Research team drives autonomous vehicle technology forward
Professors Bilin Aksun-Guvenc and Levent Guvenc of the Department of Mechanical and Aerospace Engineering are taking autonomous vehicles from the drawing board to the streets.
Their latest research is making driving more convenient for passengers, eliminating the challenges facing public transportation riders, and — most importantly — keeping those inside and outside of self-driving vehicles safe. To accomplish this, the city where the Guvencs’ research is being conducted had to first be transformed.
In June 2016, Columbus was named the winner of the U.S. Department of Transportation’s inaugural Smart City Challenge. The competition spurred cities across the nation to demonstrate their best vision for how advanced data, applications and transportation systems can be used in their regions. The Guvencs’ innovative research played a central role in making Ohio’s capital city the ideal epicenter for intelligent transportation testing.
The newly established Automated Driving Lab (ADL), directed by MAE alumnus and faculty Levent (‘92 PhD, mechanical engineering), leads the Smart City charge by using two state-of-the-art autonomous research vehicles and a hardware-in-the-loop simulator to study and develop connected and autonomous driving functions. Bilin’s socially acceptable collision avoidance (SACA) algorithm ensures that these vehicles respect the personal space of pedestrians, bikers and other cars on the road while maneuvering a collision-free path.
The Guvencs’ aim is simple: To develop a unified, replicable and scalable driving-assistance architecture that can be used throughout the U.S. to help people and goods move quickly, safely, cheaply and efficiently. Their approach utilizes interfaces for typical sensors like GPS, camera, LIDAR, RADAR and V2V modem, generic steering, throttle and brake actuators, all within a multi-agent automated driving architecture connected by generic and scalable control and decision-making blocks.
The ADL sets itself apart by partnering with the nation’s largest independent vehicle test facility: the Transportation Research Center (TRC). This one-of-a-kind center offers the lab a 7.5-mile high-speed oval test track, a 50-acre Vehicles Dynamics Area and more than 4,000 acres of road courses. “Without even going to the road, we are able to run the same experiment because we have validated models for both vehicles,” said Levent.
The Guvencs’ research isn’t just pushing driving technologies forward, it’s having an impact on the city’s most vulnerable populations.
“In Columbus, mobility directly impacts low-income areas, such as the Linden neighborhood,” said Bilin. “The infant mortality rate in this community is affected by mothers’ inability to find a way to the nearest hospital.”
While public transportation can be quite convenient, it is not perfect. The mile or so that individuals need to walk from their homes to their first bus stop and from their last stop to their final destination is the problem. This “first mile, last mile” concept is currently one of the weakest links in smart mobility.
“We don’t conduct [our research] for the sake of technology, but to help people,” said Bilin. “We want to make sure driving is safe and that no pedestrian or passenger in the car is hurt.”
With a Smart City grant from the National Science Foundation, the Guvencs and their team of graduate student researchers are developing low-speed on-demand automated shuttle technology to help solve this very challenge. This developing research blends very well with the city’s emerging Smart Columbus bus rapid transport network, which would allow buses to bypass traffic lights. The Guvencs’ autonomous electric shuttles are planned for operation in the city’s 750,000-square-foot Easton Town Center shopping complex.
Their unique approach has the ability to work in similarly sized urban and traffic-dense environments. Through the National Institute of Standards and Technology’s Global City Teams Challenge, the ADL and the city of Columbus have already connected with Boston; Portland, Oregon; Greenville, South Carolina; Washington D.C.; and Madison, Wisconsin as Smart Shuttle technical cluster collaborators.
Scalable. Replicable. And, capable of making transportation safe, efficient and fast. The Guvencs' work will have a long-lasting effect on Columbus and "smart cities" across the nation.
by Kam King, Department of Mechanical and Aerospace Engineering