How can autonomous cars best communicate with drivers?

June 24, 2019
Laboratory simulation of urban, highway and rural route driving at the U-M Transportation Research Institute. Image courtesy: UMTRI.

When working with others, clearly communicating expectations, tasks, and progress can alleviate many issues common in teamwork. What if the teammate is not another human, however, but an autonomous system helping you drive? Do drivers still want to be kept in the loop of a car’s intentions and actions?

To help answer this question, researchers at the University of Michigan led a study that put drivers in a simulation with a talking autonomous vehicle. The findings can help those who are designing autonomous vehicles to make sure drivers will adopt the autonomy and realize their safety benefits.

Continue reading ⇒

A2Sys Lab takes first in firefighting drone competition

April 2, 2019
A2Sys Lab poses with award money
Ella Atkins, Jeremy Castagno, Prince Kuevor, and Matthew Romano won the 2019 Swarm & Search AI Challenge held in Dayton, Ohio over March 29-31, 2019. Photo courtesy A2Sys Lab.

Last year, California experienced the single largest wildfire in its recorded history, a wildfire in Greece killed 100, and wildfires in the British Columbia surpassed the historic proportions seen only the year before. Water and firebreaks can fight immediate threats, but improved mapping and better planning in deploying such resources can maximize impact and minimize risk, reducing the impact of fires over an entire season.

One way to improve mapping and firefighting plans? Unmanned aerial vehicles (UAVs) and algorithms that allow them to operate autonomously.

Continue reading ⇒

Building trust between driverless car and driver

March 14, 2019
an obstacle in the road during a driving simulator
An upcoming obstacle sits in the road during a driving simulation that explores how drivers trust autonomous driving systems. Courtesy Lionel Robert.

If a driver does not trust an autonomous driving system, letting the computer take control can be as daunting as letting a teenager take the wheel. While not trusting a new driver might cause passengers to slam a phantom brake pedal or white-knuckle an arm rest, a driver who does not trust driverless systems might miss out on important safety benefits or even, as autonomous system advance, the ability to complete other tasks.

To improve trust in autonomous systems, researchers at the University of Michigan conducted virtual driving trials that found that the more information an automated driving system communicated about upcoming situations, the higher the level of trust a driver had in the system, and the better the driver performed on a task other than driving.

Continue reading ⇒

Speaking like dolphins, a robot fleet takes on underwater tasks

December 17, 2018
An underwater autonomous robot built to inspect dams, bridges, and hulls of ships practices by inspecting the side of a pool. Courtesy of Joshua Mangelson.

In a Navy shipyard in San Diego, a new generation of underwater robots are learning to communicate and collaborate in order to inspect boats, bridges, pipelines, and other underwater structures. Developed by Joshua Mangelson, a University of Michigan doctoral student in Robotics, the autonomous vehicles overcome the many challenges posed by murky water by simplifying how the robots coordinate and communicate.

Water, while the basis of life for many, means death for wireless communication. “Below a meter or two of water, Wi-Fi cuts out completely,” Mangelson said. “Same with GPS signals. This is because water attenuates electromagnetic signals very quickly, which makes underwater exploration and mapping a very interesting problem.”

Continue reading ⇒

Decentralized air traffic control for drone-laden skies

December 17, 2018

Anticipating skies crowded with crisscrossing autonomous vehicles, University of Michigan researchers have developed a future air traffic control system that allows any number of autonomous planes to safely route around each other to their final destinations.

Kunal Garg, a University of Michigan graduate student, designed the system to be utilized by autonomous fixed-wing aircraft, which require more time and space to turn than rotorcraft such as quadcopters. The work could be extended to autonomous vehicles, which follow similar turning dynamics.

In this simulation, autonomous fixed-wing aircraft change goal points and flight modes to avoid collisions based on control laws developed by Kunal Garg.
Continue reading ⇒