Selection of Lightweight CNN Models with Limited Computing Resources for Drone Collision Prediction

Main Article Content

Rifqi Nabila Zufar
David Banjerdpongchai

Abstract

The Collision Avoidance System (CAS) is a safety system created to identify and prevent collisions, primarily on drones. The CAS comprises three processes: detection, prediction, and action. The predictive process is crucial as it determines whether a collision will occur, making it the core component of the system. Most drones are equipped with cameras. A visual-based prediction involves the use of a convolutional neural network (CNN). The CNN operates by autonomously learning and extracting hierarchical characteristics from input data through convolution, pooling, and fully connected layers. Currently, there are CNN models called pretrained models that are ready to use. However, not all pretrained models are suitable for compatibility with drones as they possess computational constraints. Our objective is to establish a suitable model selection from a variety of pre-trained CNN models with lightweight architectures. The transfer learning technique is applied to customize these models with the ColANet dataset. Subsequently, we evaluate these models regarding their accuracy, model size, inference time, and power consumption. Finally, the selected model is deployed in real time on a Raspberry Pi 3B+ with data input from a DJI Tello drone camera, and the prediction performance is evaluated.

Article Details

How to Cite
Zufar, R. N. ., & Banjerdpongchai, D. (2024). Selection of Lightweight CNN Models with Limited Computing Resources for Drone Collision Prediction. ECTI Transactions on Electrical Engineering, Electronics, and Communications, 22(1). https://doi.org/10.37936/ecti-eec.2024221.251164
Section
Research Article

References

Z. Xiaoning, “Analysis of military application of UAV swarm technology,” 2020 3rd International Conference on Unmanned Systems (ICUS), Nov. 2020, doi: 10.1109/icus50048.2020.9274974.

A. Chriki, H. Touati, H. Snoussi, and F. Kamoun, “UAV-based Surveillance System: An Anomaly Detection Approach,” 2020 IEEE Symposium on Computers and Communications (ISCC), Jul. 2020, doi: 10.1109/iscc50000.2020.9219585.

P. Radoglou-Grammatikis, P. Sarigiannidis, T. Lagkas, and I. Moscholios, “A compilation of UAV applications for precision agriculture,” Computer Networks, p. 107148, May 2020, doi: 10.1016/j.comnet.2020.107148.

J. Wu, S. Wen, Y. Lan, X. Yin, J. Zhang, and Y. Ge, “Estimation of cotton canopy parameters based on unmanned aerial vehicle (UAV) oblique photography,” Plant Methods, no. 1, Dec. 2022, doi: 10.1186/s13007-022-00966-z.

G. Skorobogatov, C. Barrado, and E. Salamí, “Multiple UAV Systems: A Survey,” Unmanned Systems, no. 02, pp. 149–169, Apr. 2020, doi: 10.1142/s2301385020500090.

B. J. Brelje and J. R. R. A. Martins, “Electric, hybrid, and turboelectric fixed-wing aircraft: A review of concepts, models, and design approaches,” Progress in Aerospace Sciences, pp. 1–19, Jan. 2019, doi: 10.1016/j.paerosci.2018.06.004.

Q. Quan, Introduction to Multicopter Design and Control. Springer, 2017.

M. Hassanalian and A. Abdelkefi, “Classifications, applications, and design challenges of drones: A review,” Progress in Aerospace Sciences, pp. 99–131, May 2017, doi: 10.1016/j.paerosci.2017.04.003.

J. N. Yasin, S. A. S. Mohamed, M.-H. Haghbayan, J. Heikkonen, H. Tenhunen, and J. Plosila, “Unmanned Aerial Vehicles (UAVs): Collision Avoidance Systems and Approaches,” IEEE Access, pp. 105139–105155, 2020, doi: 10.1109/access.2020.3000064.

Z. Wei, Z. Meng, M. Lai, H. Wu, J. Han, and Z. Feng, “Anti-Collision Technologies for Unmanned Aerial Vehicles: Recent Advances and Future Trends,” IEEE Internet of Things Journal, no. 10, pp. 7619–7638, May 2022, doi: 10.1109/jiot.2021.3135578.

F. Zhuang et al., “A Comprehensive Survey on Transfer Learning,” Proceedings of the IEEE, no. 1, pp. 43–76, Jan. 2021, doi: 10.1109/jproc.2020.3004555.

R. N. Zufar and D. Banjerdpongchai, “Performance Comparison of Lightweight CNN Models for Drone Collision Avoidance Dataset,” 2023 20th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications, and Information Technology (ECTI-CON), May 2023, doi: 10.1109/ecti-con58255.2023.10153233.

D. Pedro, J. P. Matos-Carvalho, J. M. Fonseca, and A. Mora, “Collision Avoidance on Unmanned Aerial Vehicles Using Neural Network Pipelines and Flow Clustering Techniques,” Remote Sensing, no. 13, p. 2643, Jul. 2021, doi: 10.3390/rs13132643.

D. Falanga, S. Kim, and D. Scaramuzza, “How Fast Is Too Fast? The Role of Perception Latency in High-Speed Sense and Avoid,” IEEE Robotics and Automation Letters, no. 2, pp. 1884–1891, Apr. 2019, doi: 10.1109/lra.2019.2898117.

ColANet: A UAV Collision Avoidance Dataset. https://colanet.qa.pdmfc.com/ (accessed Sep. 10, 2023).

D. Pedro, A. Mora, J. Carvalho, F. Azevedo, and J. Fonseca, “ColANet: A UAV Collision Avoidance Dataset,” in IFIP Advances in Information and Communication Technology, Springer International Publishing, pp. 53–62, 2020.

E. D. Cubuk, B. Zoph, J. Shlens, and Q. V. Le, “Randaugment: Practical automated data augmentation with a reduced search space,” 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Jun. 2020, doi: 10.1109/cvprw50498.2020.00359.

Keras Team, “Keras Applications,” Keras: Deep Learning for humans. https://keras.io/api/applications/ (accessed Sep. 09, 2023).

A. Abid, P. Sinha, A. Harpale, J. Gichoya, and S. Purkayastha, “Optimizing Medical Image Classification Models for Edge Devices,” in Distributed Computing and Artificial Intelligence, Volume 1: 18th International Conference, Springer International Publishing, pp. 77–87, 2021.

R. P. Ltd, “Buy a Raspberry Pi 3 Model B+ – Raspberry Pi,” Raspberry Pi. https://www.raspberrypi.com/products/raspberry-pi-3-model-b-plus/ (accessed Sep. 23, 2023).

“Tello,” Tello Official Website-Shenzhen Ryze Technology Co., Ltd. https://www.ryzerobotics.com/tello/specs (accessed Sep. 19, 2023).

“3.3. Metrics and scoring: quantifying the quality of predictions — scikit-learn 1.3.0 documentation,” scikit-learn. https://scikit-learn.org/stable/modules/model_evaluation.html (accessed Sep. 9, 2023).

L. Yin, X. Chen, Z. Qin, Z. Zhang, J. Feng, and D. Li, “An Experimental Perspective for Computation-Efficient Neural Networks Training,” in Communications in Computer and Information Science, Springer Singapore, pp. 168–178, 2018.

Y. Liu, Y. Wang, R. Yu, M. Li, V. Sharma, and Y. Wang. “Optimizing {CNN} model inference on {CPUs},” In 2019 USENIX Annual Technical Conference (USENIX ATC 19), pp. 1025-1040, 2019.

“Energy Efficiency Guide for Mac Apps: Monitor Usage Regularly,” Apple Developer, Sep. 13, 2016. https://developer.apple.com/library/archive/documentation/Performance/Conceptual/power_efficiency_guidelines_osx/MonitoringEnergyUsage.html (accessed Sep. 9, 2023).