While they can tell us the distance between two points, roughly how long an errand may take and possibly even alert us to traffic problems ahead, they lack real computing firepower. At least, when compared to travel planning systems in, say, a Nasa spacecraft.
But that may change, and soon.
Brian C Williams, a professor in the Computer Science and Artificial Intelligence Lab at the Massachusetts Institute of Technology (MIT), is working with his research team to create software algorithms that would help drivers arrive at their destination exactly – and not a minute later – than they want to.
Williams said that while the world’s go-to online route planner, Google Maps, can recommend how to travel the roughly 200 miles (161km) from New York to Boston, the trip usually takes much longer than the original estimate. He says Google Maps "doesn't really know about uncertainty. It doesn't really know about my deadline. It just helps me know where to go along the way."
Contrast this with software his lab has developed, which can recommend which filling stations to patronise and which route to take, and even understands that Williams needs to arrive at his destination at a specific time – and oh, that he wants to eat an hour-long dinner en route.
The lab’s planning algorithms would give motorists an initial travel plan, which would adapt to externalities along the way, communicating when a parameter has changed and making recommendations to ensure a timely arrival, such as suggesting a restaurant closer to the traveller’s destination.
Williams says his artificial intelligence system, which is called Enterprise ("after a Star Trek theme", he quips) can be thought of as a smarter version of Apple’s Siri or Microsoft's Cortana voice-activated user interfaces. "Our AI is really good at coming up with plans for you, explaining why they work, monitoring them, and then we're coupling that together with speech and natural language recognition," Williams says.
We've been here before
Travel-planning algorithms have become essential software for drivers of electric cars, to help reduce range anxiety. But their roots lie in Williams's work at Nasa in the late 1990s, when he helped build autonomous systems for deep-space probes.
And while such algorithms could be used for decision-making in autonomous cars, Williams sees them as functioning as a supplementary guide, not as tools that would obviate the driver from driving. "You're still doing the driving, but it's helping to be an advisor," he says.
A proof of concept is already up and running in Williams’s lab, and could ultimately be built as a smartphone app or paired with a virtual assistant system. "Right now it's a tool that you could use without hooking into your car," Williams says.
More than just trips to the country
Williams stresses that the technology’s application is not limited to ground transport. Deep-sea robot explorers are already testing it, and Boeing has assessed it for personal autonomous air vehicles (yes, you read that correctly). Williams said the system can also be applied to manufacturing, humanoid robots, drones and – taking it back to its interstellar roots – further space exploration.
Motorists may not be waiting long for this tech, either; Williams predicts that such algorithms may be running on our smartphones within the next few years. "I think it's more of an issue of the right apps, or the right product at this point," he says. Packaging, in other words.
Nasa-calibre travel guides in our pockets? Check. Now, if only MIT could make smartphone batteries last longer (though there is a lab for that, too).
If you would like to comment on this or anything else you have seen on BBC Autos, head over to our Facebook page or message us on Twitter.