A new approach from KAIST enables electric vehicles to make smarter driving decisions by looking far ahead on the road. Using route-wide information shared through V2X, the system helps vehicles save energy—almost as efficiently as if they could see the future.

Imagine a car approaching a red light—but instead of stopping, it keeps going smoothly, knowing the light will turn green just in time. By avoiding unnecessary braking and acceleration, it saves energy and arrives more efficiently. This is the idea behind optimal control: making the smartest driving decisions by planning ahead.

This leads us to predictive control—a method that helps electric vehicles, hybrids, and fuel cell vehicles plan their energy use based on what's coming up on the road. The key enabler is V2X, or vehicle-to-everything communication. V2X allows vehicles to receive data from their surroundings:

● V2I (vehicle-to-infrastructure) shares traffic signal timing and road conditions from infrastructure such as traffic lights.

●V2V (vehicle-to-vehicle) sends information about nearby cars, such as their speed and position.

Most of today’s systems focus on this local V2X—information from nearby sources. But this only goes so far. In real-world driving situations, more detailed context is needed. Imagine a fuel cell vehicle driving toward a hill and knowing it can recharge its battery later while going downhill. With this type of foresight, the vehicle can use more energy earlier in the drive. This kind of smart, long-term planning isn’t possible with local V2X alone.

In theory, if a car knew everything about the road ahead—including traffic, slopes, and timing—it could make perfect decisions. However, this far this has been considered impossible. Predicting all of these details and calculating the best choices in real time has simply been too difficult.

A team at the KAIST CCS Graduate School of Mobility, led by Professors Kyunghwan Choi and Dongsuk Kum, recently broke through this barrier. They developed the first system that uses route-wide V2X—broad information covering the entire planned route. Instead of requiring every tiny detail, their system works with representative data for each road segment, such as how much energy is typically used and how long it takes to drive any given part.

Even more impressively, they found a way to turn this massive calculation into a simpler problem that a vehicle can solve quickly. Owing to this innovation, a vehicle can now make real-time decisions that nearly match the best-case scenario in terms of energy use.

Their study, published in IEEE Transactions on Intelligent Transportation Systems, showed that this method achieves the best possible energy efficiency within 0.1%—comparable to a system with full knowledge of the future. This breakthrough brings us closer to a future where vehicles drive more intelligently and efficiently, all thanks to better communication and smarter planning.