This technology provides the ability for a self-driving automobile to stay in its lane even in cases where traditional optical approaches don't work.
Traditional sensing approaches for map-based localization are not robust to common real-world conditions especially when visible features are obscured by leaves, snow, smoke or heavy rain. Therefore, there is a need to provide a robust vehicle localization method by augmenting current sensors.
The invention features a system for determining location data for a vehicle, using surface penetrating radar (SPR). The SPR images of a subsurface region are acquired along a vehicle track. Then, the images are compared to previous SPR images from another subsurface region that at least partially overlaps. Finally, the location data for the vehicle is determined based on the comparison.
- Can operate in poor conditions, unlike current localization methods
- The technology is not limited to ground based applications