The algorithms collectively used in autonomous vehicles lacking in GPS or in addition to it are called SLAM. Simultaneous localization and mapping is defined as follows: In robotic mapping, simultaneous localization and mapping (SLAM) is the computational problem of constructing or updating a map of an unknown environment while simultaneously keeping track of an agent's location within it.
The equations are designed to calculate the responses of the vehicle or object as it moves through an unknown environment. Without understanding of the cirtical point of origin the vehicle has no information with which to base its decision tree upon. The core of autonomy without GPS needs to understand the motion/inertia of itself to coordinate a response to a changing environment. There is no SLAM that localization is not critical to.