With no driver at the helm, self-driving vehicles are posed with the challenge of effectively communicating with their surroundings.

Drive.ai has addressed that challenge head-on, equipping its self-driving vehicles with external communication panels that convey a variety of messages to properly communicate with drivers, pedestrians, cyclists, and everyone else on the road.

“Our external communication panels are intended to mimic what an interaction with a human driver would look like. Normally, you’d make eye contact, wave someone along, or otherwise signal your intentions,” Drive.ai CEO Bijit Halder tells AUVSI via email.

“With our panels, everyone on the road is kept in the loop of the car’s intentions, ensuring maximum comfort and trust, even for people interacting with a self-driving car for the first time.”

According to Halder, Drive.ai knew early on that vehicle communication needed to be a priority to build transparency and trust with the public. The company believes that this will ultimately drive greater adoption of autonomous vehicles, and lead to safer roads.

To help the company achieve its goal of building vehicles that could properly communicate with the public, one of Drive.ai’s founders recorded herself driving around normally and analyzed all the interactions she had with other drivers, including eye contact and hand motions.

“This early research on communication between human drivers ultimately informed what we wanted to convey with the communication panels,” Halder says.

“From then on, Drive.ai has iterated designs to evolve and improve the messages shown.”

Communicating with pedestrians

Using its suite of lidar sensors, cameras, and radar, Drive.ai’s vehicle determines if any pedestrians are in or near a crosswalk it is approaching.

If the vehicle detects a pedestrian’s path will intersect the vehicle path, the car begins to slow down, preparing to stop, and the panel displays the message “Stopping for you.”

“This panel is especially important, as our internal studies found that 92 percent of pedestrians do not wait for vehicles to come to a complete stop before starting to cross,” Halder says. “Signaling that we see them and are slowing for them to cross helps support natural human behavior, which is always our goal.”

Once the vehicle comes to a complete stop for pedestrians, the panel will change to “Waiting for you,” to tell pedestrians that the vehicle is stopped as they cross safely.

Additionally, the vehicle’s rear panel says “Pedestrian crossing” so that vehicles behind Drive.ai’s vehicle know that it is stopped for safety, so they won’t try to drive around Drive.ai’s vehicle.

When no more pedestrians are detected in the crosswalk, the panel will change to “Going now, please wait,” indicating to any nearby pedestrians that the vehicle plans to move forward soon, so it’s best to wait until it passes.

“Safety is our top priority, and by communicating our vehicle’s intent with these panels, pedestrians can make the safest, most informed choice when crossing the street,” Halder says.

Shift in approach

Over the years, Drive.ai has shifted its approach to its communication panels to ensure its vehicles are communicating in the most effective way possible. The screens were initially meant to be high up on top of the car, but Drive.ai’s research showed that most people look at a car’s wheels to see when and how it will start moving. Equipped with this knowledge, Drive.ai opted to move the panels down to the sides and front of the car.

“This told us that people would be more likely to see our signs if they were closer to the wheels, and encourage a safer and more intuitive interaction between human drivers and self-driving cars on the roads,” Halder says.

Halder says the company has also evolved its external communication panels to figure out which images and messages are the most effective and easiest to read.

“It’s very important that the messages are universally and immediately understood by pedestrians and other drivers,” Halder says.

Drive.ai conducts iterative testing with people to continually refine the panels’ graphics and wording, so that it can find the “sweet spot” between too many words and too few.

For example, “Waiting for you to cross” forced text to be smaller, while “Waiting” by itself was less clear. The company ultimately settled on “Waiting for you,” because it “both conveyed the message reliably and quickly,” Halder says.

During the testing process, Drive.ai also evaluates whether the panels communicate effectively without relying on text, to accommodate non-English speakers. Drive.ai also employs animation instead of static images to better communicate meaning and to grab visual attention.

Drive.ai also grabs visual attention with its Safety Orange paint on its vehicles. Halder says the paint was added later to make the vehicle stand out.

“Other companies in the self-driving industry are black or white — aiming to blend in or be discreet,” Halder says. “Instead, we wanted to stand out, given AVs are still an emerging technology and not something people are used to (or should be afraid of!), with the aim of promoting awareness and transparency.”

Drive.ai has worked to promote awareness and transparency of autonomous vehicles not only through the design and layout of its vehicles, but also through real-world testing of its technology. Since last year, the company has been testing its vehicles in Texas, having recently finished testing in Frisco, Texas, where it received overwhelmingly positive reviews.

According to Halder, more than 98 percent of riders said they felt safe during a ride in a Drive.ai vehicle. More than 96 percent said they were happy with the overall service experience, and almost 94 percent said that they would ride in a Drive.ai self-driving car again.

Halder adds that Drive.ai’s research showed that virtually all of the messages displayed by the company’s external communication panels have been understood by 100 percent of respondents.

“The more you can effectively communicate how a self-driving car will act, the more confidence the public will have in the technology, and that trust will lead to adoption on a broader scale,” Halder says.