Probabilistic navigation system and method

A computing system comprising:

• one or more processors;
• and
• memory storing instructions that, when executed by the one or more processors, cause the computing system to perform operations comprising:
• obtaining a map of an environment comprising a plurality of objects each associated with a probability model, each of the probability models at least describing one or more probability fields of a vehicle colliding with at least one of the plurality of objects;
• obtaining, using the vehicle, one or more global factors from one or more data stores;
• obtaining, using one or more sensors, one or more local factors;
• updating the probability fields of the probability models based on the one or more global factors and the one or more local factors, wherein the global factors apply to all of the objects, and the local factors apply to a portion of the objects, wherein at least a first global factor has a first degree of influence for a particular object of the plurality of objects, and wherein the first global factor has a second degree of influence for another object of the plurality of objects, the updating including adjusting a rate of decay for at least one of the probability fields of a corresponding probability model of the plurality of probability models;
• determining a navigation for the vehicle through at least a part of the environment based at least on a total collision probability of the vehicle with at least one of the plurality of objects along the navigation according to the updated probability models;
• and
• causing, in response to determining the navigation, the navigation of the vehicle through at least the part of the environment.

A non-transitory computer readable medium comprising instructions that, when executed, cause one or more processors to perform operations comprising:

• obtaining a map of an environment comprising a plurality of objects each associated with a probability model, each of the probability models at least describing one or more probability fields of a vehicle colliding with at least one of the plurality of objects;
• obtaining, using the vehicle, one or more global factors from one or more data stores;
• obtaining, using one or more sensors, one or more local factors;
• updating the probability fields of the probability models based on the one or more global factors and the one or more local factors, wherein the global factors apply to all of the objects, and the local factors apply to a portion of the objects, wherein at least a first global factor has a first degree of influence for a particular object of the plurality of objects, and wherein the first global factor has a second degree of influence for another object of the plurality of objects, the updating including adjusting a rate of decay for at least one of the probability fields of a corresponding probability model of the plurality of probability models;
• determining a navigation for the vehicle through at least a part of the environment based at least on a total collision probability of the vehicle with at least one of the plurality of objects along the navigation according to the updated probability models;
• and
• causing, in response to determining the navigation, the navigation of the vehicle through at least the part of the environment.