Kinematic model for autonomous truck routing
The technology relates to route planning and performing driving operations in autonomous vehicles, such as cargo trucks, articulating buses, as well as other vehicles. A detailed kinematic model of the vehicle in evaluated in conjunction with roadgraph and other information to determine whether a route or driving operation is feasible for the vehicle. This can include evaluating a hierarchical set of driving rules and whether current driving conditions impact any of the rules. Driving trajectories and cost can be evaluated when pre-planning a route for the vehicle to follow. This can include determining an ideal trajectory for the vehicle to take a particular driving action. Pre-planned routes may be shared with a fleet of vehicles, and can be modified based on information obtained by different vehicles of the fleet.
1. A method of operating a vehicle in an autonomous driving mode, the method comprising:
receiving, by one or more processors of the vehicle, sensor data of the environment around the vehicle from one or more sensors of the vehicle;
determining, by the one or more processors, a plurality of possible paths for a given driving maneuver in order to follow a planned trajectory along a route before determining a kinematic feasibility of each possible path;
determining, by the one or more processors based on the received sensor data and the plurality of possible paths, the kinematic feasibility of each possible path according to a kinematic model of the vehicle;
selecting, by the one or more processors, a given one of the possible paths based on the kinematic feasibility and a hierarchy of stored rules, the hierarchy of stored rules being associated with one or more of lane indicators, physical road components or restrictions of the vehicle; and
causing the vehicle to perform a selected driving operation in the autonomous driving mode to follow the given path in accordance with the kinematic feasibility and the hierarchy of stored rules.
2. The method of claim 1 , wherein the kinematic feasibility is determined according to a swept volume of the vehicle in accordance with each possible path.
3. The method of claim 2 , wherein the swept volume includes one or more articulating sections of the vehicle.
4. The method of claim 1 , wherein the hierarchy of stored rules ranks one or more of different lane line types, curb configurations or road shoulder types.
5. The method of claim 1 , wherein selecting the given path based on the hierarchy of stored rules includes ranking paths based on expected violation of one or more of the rules.
6. The method of claim 1 , wherein determining the kinematic feasibility of each possible path includes evaluating a maneuver cost associated with that path.
7. The method of claim 6 , wherein evaluating the maneuver cost includes evaluating one or more of a number of turns, occlusions, weather limitations, neighborhood type or temporal restrictions.
8. The method of claim 6 , wherein evaluating the maneuver cost includes analyzing a cost for missing a turn, missing an exit, or missing a lane change.
9. The method of claim 1 , wherein selecting the given path is further based on a likelihood of the vehicle becoming stuck for one or more of the possible paths.
10. The method of claim 1 , further comprising pre-planning for when to change lanes in advance of the selected driving operation.
11. A method of determining an overall route for a vehicle to operate in an autonomous driving mode, the method comprising:
obtaining, by one or more processors of a computing system, at least one map comprising roadgraph information between a starting point and a destination of the overall route for the vehicle to operate in the autonomous driving mode;
determining, by the one or more processors, whether the vehicle is restricted from traveling along any segments of the overall route;
determining, by the one or more processors, an initial trajectory along the overall route based on determining whether the vehicle is restricted from traveling along any segments;
determining, by the one or more processors for the initial trajectory, a plurality of possible paths for a given driving maneuver by the vehicle before determining a kinematic feasibility of each possible path;
determining, by the one or more processors based on the plurality of possible paths, the kinematic feasibility of each possible path according to a kinematic model of the vehicle;
selecting, by the one or more processors, a given one of the possible paths based on the kinematic feasibility and a hierarchy of rules, the hierarchy of rules being associated with one or more of lane indicators, physical road components or restrictions of the vehicle;
generating, by the one or more processors, a final trajectory for the vehicle based on the given path; and
transmitting the final trajectory to the vehicle for operating in the autonomous driving mode.
12. The method of claim 11 , wherein determining the kinematic feasibility of each possible path includes evaluating at least one of a temporal restriction or a legal restriction.
13. The method of claim 12 , wherein evaluating the temporal restriction includes estimating a likelihood that the vehicle will be able to perform a driving action prior to occurrence of the temporal restriction.
14. The method of claim 11 , wherein determining the kinematic feasibility of each possible path includes evaluating a maneuver cost associated with that path.
15. The method of claim 14 , wherein the maneuver cost is based on one or more of a number of turns, occlusions, weather limitations, neighborhood type or temporal restrictions.
16. The method of claim 14 , wherein the maneuver cost includes a likelihood of the vehicle becoming stuck.
17. The method of claim 11 , wherein the hierarchy of rules includes a set of rules relating to items or objects along a roadway.
18. The method of claim 17 , wherein the items or objects include one or more of different lane line types, curb configurations or road shoulder types.
19. The method of claim 11 , wherein selecting the given path based on the hierarchy of rules includes ranking paths based on expected violation of one or more rules in the hierarchy.
20. The method of claim 11 , further comprising receiving information about restrictions or maneuver costs from one or more other vehicles that have driven along the overall route.
21. The method of claim 20 , wherein receiving the information includes receiving information about actions either taken or attempted by a given vehicle.
22. The method of claim 20 , wherein receiving the information includes receiving information observed by a given vehicle about an action taken by a different vehicle.
23. The method of claim 11 , wherein determining whether the vehicle is restricted from traveling along any segments includes determining whether the segments are legally restricted to the vehicle.