Realization of the Autonomous Driving System on the Experimental Vehicle

Abstract

Running control software on limited computing resources is considered one of the toughest problems. In this study, an autonomous driving software has been developed that can safely complete the map by tracking the lanes and avoiding obstacles on a robot vehicle with limited hardware components. The data was simplified with the image processing technique and the neural network was trained. Overfitting was prevented by hyperparameter tuning and synthetic data augmentation. In order to avoid obstacles, optical flow was calculated by detecting corners every 4 seconds and was used to find the focus of expansion of the vehicle. Time-to-collision was found with the FOE and the distance between the previous position and the current position of the detected point. Optimization was made by averaging the values of close points. The balance mechanism was created according to the TTC difference calculated on the right and left parts of the vehicle.

Keywords

• Convolutional Neural Network
• Overfitting
• Hyperparameter Tuning
• Data augmentation
• Lane tracking
• Optical Flow
• Focus of Expansion
• Time to Collision

References

1. Casser, Vincent, et al. "Depth prediction without the sensors: Leveraging structure for unsupervised learning from monocular videos." Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 33. No. 01. 2019.
2. Google LLC, “Get started with the Dev Board.” 2020.
3. Caruana, Rich, and Alexandru Niculescu-Mizil. "An empirical comparison of supervised learning algorithms." Proceedings of the 23rd international conference on Machine learning. 2006.
4. Zhu, Xiaojin, and Andrew B. Goldberg. "Introduc-tion to semi-supervised learning."Synthesis lec-tures on artificial intelligence and machine learning 3.1 (2009): 1-130.
5. Alpaydin, Ethem. “Introduction to machine learn-ing.” MIT press, 2020.
6. Schalkoff, Robert J. "Pattern recognition." Wiley Encyclopedia of Computer Science and Engineer-ing (2007).
7. Albawi, Saad, Tareq Abed Mohammed, and Saad Al-Zawi. "Understanding of a convolutional neural network."2017 International Conference on Engi-neering and Technology (ICET). Ieee, 2017.
8. O’Donovan, Peter. "Optical flow: Techniques and applications."International Journal of Computer Vision (2005): 1-26.

9. Beauchemin, Steven S., and John L. Barron. "The computation of optical flow."ACM computing sur-veys (CSUR) 27.3 (1995): 433-466.
10. Barron, John L., and Neil A. Thacker. "Tutorial: Computing 2D and 3D optical flow."Imaging sci-ence and biomedical engineering division, medical school, university of manchester 1 (2005).
11. Bounini, Farid, et al. “Autonomous vehicle and real time road lanes detection and tracking.” 2015 IEE Vehicle Power and Propulsion Conference (VPPC). IEEE, 2015
12. Han, J., et al. “Road boundary detection and track-ing for structured and unstructured roads using a 2D lidar sensor.” International Journal of Automo-tive Technology 15.4 (2014): 611-623
13. Souhila, Kahlouche, and Achour Karim. "Optical flow based robot obstacle avoidance."International Journal of Advanced Robotic Systems 4.1 (2007): 2.
14. Saravanan, C. "Color image to grayscale image conversion."2010 Second International Conference on Computer Engineering and Applications. Vol. 2. IEEE, 2010.
15. Xu, Zhao, Xu Baojie, and Wu Guoxin. "Canny edge detection based on Open CV. "2017 13th IEEE in-ternational conference on electronic measurement & instruments (ICEMI). IEEE, 2017.
16. Edwards, Allen L. An introduction to linear regres-sion and correlation. No. 04; QA278. 2, E3 1984.. 1984
17. Montgomery, Douglas C., Elizabeth A. Peck, and G. Geoffrey Vining. Introduction to linear regression analysis. John Wiley & Sons, 2021.
18. Bergstra, James, et al. "Algorithms for hyper-parameter optimization."25th annual conference on neural information processing systems (NIPS 2011). Vol. 24. Neural Information Processing Sys-tems Foundation, 2011.
19. Feurer, Matthias, and Frank Hutter. “Hyperparame-ter optimization.” Automated Machine Learning. Springer, Cham, 2019. 3-33.
20. Mikołajczyk, Agnieszka, and Michał Grochowski.
21. "Data augmentation for improving deep learning in image classification problem."2018 international interdisciplinary PhD workshop (IIPhDW). IEEE, 2018.
22. Srivastava, Nitish, et al. "Dropout: a simple way to prevent neural networks from overfitting." The journal of machine learning research 15.1 (2014): 1929-1958.
23. Ciliberto, Carlo, Lorenzo Rosasco, and Alessandro Rudi. "A consistent regularization approach for structured prediction." Advances in neural infor-mation processing systems 29 (2016): 4412-4420.
24. Hawkins, Douglas M. "The problem of overfitting. "Journal of chemical information and computer sciences 44.1 (2004): 1-12.
25. Takahashi, Ryo, Takashi Matsubara, and Kuniaki Uehara. "Data augmentation using random image cropping and patching for deep cnns." IEEE Trans-actions on Circuits and Systems for Video Tech-nology 30.9 (2019): 2917-2931.
26. Fleet, David, and Yair Weiss. "Optical flow estima-tion." Handbook of mathematical models in com-puter vision. Springer, Boston, MA, 2006. 237-257.
27. Viswanathan, Deepak Geetha. "Features from ac-celerated segment test (fast)." Proceedings of the 10th workshop on Image Analysis for Multimedia Interactive Services, London, UK. 2009.
28. Lucas, B. and Kanade, T. 1981. An iterative image registration technique with an application to stereo vision.In Proc. Seventh International Joint Confer-ence on Artificial Intelligence, Vancouver, Canada, pp. 674–679.
29. Levenberg, Kenneth. "A method for the solution of certain non-linear problems in least squares." Quar-terly of applied mathematics 2.2 (1944): 164-168.
30. Gander, Walter. "Algorithms for the QR decompo-sition." Res. Rep 80.02 (1980): 1251-1268.
31. Satti, Satish Kumar, et al. “A machine learning ap-proach for detecting and tracking road boundary lanes.” ICT Express 7.1 (2021): 99-103