Location specific assistance for autonomous vehicle control system
First Claim
1. A system to provide location specific assistance, the system comprising:
- a processor configured to;
determine with a first autonomous vehicle control system how to navigate a first vehicle based on a first artificial intelligence learning model supplied with first sensor data from at least one sensor of the first vehicle;
receive a second artificial intelligence learning model in response to a failure of the first autonomous vehicle control system to determine how to navigate the first vehicle past a condition present at a geographic location based on the first artificial intelligence learning model, the second artificial intelligence learning model is usable for a predetermined timeframe, the second artificial intelligence learning model trained with second sensor data collected when the condition was present at the geographic location and a course past the condition was determined by a second autonomous vehicle control system and/or followed by a person-driven vehicle;
find the course based on an application of the second artificial intelligence learning model within the predetermined timeframe to third sensor data sensed in an environment of the first vehicle at the geographic location; and
cause the first vehicle to follow the course.
5 Assignments
0 Petitions
Accused Products
Abstract
Systems and methods to provide location specific assistance are presented. A condition at a geographic location may be identified that is sensed in an environment of a vehicle and that a first autonomous vehicle control system is unable to, without location specific assistance, perceive, interpret and/or react to if navigating an area. A course at the geographic location that was previously determined by a second autonomous vehicle control system and/or followed by a person-driven vehicle when the condition was present at the geographic location may be found. The vehicle may be caused to follow the course previously determined by a second autonomous vehicle control system and/or followed by the person-driven vehicle.
55 Citations
14 Claims
-
1. A system to provide location specific assistance, the system comprising:
a processor configured to; determine with a first autonomous vehicle control system how to navigate a first vehicle based on a first artificial intelligence learning model supplied with first sensor data from at least one sensor of the first vehicle; receive a second artificial intelligence learning model in response to a failure of the first autonomous vehicle control system to determine how to navigate the first vehicle past a condition present at a geographic location based on the first artificial intelligence learning model, the second artificial intelligence learning model is usable for a predetermined timeframe, the second artificial intelligence learning model trained with second sensor data collected when the condition was present at the geographic location and a course past the condition was determined by a second autonomous vehicle control system and/or followed by a person-driven vehicle; find the course based on an application of the second artificial intelligence learning model within the predetermined timeframe to third sensor data sensed in an environment of the first vehicle at the geographic location; and cause the first vehicle to follow the course. - View Dependent Claims (2, 3)
-
4. An autonomous vehicle, the autonomous vehicle comprising:
a processor configured to; determine that, due to a condition at a geographic location sensed in an environment of the autonomous vehicle, a first autonomous vehicle control system, based on a first artificial intelligence learning model, is unable to cause the autonomous vehicle to navigate past the condition; receive a second artificial intelligence learning model in response to a determination that the first autonomous vehicle control system is unable to cause the autonomous vehicle to navigate past the condition at the geographic location, the second artificial intelligence learning model only usable for a predetermined timeframe, the second artificial intelligence learning model specifically trained with information collected when the condition was present at the geographic location and a course past the condition was determined by a second autonomous vehicle control system and/or followed by a person-driven vehicle; find the course at the geographic location based on the second artificial learning model; and cause the autonomous vehicle to follow the course previously determined by the second autonomous vehicle control system and/or followed by the person-driven vehicle. - View Dependent Claims (5, 6, 7, 8, 9, 10)
-
11. A method comprising:
determining a condition at a geographic location that is sensed in an environment of a vehicle, the condition preventing a first autonomous vehicle control system, which is based on a first artificial learning model, from navigating an area that includes the geographic location, and/or the condition causing a confidence level of the first autonomous vehicle control system in navigating the area to fall below a threshold level; receiving a second artificial intelligence learning model usable for a predetermined timeframe, the second artificial intelligence learning model trained with sensor data collected when the condition was present at the geographic location and a course past the condition was determined by a second autonomous vehicle control system and/or followed by a person-driven vehicle; finding the course at the geographic location based on application of the second artificial intelligence learning model; and causing the vehicle to follow the course at the geographic location past the condition. - View Dependent Claims (12, 13)
-
14. A non-transitory computer readable storage medium comprising computer executable instructions executable by at least one processor to:
-
determine a condition at a geographic location that is sensed in an environment of a vehicle, the condition preventing a first autonomous vehicle control system, which is based on a first artificial learning model, from navigating an area that includes the geographic location, and/or the condition causing a confidence level of the first autonomous vehicle control system in navigating the area to fall below a threshold level; receive a second artificial intelligence learning model usable for a predetermined timeframe, the second artificial intelligence learning model trained with sensor data collected when the condition was present at the geographic location and a course past the condition was determined by a second autonomous vehicle control system and/or followed by a person-driven vehicle; find the course at the geographic location based on application of the second artificial intelligence learning model; and cause the vehicle to follow the course past the condition.
-
Specification