Summary of Dance Of the Ads: Orchestrating Failures Through Historically-informed Scenario Fuzzing, by Tong Wang and Taotao Gu and Huan Deng and Hu Li and Xiaohui Kuang and Gang Zhao
Dance of the ADS: Orchestrating Failures through Historically-Informed Scenario Fuzzing
by Tong Wang, Taotao Gu, Huan Deng, Hu Li, Xiaohui Kuang, Gang Zhao
First submitted to arxiv on: 5 Jul 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Neural and Evolutionary Computing (cs.NE); Software Engineering (cs.SE)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A machine learning-based approach called ScenarioFuzz is introduced for verifying the safety of autonomous driving systems (ADS) without relying on predefined scenarios. The methodology leverages map road networks and historical test data to extract essential information, which is then used to form a foundational scenario seed corpus. This corpus is enriched with pertinent details and provides the necessary boundaries for fuzz testing. The approach integrates mutators, mutation techniques, and a graph neural network model to predict and filter out high-risk scenario seeds, optimizing the fuzzing process. Compared to other methods, ScenarioFuzz reduces time costs by an average of 60.3% while increasing the number of error scenarios discovered per unit of time by 103%. The paper also proposes a self-supervised collision trajectory clustering method that identifies and summarizes 54 high-risk scenario categories prone to inducing ADS faults. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary ScenarioFuzz is a new way to test autonomous driving systems (ADS) to make sure they are safe. It works by looking at maps of roads and using that information, along with past testing data, to create scenarios for the system to try. The approach uses special algorithms and computer models to predict which scenarios might cause problems and filters out the ones that won’t. This makes the testing process more efficient and helps find more errors in the system. In fact, tests showed that ScenarioFuzz can reduce the time it takes to test by 60% while finding 103% more errors than other methods. The paper also proposes a new way to group together scenarios that might cause problems, helping developers create safer ADS. |
Keywords
» Artificial intelligence » Clustering » Graph neural network » Machine learning » Self supervised
