Zhi Hao Lin
Yuan Shen
Bhargav Chandaka
We present Sim-on-Wheels, a safe, realistic, and vehicle-in-loop framework to test autonomous vehicles’ performance in the real world under safety-critical scenarios. Sim-on-wheels runs on a self-driving vehicle operating in the physical world. It creates virtual traffic participants with risky behaviors and seamlessly inserts the virtual events into images perceived from the physical world in real-time. The manipulated images are fed into autonomy, allowing the self-driving vehicle to react to such virtual
Sim-on-Wheels is highly flexible and generates scenarios with different difficulty levels. The scenarios include static human/objects, a traffic intersection, and jaywalkers with different speeds and occluding walls.
We can easily adapt to any environment as long as the scenario is compatible with the environment mapping. For example, we ran our Sim-on-Wheels on KITTI-360 data to simulate scenarios on the street roads as shown below:
Zhi Hao Lin
Yuan Shen
Bhargav Chandaka