IATG: Interpretation-analysis-based Testing Method for Autonomous Driving Software

被引：0
|
作者：
Xie R.-L. [1 ]
Cui Z.-Q. [1 ]
Chen X. [2 ]
Zheng L.-W. [1 ]
机构：
[1] School ofcomputer Science, Beijing Information Science and Technology University, Beijing
[2] School of Information Science and Technology, Nantong University, Nantong
来源：
Ruan Jian Xue Bao/Journal of Software | 2024年 / 35卷 / 06期
关键词：
autonomous driving software; deep neural network (DNN); interpretation method; software testing;
D O I：
10.13328/j.cnki.jos.006836
中图分类号：
学科分类号：
摘要：
Autonomous driving software based on deep neural network (DNN) has become the most popular solution. Like traditional software, DNN can also produce incorrect output or unexpected behaviors, and DNN-based autonomous driving software has caused serious accidents, which seriously threaten life and property safety. Therefore, how to effectively test DNN-based autonomous driving software has become an urgent problem. Since it is difficult to predict and understand the behavior of DNNs, traditional software testing methods are no longer applicable. Existing autonomous driving software testing methods are implemented byadding pixel-level perturbations to original images or modifying the whole image to generate test data. The generated test data are quite different from the real images, and the perturbation-based methods are difficult to be understood. To solve the above problem, this study proposes a test data generation method, namely interpretability-analysis-based test data generation (IATG). Firstly, it uses the interpretation method for DNNs to generate visual explanations of decisions made by autonomous driving software and chooses objects in the original images that have significant impacts on the decisions. Then, it generates test data by replacing the chosen objects with other objects with the same semantics. The generated test data are more similar to the real image, and the process is more understandable. As an important part of the autonomous driving software’s decision-making module, the steering angle prediction model is used to conduct experiments. Experimental results show that the introduction of the interpretation method effectively enhances the ability of IATG to mislead the steering angle prediction model. Furthermore, when the misleading angle is the same, the test data generated by IATG are more similar to the real image than DeepTest; IATG has a stronger misleading ability than semSensFuzz, and the interpretation analysis based important object selection method of IATG can effectively improve the misleading ability of semSensFuzz. © 2024 Chinese Academy of Sciences. All rights reserved.
引用
收藏
页码：2753 / 2774
页数：21
相关论文
共 45 条
  • [1] Yu K, Jia L, Chen YQ, Xu W., Deep learning: Yesterday, today, and tomorrow, Journal ofcomputer Research and Development, 50, 9, pp. 1799-1804, (2013)
  • [2] LeCun Y, Bengio Y, Hinton G., Deep learning, Nature, 521, 7553, pp. 436-444, (2015)
  • [3] Chen CY, Seff A, Kornhauser A, Xiao JX., DeepDriving: Learning affordance for direct perception in autonomous driving, Proc. of the 2015 IEEE Int’l Conf. oncomputer Vision, pp. 2722-2730, (2015)
  • [4] Tesla hits a white truck again with two passengers rushing to ICU, (2021)
  • [5] Zhou HS, Li W, Kong ZL, Guo JF, Zhang YQ, Yu B, Zhang LM, Liu C., DeepBillboard: Systematic physical-world testing of autonomous driving systems, Proc. of the 42nd Int’l Conf. on Software Engineering, pp. 347-358, (2020)
  • [6] Tian YC, Pei KX, Jana S, Ray B., DeepTest: Automated testing of deep-neural-network-driven autonomous cars, Proc. of the 40th Int’l Conf. on Software Engineering, pp. 303-314, (2018)
  • [7] Zhang MS, Zhang YQ, Zhang LM, Liu C, Khurshid S., DeepRoad: GAN-based metamorphic testing and input validation framework for autonomous driving systems, Proc. of the 33rd IEEE/ACM Int’l Conf. on Automated Software Engineering, pp. 132-142, (2018)
  • [8] Csurka G, Perronnin F., An efficient approach to semantic segmentation, Int’l Journal ofcomputer Vision, 95, 2, pp. 198-212, (2011)
  • [9] Woodlief T, Elbaum S, Sullivan K., Semantic image fuzzing of AI perception systems, Proc. of the 44th Int’l Conf. on Software Engineering, pp. 1958-1969, (2022)
  • [10] Zhu XL, Wang HC, You HM, Zhang WH, Zhang YY, Liu S, Chen JJ, Wang Z, Li KQ., Survey on testing of intelligent systems in autonomous vehicles, Ruan Jian Xue Bao/Journal of Software, 32, 7, pp. 2056-2077, (2021)
12345