Researchers Develop Statistically Realistic Simulation for Autonomous Vehicle Testing

Category Machine Learning

tldr #

The University of Michigan researchers developed a simulation of driving conditions to test autonomous vehicles called the Neural Naturalistic Driving Environment (NeuralNDE). The Unique AI-approach used in the model allows it to generate safety-critical events a thousand times more frequently than they would usually occur in real driving, helping to ensure the safety of AVs before other cars, cyclists, and pedestrians ever cross its path.

content #

The first statistically realistic roadway simulation has been developed by researchers at the University of Michigan. While it currently represents a particularly perilous roundabout, future work will expand it to include other driving situations for testing autonomous vehicle software.

The simulation is a machine-learning model that trained on data collected at a roundabout on the south side of Ann Arbor, recognized as one of the most crash-prone intersections in the state of Michigan and conveniently just a few miles from the offices of the research team.

The NeuralNDE is able to recreate safety-critical events a thousand times more frequently than they occur in real driving.

Known as the Neural Naturalistic Driving Environment or NeuralNDE, it turned that data into a simulation of what drivers experience everyday. Virtual roadways like this are needed to ensure the safety of autonomous vehicle software before other cars, cyclists and pedestrians ever cross its path.

"The NeuralNDE reproduces the driving environment and, more importantly, realistically simulates these safety-critical situations so we can evaluate the safety performance of autonomous vehicles," said Henry Liu, U-M professor of civil engineering and director of Mcity, a U-M-led public-private mobility research partnership.

The machine-learning model used by NeuralNDE is called bayesian machine learning.

Liu is also director of Center for Connected and Automated Transportation and corresponding author of the study in Nature Communications. The publication has been featured as an Editor's Highlight.

Safety critical events, which require a driver to make split-second decisions and take action, don't happen that often. Drivers can go many hours between events that force them to slam on the brakes or swerve to avoid a collision, and each event has its own unique circumstances.

The roundabout featured in the NeuralNDE is known as the State Street/Ellsworth Road roundabout in Ann Arbor.

Together, these represent two bottlenecks in the effort to simulate our roadways, known as the "curse of rarity" and the "curse of dimensionality" respectively. The curse of dimensionality is caused by the complexity of the driving environment, which includes factors like pavement quality, the current weather conditions, and the different types of road users including pedestrians and bicyclists.

To model it all, the team tried to see it all. They installed sensor systems on light poles which continuously collect data at the State Street/Ellsworth Road roundabout.

SAFE TEST, the system developed by the university, only needs about 0.01% of the testing miles required to ensure a safe autonomous vehicle experience.

"The reason that we chose that location is that roundabouts are a very challenging, urban driving scenario for autonomous vehicles. In a roundabout, drivers are required to spontaneously negotiate and cooperate with other drivers moving through the intersection. In addition, this particular roundabout experiences high traffic volume and is two lanes, which adds to its complexity," said Xintao Yan, a Ph.D. student in civil and environmental engineering and first author of the study, who is advised by Liu.

Mcity is a public-private mobility research partnership started at the University of Michigan.

The NeuralNDE serves as a key component of the CCAT Safe AI Framework for Trustworthy Edge Scenario Tests, or SAFE TEST, a system developed by Liu's team that uses artificial intelligence to reduce the testing miles required to ensure the safety of autonomous vehicles by 99.99%.

It essentially breaks the "curse of rarity," introducing safety-critical incidents a thousand times more frequently than they occur in real driving. The NeuralNDE is also critical to a project designed to enable the Mcity Test Facility to be used for remote testing of AV software.

The publication of the NeuralNDE's results was featured as an Editor's Highlight by Nature Communications.

But unlike a traditional machine-learning model, which outputs a prediction containing little insight as to how it arrived at that answer, the team developed a model using bayesian machine learning. It builds on previous work from Yan and Liu that applied bayesian machine learning to traffic incident detection.

hashtags #
worddensity #