74. Springer, Cham (2023). https://doi.org/10.1007/978-3-031-28469-4_6
Szemenyei, M., Estivill-Castro, V.: Robo: robust, fully neural object detection for robot soccer. In: Chalup, S., Niemueller, T., Suthakorn, J., Williams, M.A. (eds.) RoboCup 2019. LNCS, vol. 11531, pp. 309–322. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-35699-6_24
Thielke, F., Hasselbring, A.: A JIT compiler for neural network inference. In: Chalup, S., Niemueller, T., Suthakorn, J., Williams, M.A. (eds.) RoboCup 2019. LNCS, vol. 11531, pp. 448–456. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-35699-6_36
Yao, Z., Douglas, W., O’Keeffe, S., Villing, R.: Faster yolo-lite: faster object detection on robot and edge devices. In: Alami, R., Biswas, J., Cakmak, M., Obst, O. (eds.) RoboCup 2021. LNCS, vol. 13132, pp. 226–237. Springer, Cham (2022). https://doi.org/10.1007/978-3-030-98682-7_19
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Specchi, G. et al. (2024). Structural Pruning for Real-Time Multi-object Detection on NAO Robots. In: Buche, C., Rossi, A., Simões, M., Visser, U. (eds) RoboCup 2023: Robot World Cup XXVI. RoboCup 2023. Lecture Notes in Computer Science(), vol 14140. Springer, Cham. https://doi.org/10.1007/978-3-031-55015-7_17
Download citation
DOI: https://doi.org/10.1007/978-3-031-55015-7_17
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-55014-0
Online ISBN: 978-3-031-55015-7
eBook Packages: Computer ScienceComputer Science (R0)