Lane changes for autonomous vehicles involving traffic stacks at intersection
Aspects of the disclosure provide for controlling an autonomous vehicle. For instance, one or more processors of one or more first systems of the autonomous vehicles may receive a signal indicating a predicted traffic stack for a lane in which the autonomous vehicle is currently traveling. In response to the received signal, costs of edges of a roadgraph between the autonomous vehicle and a location of the predicted traffic stack may be adjusted in order to encourage the autonomous vehicle to change lanes in response to the predicted traffic stack. A route may be generated to a destination based on at least one of the adjusted costs. The route may be provided to one or more second systems of the autonomous vehicle in order to control the autonomous vehicle according to the route.
1. A method for controlling an autonomous vehicle, the method comprising:
receiving, by one or more processors of one or more first systems of the autonomous vehicle, a first signal indicating a first predicted traffic stack for a first lane in which the autonomous vehicle is currently traveling;
receiving, by the one or more processors, a second signal indicating a second predicted traffic stack in a second lane adjacent to the first lane;
in response to the received first signal or the received second signal, adjusting, by the one or more processors, costs of edges of a roadgraph based on which of the first predicted traffic stack or the second predicted traffic stack is a worse traffic stack, wherein the costs include costs of first edges between the autonomous vehicle and a location of the first predicted traffic stack or costs of second edges of the roadgraph between the autonomous vehicle and a location of the second predicted traffic stack;
generating, by the one or more processors, a route to a destination based on the adjusted costs; and
providing, by the one or more processors, the route to one or more second systems of the autonomous vehicle in order to control the autonomous vehicle according to the route.
2. The method of claim 1 , wherein the first or second predicted traffic stack is located at an intersection.
3. The method of claim 1 , wherein the one or more first systems is a planning system of the autonomous vehicle configured to generate trajectories according to the route for the autonomous vehicle to follow.
4. The method of claim 1 , wherein the one or more second systems include a routing system of the autonomous vehicle configured to generate the route.
5. The method of claim 1 , wherein adjusting the costs includes increasing the costs.
6. The method of claim 1 , wherein the first edges correspond to the first lane and the second edges correspond to the second lane.
7. The method of claim 1 , wherein the first and second predicted traffic stacks are based on behavior predictions of other road users and occlusion information of the other road users, and the occlusion information includes information about at least one other road user that was detected by the one or more first systems and subsequently became occluded from the one or more first systems.
8. The method of claim 1 , wherein the second signal further indicates which of the first predicted traffic stack and the second predicted traffic stack is the worse traffic stack, and the adjusting is further based on which of the predicted traffic stack and the second predicted traffic stack is the worse traffic stack.
9. A system for controlling an autonomous vehicle, the system comprising one or more processors configured to:
receive a first signal indicating a first predicted traffic stack for a first lane in which the autonomous vehicle is currently traveling;
in response to the received first signal, adjust first costs of first edges of a roadgraph between the autonomous vehicle and a location of the first predicted traffic stack in order to encourage the autonomous vehicle to change lanes in response to the first predicted traffic stack;
before, during, or after receipt of the first signal, receive a second signal indicating a second predicted traffic stack in a second lane adjacent to the first lane, wherein the second signal further indicates which of the first predicted traffic stack and the second predicted traffic stack is a worse traffic stack;
in response to the received second signal, adjust second costs of second edges of the roadgraph in the second lane;
adjust the first edges and the second edges further based on which of the first predicted traffic stack and the second predicted traffic stack is a worse traffic stack;
generate a route to a destination based on at least one of the adjusted first costs and based on at least one of the adjusted second costs; and
provide the route to one or more second systems of the autonomous vehicle in order to control the autonomous vehicle according to the route.
10. The system of claim 9 , wherein the first or second predicted traffic stack is located at an intersection.
11. The system of claim 9 , wherein the system is a planning system of the autonomous vehicle configured to generate trajectories according to the route for the autonomous vehicle to follow.
12. The system of claim 9 , further comprising the one or more second systems, wherein the one or more second systems include a routing system of the autonomous vehicle configured to generate the route.
13. The system of claim 9 , wherein the one or more processors are further configured to adjust the first or second costs includes increasing the first or second costs.
14. The system of claim 9 , wherein the first edges correspond to the first lane and the second edges correspond to the second lane.
15. The system of claim 9 , wherein the first and second predicted traffic stacks are based on behavior predictions of other road users and occlusion information of the other road users.
16. An autonomous vehicle comprising:
memory storing a roadgraph; and
a planning system comprising one or more processors configured to:
receive a first signal indicating a first predicted traffic stack for a first lane in which the autonomous vehicle is currently traveling;
receive a second signal indicating a second predicted traffic stack in a second lane adjacent to the first lane;
in response to the received first signal or the received second signal, adjust costs of edges of a roadgraph based on which of the first predicted traffic stack or the second predicted traffic stack is a worse traffic stack, wherein the costs include costs of first edges between the autonomous vehicle and a location of the first predicted traffic stack or costs of second edges of the roadgraph between the autonomous vehicle and a location of the second predicted traffic stack;
generate a route to a destination based on at least one of the adjusted costs; and
provide the route to one or more second systems of the autonomous vehicle in order to control the autonomous vehicle according to the route.
17. The autonomous vehicle of claim 16 , further comprising the one or more second systems.
18. The autonomous vehicle of claim 16 , the one or more second systems include a routing system of the autonomous vehicle configured to generate the route.
19. The autonomous vehicle of claim 16 wherein, to adjust the costs, the one or more processors are further configured to:
when the first predicted traffic stack is determined to be the worse traffic stack, increase the costs of the first edges by a value greater than costs of edges in any other lanes; and
when the second predicted traffic stack is determined to be the worse traffic stack, increase the costs of the second edges by a value greater than costs of edges in any other lanes.
20. The autonomous vehicle of claim 16 , wherein the first and second predicted traffic stacks are based on behavior predictions of other road users and occlusion information of the other road users, and the occlusion information includes information about at least one other road user that was detected and subsequently became occluded.