Apr 2, 2019


Electrification, ride-sharing and autonomous vehicles are the next big change in the world of mobility. It diminishes traffic and congestion, leads to safer roads and improves urban areas. However, getting a new technology to market takes time. It takes even longer if that technology has anything to do with people’s safety. You want to make sure the people interacting with your product are as safe as possible—and that takes a lot of simulation, testing and readjusting.

For example, imagine you’re developing an autonomous shuttle system for a college campus. In addition to being safe, it needs to be effective and efficient to get students to class on time. But campuses are notorious for their huge fluctuations in vehicle and foot traffic: In the early mornings or the summer, they can be ghost towns; yet you’ll see throngs of people, bicycles, and cars just before the start of the most popular class times.

Input from multiple types of sensors, when processed by an excellent autonomy system, is a great start in navigating those complex situations. But “training” the autonomous vehicle to deal with them could take more time than you’d like and releasing the vehicle to the market too soon presents a liability you don’t need.

That’s where teleoperation comes in. A highly trained, off-site human can use all the sensor data your autonomy software uses but make decisions it can’t or hasn’t yet learned to and act on them to keep the passengers and surrounding humans safe. Consider these three examples:

Campus safety officers
An autonomous vehicle can navigate normal campus conditions safely. But it may come to a stop when it detects a human in the road, not recognizing that it is a campus police officer directing traffic. A teleoperator can recognize the hand signals and takeover the controls long enough to appropriately guide the shuttle safely through the situation until the coast is clear.

College campuses are subject to the same extremes of weather as everywhere else. That means that your automated shuttle needs to be able to deal with high winds, heavy rain, fog and other driving challenges presented by Mother Nature. An IMU and other sensors can help the car deal with weather-related risks, but it would take quite a while to learn all of them. Better to have a teleoperator with maps and screens standing by to help navigate if things get too complex and determine when it’s just not safe for the shuttle to continue operation.

The human element
Campuses may be hesitant to invest in shuttles that have no human element. According to AAA’s annual automated vehicle survey, 71 percent of people are afraid to ride in fully self-driving vehicles. AAA believes the key to helping consumers feel more comfortable with fully self-driving vehicles will be bridging the gap between the perception of automated vehicle technology and the reality of how it actually works in today’s cars. Knowing that there is an actual human being there to keep them safe represents an instrumental value and can allay passengers’ concerns about riding in an autonomous vehicle.

Our teleoperation software can fit seamlessly into an autonomous shuttle’s existing technology, making it easier to get them to market and doing what they’re designed to do: help college campuses run more efficiently and safely.

Learn more about Designated Driver and our approach to teleoperation services.