The University of Michigan has announced the release of the first open-source digital twin of its Mcity Test Facility, a pioneering research center dedicated to advancing connected and autonomous vehicle technologies. This innovative virtual representation of the Mcity Test Facility is now freely available to researchers worldwide, marking a significant milestone in the field of mobility systems testing.
A digital twin functions as a virtual counterpart to a physical environment, allowing for robust simulation and testing by facilitating data exchange between the real and digital realms. With support from the National Science Foundation (NSF), Mcity's newly launched digital twin exemplifies a groundbreaking step forward as the inaugural open-source digital twin specifically designed for mobility system testing. Leveraging this virtual platform, researchers can now conduct simulations that promise enhanced speed, safety, and cost-effectiveness compared to traditional testing methods.
“This takes our almost 10-year-old track and puts the digital replica directly over it,” stated Greg Stevens, Mcity's director of research. He further described the innovation as a 'living, breathing manifestation' of the original track, which enables mixed reality testing and development. Through the digital twin, researchers can access the Mcity Test Facility's diverse features, including various road materials, markings, signals, and intersections, facilitating the trial of autonomous driving algorithms without the need to physically visit Ann Arbor.
At the heart of Mcity's digital twin is TeraSim, an open-source traffic simulator crafted by Mcity researchers. This tool introduces virtual road users such as pedestrians, cyclists, and other vehicles, along with generating critical safety events that mimic potential collisions. Calibrated with real-world data, TeraSim offers the ability to simulate both typical and high-risk driving scenarios, thus providing a comprehensive testing environment.
Originating in 2015, the Mcity Test Facility was envisioned as a purpose-built proving ground essential for the exploration of connected and automated vehicles. The facility boasts a simulated downtown area containing urban streets, a 1,000-foot straightaway, access ramps, a traffic circle, and a myriad of road surfaces, each featuring diverse markings and crossing types. Other key elements include traffic signals, a bridge deck, underpasses, guardrails, barriers, and crash attenuators. As Mcity continues to evolve, the facility is consistently enriched with new features to broaden its reach within the research community.
The development of the digital twin was made possible through a $5.1 million grant awarded by the NSF in 2022, which funded the advancement of Mcity’s digital infrastructure, effectively enabling what is referred to as Mcity 2.0. This grant also facilitated the creation of the digital twin, laying the groundwork for remote testing capabilities worldwide.
Proponents of autonomous and connected vehicle technologies, including various mobility officials, argue that these advancements hold the potential to create safer, more efficient, and equitable modes of transportation. Nonetheless, significant research is still required before such benefits can be widely realized. While real-world testing is both costly and time-consuming, virtual testing—bolstered by digital twin technology—allows for comprehensive trials in which control software can achieve high levels of safety prior to real-world implementation.
“You can drive millions of miles in your AV in a digital twin built off of a real-world environment before your AV actually touches the real world,” explained Darian Hogue, a Mcity software engineer instrumental in the digital twin's development. Digital twins offer users the ability to manipulate numerous variables, such as pedestrian traffic patterns, which remain inherently unpredictable in the physical world. This control accelerates simulated testing, potentially expediting the path to mass deployment of autonomous vehicles.
The debut of Mcity’s remote capabilities in October marked a new era for researchers globally, who can now refine their autonomous driving algorithms by connecting to Mcity's cloud-based digital infrastructure from anywhere. Employing advanced 5G wireless communications, researchers can remotely operate physical vehicles and manipulate traffic signals located within Mcity's test streets, all while receiving real-time data feedback and safeguarding proprietary information.
The introduction of this open-source digital twin underscores Mcity’s commitment to fostering collaborative research and innovation within the autonomous vehicle sector. With this unparalleled virtual resource now widely accessible, the essential testing and development of emerging mobility solutions can proceed with unprecedented scale and efficiency.
.png)
