Researchers at Chalmers University of Technology in Sweden have developed a new optimization method that could make large fleets of autonomous electric taxis more efficient and reliable, according to a report by Chinese automotive news outlet D1EV.
The method is designed to handle the complex task of dispatching and managing robotaxi fleets while accounting for the uncertainties inherent in real traffic systems. Autonomous electric ride-hailing services, also known as "on-demand autonomous electric mobility," have the potential to reduce the cost of individual trips by up to 60% compared with traditional taxis, and significantly cut emissions per trip, the report said.
Balancing supply and demand
Managing such fleets is extremely complex. In a robotaxi operation, urban vehicle networks can easily fall out of balance between supply and demand. For example, during the morning rush hour, people want to travel to the city center, but in the afternoon the pattern reverses. To meet peak demand, autonomous vehicles often need to drive empty to pick up passengers. If the vehicles are electric, they also need to stop to recharge—but deciding where to charge and for how long is a tricky problem, said Sten Elling Tingstad Jacobsen, a doctoral student in control engineering at Chalmers' Department of Electrical Engineering.
The new optimization approach aims to achieve near-optimal dispatch and control of large-scale fleets while fully considering various uncertainties in real traffic systems. The research was published in the academic journal Transportation Research Part C: Emerging Technologies.
Implications for China's EV industry
While the research originates in Sweden, it has direct relevance to China's rapidly developing autonomous driving sector. Chinese companies such as Baidu, Pony.ai, and AutoX are already operating robotaxi services in several cities, and the optimization of fleet management is a key challenge for scaling these operations. The findings could inform the development of more efficient fleet management algorithms for Chinese autonomous ride-hailing services, potentially reducing costs and improving service reliability.
According to D1EV, the research underscores the importance of advanced control and optimization techniques in making autonomous electric mobility a viable and sustainable transportation option. As the industry moves toward commercialization, such methods will be crucial in balancing efficiency, cost, and environmental benefits.
