Determining autonomous vehicle routes

1. An autonomous vehicle comprising:

one or more sensors for detecting an object in an environment surrounding the autonomous vehicle; and

a vehicle computing system comprising one or more processors, wherein the vehicle computing system is programmed or configured to:

receive canonical route data associated with a canonical route, the canonical route comprising a plurality of interconnected roadways in a geographic location, the plurality of interconnected roadways satisfying at least one condition based on trip data associated with a plurality of trips executed by a plurality of autonomous vehicles, a first trip of the plurality of trips describing a traversal of a first portion of the plurality of roadways in a geographic location by a first autonomous vehicle of the plurality of autonomous vehicles, and a second trip of the plurality of trips describing a traversal of a second portion of the plurality of roadways in the geographic location by a second autonomous vehicle of the plurality of autonomous vehicles, the at least one condition being based at least in part on a proportion of trips from the plurality of trips that are covered by the canonical route; and

control travel of the autonomous vehicle based on sensor data from the one or more sensors and the canonical route data associated with the canonical route.

2. The autonomous vehicle of claim 1 , wherein the trip data comprises at least one of the following:

a number of interventions associated with the plurality of trips by the plurality of autonomous vehicles;

a number of hazards associated with the plurality of trips; or

any combination thereof.

3. The autonomous vehicle of claim 1 , wherein the vehicle computing system is further programmed or configured to:

receive the trip data associated with the plurality of trips by the plurality of autonomous vehicles;

generate the canonical route based on the trip data; and

provide the canonical route data associated with the canonical route.

4. The autonomous vehicle of claim 1 , wherein the vehicle computing system is further programmed or configured to:

receive map data associated with a map of the geographic location; and

determine the plurality of roadways in the geographic location based on the map data and autonomy criteria, wherein the autonomy criteria is associated with an indication of whether the autonomous vehicle can travel on the plurality of roadways.

5. The autonomous vehicle of claim 1 , wherein the vehicle computing system is further programmed or configured to:

determine a subset of roadways of the plurality of roadways associated with trip data that satisfies a threshold value, wherein each roadway of the subset of roadways must be traversed in a route;

determine a shortest path between each pair of roadways of the subset of roadways;

determine a shortest route in the subset of roadways that includes each roadway of the subset of roadways;

determine a shortest route in the plurality of roadways based on an order of the subset of roadways in the shortest route of the subset of roadways and the shortest path between each pair of roadways in the subset of roadways; and

wherein the canonical route comprises roadways in the shortest route in the plurality of roadways.

6. The autonomous vehicle of claim 1 , wherein the vehicle computing system is further to:

determine a trimmed set of roadways of the plurality of roadways associated with trip data that satisfies a threshold value, wherein each roadway of the trimmed set of roadways must not be traversed in a route;

determine a subset of roadways based on the trimmed set of roadways, wherein the subset of roadways comprises a largest strongly connected set of roadways remaining after removing the trimmed set of roadways from the plurality of roadways;

determine a shortest path between pairs of roadways in the subset of roadways;

determine, for each of a plurality of random subsets of roadways in the subset of roadways, a shortest route that includes each roadway of that random subset of roadways;

apply an objective function derived based on the trip data to each of the shortest routes determined for each random subset of the plurality of random subsets of roadways to determine an objective function value for that shortest route;

determine a preferred subset of roadways based on the objective function values of the plurality of shortest routes;

determine a shortest route in the plurality of roadways based on an order of the preferred subset of roadways and the shortest path between each pair of roadways in the subset of roadways; and

wherein the canonical route comprises roadways in the shortest route in the plurality of roadways.

7. A method of controlling autonomous vehicle, comprising:

receiving, by a vehicle computing system of the autonomous vehicle, canonical route data associated with a canonical route, the canonical route comprising a plurality of interconnected roadways in a geographic location, the plurality of interconnected roadways satisfying at least one condition derived based on trip data associated with a plurality of trips executed by a plurality of autonomous vehicles, a first trip of the plurality of trips describing a traversal of a first portion of the plurality of roadways in a geographic location by a first autonomous vehicle of the plurality of autonomous vehicles, and a second trip of the plurality of trips describing a traversal of a second portion of the plurality of roadways in the geographic location by a second autonomous vehicle of the plurality of autonomous vehicles, the at least one condition being based at least in part on a proportion of trips from the plurality of trips that are covered by the canonical route; and

controlling, by the vehicle computing system, travel of the autonomous vehicle, the controlling being based on sensor data from one or more sensors of the autonomous vehicle and the canonical route data associated with the canonical route.

8. The method of claim 7 , wherein the trip data comprises at least one of the following:

a number of interventions associated with the plurality of trips by the plurality of autonomous vehicles;

a number of hazards associated with the plurality of trips; or

any combination thereof.

9. The method of claim 7 , further comprising:

receiving the trip data associated with the plurality of trips by the plurality of autonomous vehicles;

generating the canonical route based on the trip data; and

providing the canonical route data associated with the canonical route.

10. The method of claim 7 , further comprising:

receiving map data associated with a map of the geographic location; and

determining the plurality of roadways in the geographic location based on the map data and autonomy criteria, wherein the autonomy criteria is associated with an indication of whether the autonomous vehicle can travel on the plurality of roadways.

11. The method of claim 7 , further comprising:

determining a subset of roadways of the plurality of roadways associated with trip data that satisfies a threshold value, wherein each roadway of the subset of roadways must be traversed in a route;

determining a shortest path between each pair of roadways of the subset of roadways;

determining a shortest route in the subset of roadways that includes each roadway of the subset of roadways;

determining a shortest route in the plurality of roadways based on an order of the subset of roadways in the shortest route of the subset of roadways and the shortest path between each pair of roadways in the subset of roadways; and

wherein the canonical route comprises roadways in the shortest route in the plurality of roadways.

12. The method of claim 7 , further comprising:

determining a trimmed set of roadways of the plurality of roadways associated with trip data that satisfies a threshold value, wherein each roadway of the trimmed set of roadways must not be traversed in a route;

determining a subset of roadways based on the trimmed set of roadways, wherein the subset of roadways comprises a largest strongly connected set of roadways remaining after removing the trimmed set of roadways from the plurality of roadways;

determining a shortest path between pairs of roadways in the subset of roadways;

determining, for each of a plurality of random subsets of roadways in the subset of roadways, a shortest route that includes each roadway of that random subset of roadways;

applying an objective function derived based on the trip data to each of the shortest routes determined for each random subset of the plurality of random subsets of roadways to determine an objective function value for that shortest route;

determining a preferred subset of roadways based on the objective function values of the plurality of shortest routes;

determining a shortest route in the plurality of roadways based on an order of the preferred subset of roadways and the shortest path between each pair of roadways in the subset of roadways; and

wherein the canonical route comprises roadways in the shortest route in the plurality of roadways.

13. A non-transitory computer-readable medium comprising instructions thereon that, when executed by at least one processor, cause the at least one processor to perform operations comprising:

receiving canonical route data associated with a canonical route, the canonical route comprising a plurality of interconnected roadways in a geographic location, the plurality of interconnected roadways satisfying at least one condition based on trip data associated with a plurality of trips executed by a plurality of autonomous vehicles, a first trip of the plurality of trips describing a traversal of a first portion of the plurality of roadways in a geographic location by a first autonomous vehicle of the plurality of autonomous vehicles, and a second trip of the plurality of trips describing a traversal of a second portion of the plurality of roadways in the geographic location by a second autonomous vehicle of the plurality of autonomous vehicles, the at least one condition being based at least in part on a proportion of trips from the plurality of trips that are covered by the canonical route; and

controlling travel of the autonomous vehicle, the controlling being based on sensor data from one or more sensors of the autonomous vehicle and the canonical route data associated with the canonical route.

14. The medium of claim 13 , wherein the trip data comprises at least one of the following:

a number of interventions associated with the plurality of trips by the plurality of autonomous vehicles;

a number of hazards associated with the plurality of trips by the plurality of autonomous vehicles; or

any combination thereof.

15. The medium of claim 13 , the operations further comprising:

receiving the trip data associated with the plurality of trips by the plurality of autonomous vehicles;

generating the canonical route based on the trip data; and

providing the canonical route data associated with the canonical route.

16. The medium of claim 13 , the operations further comprising:

receiving map data associated with a map of the geographic location; and

determining the plurality of roadways in the geographic location based on the map data and autonomy criteria, wherein the autonomy criteria is associated with an indication of whether the autonomous vehicle can travel on the plurality of roadways.

17. The medium of claim 13 , the operations further comprising:

determining a subset of roadways of the plurality of roadways associated with trip data that satisfies a threshold value, wherein each roadway of the subset of roadways must be traversed in a route;

determining a shortest path between each pair of roadways of the subset of roadways;

determining a shortest route in the subset of roadways that includes each roadway of the subset of roadways;

determining a shortest route in the plurality of roadways based on an order of the subset of roadways in the shortest route of the subset of roadways and the shortest path between each pair of roadways in the subset of roadways; and

wherein the canonical route comprises roadways in the shortest route in the plurality of roadways.

18. The medium of claim 13 , the operations further comprising:

determining a trimmed set of roadways of the plurality of roadways associated with trip data that satisfies a threshold value, wherein each roadway of the trimmed set of roadways must not be traversed in a route;

determining a subset of roadways based on the trimmed set of roadways, wherein the subset of roadways comprises a largest strongly connected set of roadways remaining after removing the trimmed set of roadways from the plurality of roadways;

determining a shortest path between pairs of roadways in the subset of roadways;

determining, for each of a plurality of random subsets of roadways in the subset of roadways, a shortest route that includes each roadway of that random subset of roadways;

applying an objective function derived based on the trip data to each of the shortest routes determined for each random subset of the plurality of random subsets of roadways to determine an objective function value for that shortest route;

determining a preferred subset of roadways based on the objective function values of the plurality of shortest routes;

determining a shortest route in the plurality of roadways based on an order of the preferred subset of roadways and the shortest path between each pair of roadways in the subset of roadways; and

wherein the canonical route comprises roadways in the shortest route in the plurality of roadways.