A new study from Coventry University is answering a crucial question about the future of self-driving cars: How many automated vehicles (AVs) can one person effectively supervise at the same time? The research, published in Computers in Human Behavior, found that an individual can best manage between five and seven vehicles, with five being the optimal number for maintaining safety and decision-making.
How the Research Was Done?
Scientists designed an experiment focused specifically on the monitoring task, rather than direct interaction with AVs. Twenty-four volunteers acted as remote monitors for groups of three, five, seven, or nine simulated self-driving taxis navigating a realistic model of Coventry. Their job was to watch, assess, and, if necessary, hand over problematic vehicles to remote drivers. The team measured how quickly and accurately the monitors responded to events, as well as their workload and situation awareness.
Main Discoveries:
- Best Results at Five Cars: The study found maximum situation awareness and fastest reaction times when participants were monitoring five AVs. Supervisors could temporarily oversee up to nine cars during busy periods, but their ability to catch problems and make good decisions dropped with higher numbers.
- More Isn’t Always Better: Some monitors preferred having nine screens to watch, saying it kept them alert. However, with more AVs, people felt mentally stretched, missed important messages, and called for help less often than needed.
- Problems with Too Few Cars: Supervising only three vehicles led to micromanagement and unnecessary interventions, likely because people had spare mental capacity and overanalyzed simple situations.
What About Developing Cities?
While these findings offer a helpful guideline for modern cities with regulated traffic, the situation is more complicated in developing countries and older urban areas. In many developing cities, traffic rules are often ignored, road surfaces are poor, and roads are crowded with all types of vehicles and pedestrians. Unlike the well-designed highways in newer cities, historic cities have narrow, congested roads full of unpredictable challenges.
This means remote operators supervising self-driving cars in such environments would face far greater complexities. They would need to watch out for sudden lane changes, vehicles driving against traffic, unexpected street vendors, livestock, and poorly marked intersections. The constant need to interpret erratic behavior and overcome lack of signage or road markings increases mental workload and the risk of missing critical incidents. As a result, it is likely that the optimal number of AVs one operator could safely supervise would be lower in these settings, and control centers might need to adapt with better training, local knowledge, and advanced alert systems to ensure safe and effective monitoring.
Why It Matters?
As self-driving cars become more common, companies need to know how to staff remote monitoring centers for safety and efficiency. This study suggests that assigning each supervisor five vehicles is optimal, with flexibility for short-term surges. However, the number should be adjusted based on local road realities. Good communication from vehicles and smart interfaces are vital, especially in cities with challenging driving conditions. The study’s results can help cities, companies, and planners create safer and more responsive fleets of automated taxis and delivery vehicles worldwide.
