LGSVL Simulator: A High Fidelity Simulator for Autonomous Driving

Guodong Rong,Byung Hyun Shin,Hadi Tabatabaee,Qiang Lu,Steve Lemke,Mārtiņš Možeiko,Eric Boise,Geehoon Uhm,Mark Gerow,Shalin Mehta,Eugene Agafonov,Tae Hyung Kim,Eric Sterner,Keunhae Ushiroda,Michael Reyes,Dmitry Zelenkovsky,Seonman Kim
2020-06-22
Abstract:Testing autonomous driving algorithms on real autonomous vehicles is extremely costly and many researchers and developers in the field cannot afford a real car and the corresponding sensors. Although several free and open-source autonomous driving stacks, such as Autoware and Apollo are available, choices of open-source simulators to use with them are limited. In this paper, we introduce the LGSVL Simulator which is a high fidelity simulator for autonomous driving. The simulator engine provides end-to-end, full-stack simulation which is ready to be hooked up to Autoware and Apollo. In addition, simulator tools are provided with the core simulation engine which allow users to easily customize sensors, create new types of controllable objects, replace some modules in the core simulator, and create digital twins of particular environments.
Robotics,Machine Learning,Systems and Control
What problem does this paper attempt to address?
The problem that this paper attempts to solve is that the cost of testing autonomous vehicle algorithms on real vehicles is too high, and many researchers and developers cannot afford the cost of actual vehicles and their corresponding sensors. Although there are free and open - source autonomous driving software stacks such as Autoware and Apollo, there are limited choices of open - source simulators that go with them. Therefore, the paper introduces the LGSVL simulator, which is a high - fidelity autonomous driving simulator designed to solve the above - mentioned problems. Specifically, the LGSVL simulator provides end - to - end, full - stack simulation capabilities and can be seamlessly connected with autonomous driving software stacks such as Autoware and Apollo. In addition, the simulator also provides a series of tools that allow users to easily customize sensors, create new controllable object types, replace certain modules in the core simulator, and create digital twins of specific environments. Through these functions, the LGSVL simulator not only reduces the R & D cost of autonomous driving technology, but also improves the safety and efficiency of testing. It can generate extreme scenarios that are difficult to encounter in reality, support the generation of large - scale data sets, and thus accelerate the development and verification of autonomous driving technology.