Improved GPS could boost self-driving cars and other tech

We've seen other efforts to improve GPS location to centimeter-level accuracy using what's called "differential GPS," that makes use of ground-based reference points in addition to satellite GPS data. This latest effort from the University of California - Riverside (UCR) seems similar in that it's basically a software-based approach.

What's perhaps most revolutionary about the advance is not just the improved level of accuracy, but just how efficiently centimeter-accurate positioning is established.

"Achieving this level of accuracy with computational loads that are suitable for real-time applications on low-power processors will not only advance the capabilities of highly specialized navigation systems – like those used in driverless cars and precision agriculture – but it will also improve location services accessed through mobile phones and other personal devices, without increasing their cost," said UCR professor Jay Farrell, who led the research.

He claims that until now, achieving such accuracy has been computationally expensive, but the new approach returns a highly accurate position with several orders of magnitude fewer computations. That could make this level of GPS more practical in a number of emerging real world applications.

"To fulfill both the automation and safety needs of driverless cars, some applications need to know not only which lane a car is in, but also where it is in that lane – and need to know it continuously at high rates and high bandwidth for the duration of the trip," said Farrell.

Two members of the team that conducted the research have since moved on to work at Qualcomm and Google. The research was published recently in IEEE'sTransactions on Control Systems Technology.