System and method for autonomous vehicle fleet routing

1. A system providing routing instructions to one or more autonomous vehicles, the system comprising:

a subject autonomous vehicle requiring a route plan, the subject autonomous vehicle providing sensor data from the subject autonomous vehicle;

a fleet of vehicles comprising a plurality of autonomous vehicles different from the autonomous vehicle, wherein each of the plurality of autonomous vehicles provide sensor data from the plurality of autonomous vehicles;

a routing server, including a computer processor, that:

collects sensor data from the subject autonomous vehicle and sensor data from the plurality of autonomous vehicles;

identifies source and destination data of the subject autonomous vehicle;

generates the route plan for the autonomous vehicle based on sensor data from the subject autonomous vehicle, sensor data from the plurality of autonomous vehicles, and source and destination data of the subject autonomous vehicle, and

controls movement of the autonomous vehicle in accordance with the route plan that is generated;

wherein the routing server generates the route plan for the autonomous vehicle by:

applying a probabilistic algorithm to one or more route segments of one or more proposed route plans, by:

identifying proposed maneuvers in the one or more route segments;

assigning to each proposed maneuver a probability of success and a probability of failure of accomplishing the proposed maneuver by the autonomous vehicle;

associating a path of failure with the probability of failure and path of success with the probability of success;

identifying (a) a probabilistic cost of the path of failure and (b) a probabilistic cost of the path of success; and

identifying a probabilistic maneuver cost based on the sum of (a) the probabilistic cost of the path of failure and added to (b) the probabilistic cost of the path of success; and

identifying one or more route segments for the route plan of the autonomous vehicle based on the probabilistic maneuver cost based on the application of the probabilistic algorithm.

2. The system of claim 1 , further comprising:

a remote assistance interface that is configured to remotely assist the one or more autonomous vehicles, wherein the remote assistance interface is different from the routing server, and

wherein the routing server collects remote assistance data from the remote assistance interface and generates the route plan for the autonomous vehicle further based on the collected remote assistance data.

3. The system of claim 1 , further comprising:

a plurality of user interfaces, wherein the routing server collects vehicle demand data from the plurality of user interfaces or a ride sharing platform, and

wherein the routing server generates the route plan for the autonomous vehicle further based on the collected vehicle demand data.

4. The system of claim 1 , wherein:

the path of success comprises a first path to reach the destination given that the proposed maneuver has been performed correctly; and

the path of failure comprises a second path to reach the destination, given that the proposed maneuver has not been performed correctly.

5. The system of claim 4 , wherein:

the path of success comprises a cost-optimized first path to reach the destination given that the proposed maneuver has been performed correctly; and

the path of failure comprises a cost-optimized second path to reach the destination, given that the proposed maneuver has not been performed correctly, such that the cost-optimized second path has a higher probabilistic cost than the cost-optimized first path but a lower score probabilistic cost as compared with all other paths in which the maneuver has not been performed correctly, wherein the probabilistic cost is associated with travelling down a road section.

6. The system of claim 5 , wherein the probabilistic maneuver cost is equal to the summation of (a) a probability of success for the maneuver multiplied by a success cost for the maneuver, for the cost-optimized first path added to (b) a probability of failure for the maneuver multiplied by a failure cost for the maneuver, for the cost-optimized second path.

7. A method providing routing instructions to an autonomous vehicle, the method comprising:

identifying a destination of the autonomous vehicle;

identifying a starting position or an initial location of the autonomous vehicle;

receiving autonomous vehicle sensor data;

receiving one or more routing goals for a routing plan for the autonomous vehicle;

generating one or more route modification parameters; and

generating a route plan for the autonomous vehicle based on the destination, the starting position or the initial location, the one or more route modification parameters and the one or more routing goals, via a computer processor; and

controlling movement of the autonomous vehicle in accordance with the route plan that is generated, via the computer processor;

wherein generating the route plan for the autonomous vehicle includes, via the computer processor:

applying a probabilistic algorithm to one or more route segments of one or more proposed route plans, and

identifying one or more route segments for the route plan of the autonomous vehicle based on the application of the probabilistic algorithm;

wherein applying the probabilistic algorithm includes:

identifying proposed maneuvers in the one or more route segments;

assigning to each proposed maneuver a probability of success and a probability of failure of accomplishing the proposed maneuver by the autonomous vehicle;

associating a path of failure with the probability of failure and path of success with the probability of success; and

identifying (a) a probabilistic cost of the path of failure and (b) a probabilistic cost of the path of success;

wherein applying the probabilistic algorithm further includes identifying a probabilistic maneuver cost based on the sum of (a) the probabilistic cost of the path of failure and added to (b) the probabilistic cost of the path of success, and

wherein generating the route plan for the autonomous vehicle is further based on the probabilistic maneuver cost.

8. The method of claim 7 , further comprising:

receiving vehicle demand data, the vehicle demand data indicating one or more desires or estimated desire for the autonomous vehicle to be present in a specific region or a specific location.

9. The method of claim 7 , wherein the routing goals establish a set of routing parameters that identify metrics from an analysis of the sensor data that are considered in determining routes or route modifications for the routing plan of the autonomous vehicles.

10. The method of claim 7 , further comprising:

receiving blacklist data from a remote assistance interface, wherein blacklist data identifies one or more conditions or events rendering one or more lanes or one or more road segments not available to be used in generating the route plan for the autonomous vehicle.

11. The method of claim 7 , further comprising:

receiving avoidance scenario data from one or more autonomous vehicles in a fleet of autonomous vehicles, wherein the fleet includes a plurality of autonomous vehicles other than the autonomous vehicle for which the route plan is generated,

generating blacklist data from the avoidance scenario data, wherein blacklist data identifies one or more conditions or events rendering one or more lanes or one or more road segments not available to be used in generating the route plan for the autonomous vehicle;

wherein generating the route plan for the autonomous vehicle is further based on the identified blacklist data.

12. The method of claim 7 , wherein receiving autonomous vehicle sensor data includes:

receiving sensor data from the autonomous vehicle,

receiving sensor data from a fleet of autonomous vehicles, wherein the fleet includes a plurality of autonomous vehicles other than the autonomous vehicle for which the route plan is generated, and

wherein generating the route plan for the autonomous vehicle is further based on both sensor data from the autonomous vehicle and the fleet of autonomous vehicles.

13. The method of claim 7 , wherein the destination of the autonomous vehicle and the starting position or the initial location of the autonomous vehicle for the route plan are identified by disparate entities,

generating autonomous vehicle control data based on the generated route plan for the autonomous vehicle, and

controlling autonomous vehicles even when the destination and the initial starting position or initial location of the autonomous vehicle for the route plan are identified by disparate entities.

14. The method of claim 7 , wherein the autonomous vehicle sensor data comprises data relating to a state of the autonomous vehicle and a state of circumstances around the autonomous vehicle.

15. The method of claim 7 , further comprising:

receiving construction data and traffic data,

wherein generating the route plan for the autonomous vehicle is further based on the construction data and the traffic data.

16. The method of claim 7 , wherein:

the path of success comprises a first path to reach the destination given that the proposed maneuver has been performed correctly; and

the path of failure comprises a second path to reach the destination, given that the proposed maneuver has not been performed correctly.

17. The method of claim 16 , wherein:

the path of success comprises a cost-optimized first path to reach the destination given that the proposed maneuver has been performed correctly; and

the path of failure comprises a cost-optimized second path to reach the destination, given that the proposed maneuver has not been performed correctly, such that the cost-optimized second path has a higher probabilistic cost than the cost-optimized first path but a lower probabilistic cost as compared with all other paths in which the maneuver has not been performed correctly, wherein the probabilistic cost is associated with travelling down a road section.

18. The method of claim 17 , wherein the probabilistic maneuver cost is equal to the summation of (a) a probability of success for the maneuver multiplied by a success cost for the maneuver, for the cost-optimized first path added to (b) a probability of failure for the maneuver multiplied by a failure cost for the maneuver, for the cost-optimized second path.