Research explores the impact of automated vehicles on drivers, traffic and infrastructure.

Proponents of automated vehicles say this technology will bring efficiency, safety and environmental benefits to the world's roadways. But the process of introducing these cars into society presents a major unknown: How will automated vehicles interact with their surroundings? 

"Not only will our driving environment change drastically, but we can expect the interaction between automated and regular vehicles to be different from the interaction between regular vehicles," says Dr. Alireza Talebpour, assistant professor in the Zachry Department of Civil Engineering.

By analyzing traffic patterns and developing models to predict driver behavior, Talebpour is addressing the specific challenges of adapting driverless vehicle behavior to complex traffic scenarios. 

Automated driving on campus

At Texas A&M University, civil engineering researchers are testing a driverless vehicle right on campus. The group's automated vehicle research car is part of a study exploring how this technology behaves in everyday driving scenarios, including interactions with regular cars as well as pedestrians. The researchers are also studying how this technology affects individuals' driving habits and driver and pedestrian safety.

Through fine-tuned performance and the elimination of human error, automated cars may significantly streamline everyday transportation. However, some driving situations call for a high degree of responsiveness and maneuverability that is difficult to replicate. The Texas A&M team is using deep neural networks, pattern recognition and data visualization to equip automated vehicles for safety and efficiency in complex environments.

Impact on the built environment 

Automated vehicles will have far-reaching implications for society's infrastructure. For example, automated vehicles are capable of following other vehicles extremely closely, at distances considered unsafe for manned vehicles. Because this feature would drastically increase a road's vehicle capacity without reducing the flow of traffic, the number of cars traveling this road would likely rise. This increased use could potentially affect the longevity of essential infrastructure, such as roads and bridges. 

The vehicles' effects on infrastructure would likely extend beyond transportation systems, too. "How people move around has determined the design and orientation of buildings, roadways and a great deal of urban and suburban development," says Dr. Robin Autenrieth, head of the civil engineering department. "Automated vehicles could significantly influence the future of the entire built environment."

The transition to automation 

Vehicles can have different degrees of automation. The National Highway Traffic Safety Administration identified five levels of automation: Zero describes a car that requires full manual control, while four indicates total automation. A level-four vehicle can perform all maneuvers and reach its destination without the aid of a driver. 

Because the first generation of automated vehicles will most likely share the road with traditional vehicles, full automation is not yet a realistic goal. Most developers are focused on achieving level-three automation, in which the car is capable of performing all safety-critical functions within certain driving conditions, but must cede control to the driver for unexpected or challenging driving situations or maneuvers. To accomplish this, research is underway to develop a reliable mechanism for a safe and reliable transition from automated to regular driving.

"There will be a wealth of possible changes to foresee and acclimate ourselves to before automated vehicles can fully integrate," says Talebpour. "Our research is helping to get a step closer to the introduction of automated vehicles into daily commuting." 

Dr. Alireza Talebpour
Assistant Professor
Zachry Department of Civil Engineering