- Abstract: We present AutonoVi-Sim, a novel high-fidelity simulation platform for autonomous driving data generation and driving strategy testing. AutonoVi-Sim is a collection of high-level extensible modules which allows the rapid development and testing of vehicle configurations and facilitates construction of complex traffic scenarios. Autonovi-Sim supports multiple vehicles with unique steering or acceleration limits, as well as unique tire parameters and dynamics profiles. Engineers can specify the specific vehicle sensor systems and vary time of day and weather conditions to generate robust data and gain insight into how conditions affect the performance of a particular algorithm. In addition, AutonoVi-Sim supports navigation for non-vehicle traffic participants such as cyclists and pedestrians, allowing engineers to specify routes for these actors, or to create scripted scenarios which place the vehicle in dangerous reactive situations. Autonovi-Sim facilitates training of deep-learning algorithms by enabling data export from the vehicle's sensors, including camera data, LIDAR, relative positions of traffic participants, and detection and classification results. Thus, AutonoVi-Sim allows for the rapid prototyping, development and testing of autonomous driving algorithms under varying vehicle, road, traffic, and weather conditions.
- TL;DR: We present a simulation framework for testing autonomous driving algorithms and generating synthetic data for deep learning approaches. Our simulation framework includes vehicles, pedestrians, and dynamic weather.
- Keywords: motion planning, deep learning, synthetic data, autonomous driving