AutoSceCraft: Generate Various Driving Scenarios from Scratch for Autonomous Driving Systems
Authors:
- Wenxing LanWLWenxing Lan
- Jialin LiuJLJialin Liu
- Bo YuanBYBo Yuan
- Xin YaoXYXin Yao
Open Access
- 1 April 2026
- journal article
- Published by Tsinghua University Press in Tsinghua Science and Technology
- Vol. 31 (2) , 1282-1305
Abstract
Autonomous Driving Systems (ADS) are safety-critical. Abundant and various driving scenarios are required to train accurate and robust models, and comprehensively test each module of autonomous driving systems (i.e., perception, tracking, prediction, planning, and control modules). However, collecting driving scenario data from the real physical world is expensive and inefficient. Most existing works generate simulated driving scenarios by varying the behaviors of dynamic objects on simple road networks (e.g., highways), while the influence of roadside structures and scenarios with complex road networks are not considered. This paper proposes a novel driving scenario generation approach, Automated Scenario Crafting (AutoSceCraft), to automatically produce abundant driving scenarios containing various road networks, traffic rules, roadside structures, and dynamic objects at low cost. To validate the effectiveness and efficiency of our proposed framework, AutoSceCraft is integrated into three popular driving simulators, including SMARTS, esmini, and CARLA. Numerical experiments and scenario visualization results show that AutoSceCraft can generate effectively and efficiently various driving scenarios from scratch for testing and training various modules (including perception, prediction, and planning modules) within autonomous driving systems.Keywords
This publication has 42 references indexed in Scilit:
- GLENet: Boosting 3D Object Detectors with Generative Label Uncertainty EstimationInternational Journal of Computer Vision, 2023
- Learning naturalistic driving environment with statistical realismNature Communications, 2023
- Dense reinforcement learning for safety validation of autonomous vehiclesNature, 2023
- A Deep Learning Framework for Generation and Analysis of Driving Scenario TrajectoriesSN Computer Science, 2023
- 6-Layer Model for a Structured Description and Categorization of Urban Traffic and EnvironmentIEEE Access, 2021
- Intelligent driving intelligence test for autonomous vehicles with naturalistic and adversarial environmentNature Communications, 2021
- A survey of deep learning techniques for autonomous drivingJournal of Field Robotics, 2019
- On the Hermite Interpolation Problem with clothoidsJournal of Computational and Applied Mathematics, 2018
- Are we ready for autonomous driving? The KITTI vision benchmark suite2012
- Search-Based Procedural Content Generation2010