Predictive Cruise Control

How sophisticated is cruise control really? How does it adapt to elevation changes? If you're cruising on the interstate and about to go up a long grade, is cruise control really going to do a more efficient job than a good human driver would? It's going to do a good job of keeping the vehicle close to the speed you set it at. But it might use much more fuel to do so than a person who can anticipate the hill and carry that momentum up the hill.

The issue is that cruise control systems only react to conditions that it is already experiencing. Adaptive cruise control is beginning to make inroads on anticipating emergency conditions and taking preventive action. That's for safety. Let's talk about energy efficiency.

Thinking along the lines of moving from cars that are driven to cars that are driverless, let's say you're driving your daily commute. It's the same route day in and day out, and there are some hills both ways. Now let's say this route is over an interstate and you're running predictive cruise control (call it PCC) in the peak efficiency speed range. As you come up to the hill the system speeds the car up by a couple of miles per hour in order to build momentum and help push you up the grade. But as it nears the peak it backs off a hair because on the other side you're going to exceed your set speed anyway as you start going downhill. This way it takes less energy to move the car over that hill.

This is repeated for every grade you encounter on the portion of the drive where you use PCC. It will end up saving a lot of fuel. Not only is that important from an environmental perspective, it's also going to be more important in electric vehicles while battery storage capacity is still far behind gasoline and diesel. There just won't be nearly as much energy on board, and PCC can stretch the car's range a little bit more.

Now, how does the system know how to predict a grade, where it begins to rise and where it peaks? GPS is one possibility. If the car's navigation system knew where it was headed and what the grades were, it could map where to adjust speeds. It could also "learn" routes it regularly travels on and build up a knowledge base. Then there's always DSRC. If it gets implemented and provides local road data to passing vehicles, topographical data could be included.

Vehicles with these capabilities are already being tested, and they're going to be important technologies as we move forward.