New system gives in-city GPS navigation a big boost
Many of us use our vehicle navigation systems on a daily basis, and as self-driving cars come into common use – assuming they do – such systems will become even more important. Unfortunately, however, the GPS technology that’s integral to vehicle navigation can be thwarted by obstacles such as tall buildings. A team of researchers at Spain’s Universidad Carlos III de Madrid (UC3M) are attempting to address that problem, with a system that is said to drastically boost GPS accuracy in city driving.
According to the researchers, commercial GPS systems have a margin of error of about 15 meters (49 feet) when in an open field, where they have a clear line of sight to the satellites that they use to triangulate their location. In an urban setting, however, where signals can get bounced off high-rises and other obstacles, that figure goes up to about 50 meters (164 feet).
When a system can’t “see” the sky at all, such as when the vehicle is going through a tunnel, the GPS stops working altogether. Some systems have a feature that takes over at such points, where the vehicle’s location is estimated on a map based on its last known position and heading, although these are essentially just educated guesses.
The UC3M team took an existing GPS unit, and combined it with a computer and an inertial measurement unit – that unit contains three accelerometers and three gyroscopes, which keep track of the speed and orientation of the vehicle at all times. Custom software on the computer combines data from the GPS and the inertial measurement unit, to consistently and accurately track the location of the vehicle to within one to two meters (3.3 to 6.6 feet) in urban settings.
According to the researchers, the hardware used in the system is inexpensive, and can be installed in any vehicle. The prototype is currently being used in the university’s IVVI (Intelligent Vehicle based on Visual Information), which is also serving as a testbed for technologies such as lane departure warning and pedestrian detection.
The team now hopes to bring the cost of these technologies down further, by taking advantage of the sensors and processing power already present in drivers’ smartphones.
Source: Universidad Carlos III de Madrid