With the rapid development of autonomous driving systems (ADS), especially the increasing adoption of deep neural networks (DNNs) as their core components, effective quality assurance methods for ADS have attracted growing interests in both academia and industry. In this paper, we report a new testing platform ADEPT we have developed, aiming to provide practically realistic and comprehensive testing facilities for DNN-based ADS. ADEPT is based on the virtual simulator CARLA and provides numerous testing facilities such as scene construction, ADS importation, test execution and recording, etc. In particular, ADEPT features two distinguished test scenario generation strategies designed for autonomous driving. First, we make use of real-life accident reports from which we leverage natural language processing to fabricate abundant driving scenarios. Second, we synthesize physically-robust adversarial attacks by taking the feedback of ADS into consideration and thus are able to generate closed-loop test scenarios. The experiments confirm the efficacy of the platform. A video demonstrating the usage of ADEPT can be found at https://youtu.be/evMorf0uR_s.

Sen Wang

Nanjing University

Zhuheng Sheng

Nanjing University

Jingwei Xu

Taolue Chen

University of Surrey, UK

Junjun Zhu

Nanjing University

Shuhui Zhang

Nanjing University

Yuan Yao

Nanjing University

Xiaoxing Ma

Nanjing University

China