Extreme Scenario Generation Based on Adversarial Attack

The transportation field requires a large number of simulation scenarios for testing. At present, there is relatively little research on the generation of extreme scenarios. In this paper, we give the definition of extreme scenarios, which are prone to problems, and divide them into two categories: the extreme scenarios based on primitive value and the extreme scenarios based on primitive coupling. This paper focuses on the second which considers the coupling effect of different primitives in the scenarios, using the methods of adversarial attack: FGSM, FGSM-target, BIM, ILCM, PGD and strategically-timed attack. Using vehicle agent for test, the first five methods prove the feasibility and effectiveness of extreme scenario generation, and the sixth method simplifies the generation process.

[1]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[2]  Ming-Yu Liu,et al.  Tactics of Adversarial Attack on Deep Reinforcement Learning Agents , 2017, IJCAI.

[3]  Nico Kaempchen,et al.  Highly Automated Driving on Freeways in Real Traffic Using a Probabilistic Framework , 2012, IEEE Transactions on Intelligent Transportation Systems.

[4]  Samy Bengio,et al.  Adversarial Machine Learning at Scale , 2016, ICLR.

[5]  Tom Schaul,et al.  Dueling Network Architectures for Deep Reinforcement Learning , 2015, ICML.

[6]  Markus Maurer,et al.  Autonomous Vehicle Guidance on Braunschweig's inner ring road within the Stadtpilot Project , 2011, 2011 IEEE Intelligent Vehicles Symposium (IV).

[7]  William Whittaker,et al.  Autonomous driving in urban environments: Boss and the Urban Challenge , 2008, J. Field Robotics.

[8]  David Silver,et al.  Deep Reinforcement Learning with Double Q-Learning , 2015, AAAI.

[9]  Simon M. Lucas,et al.  A Survey of Monte Carlo Tree Search Methods , 2012, IEEE Transactions on Computational Intelligence and AI in Games.

[10]  Satoshi Masuda.  Software Testing Design Techniques Used in Automated Vehicle Simulations , 2017, 2017 IEEE International Conference on Software Testing, Verification and Validation Workshops (ICSTW).

[11]  Mohamed Benmimoun.  Effective Evaluation of Automated Driving Systems , 2017 .

[12]  Samy Bengio,et al.  Adversarial examples in the physical world , 2016, ICLR.

[13]  Mikhail Pavlov,et al.  Deep Attention Recurrent Q-Network , 2015, ArXiv.

[14]  Frank Barickman,et al.  Key Considerations in the Development of Driving Automation Systems , 2015 .

[15]  Jonathon Shlens,et al.  Explaining and Harnessing Adversarial Examples , 2014, ICLR.

[16]  Nidhi Kalra,et al.  Driving to Safety , 2016 .

[17]  Nidhi Kalra,et al.  Autonomous Vehicle Technology: A Guide for Policymakers , 2014 .

[18]  Aleksander Madry,et al.  Towards Deep Learning Models Resistant to Adversarial Attacks , 2017, ICLR.