Systems and methods for generating basis paths for autonomous vehicle motion control
Systems and methods for basis path generation are provided. In particular, a computing system can obtain a target nominal path. The computing system can determine a current pose for an autonomous vehicle. The computing system can determine, based at least in part on the current pose of the autonomous vehicle and the target nominal path, a lane change region. The computing system can determine one or more merge points on the target nominal path. The computing system can, for each respective merge point in the one or more merge points, generate a candidate basis path from the current pose of the autonomous vehicle to the respective merge point. The computing system can generate a suitability classification for each candidate basis path. The computing system can select one or more candidate basis paths based on the suitability classification for each respective candidate basis path in the plurality of candidate basis paths.
1. A computer-implemented method for controlling an autonomous vehicle, comprising:
determining a target lane for the autonomous vehicle that is different from a current lane in which the autonomous vehicle is operating;
identifying a merge point in the target lane, the merge point describing a location between two vehicles in the target lane;
determining that the autonomous vehicle should enter the target lane at the merge point, wherein determining that the autonomous vehicle should enter the target lane comprises:
processing a plurality of features with a learned lane change model to determine that a lane change criterium is satisfied, wherein the lane change criterium comprises a limit on one or more of the plurality of features, and wherein the lane change criterium is indicative of a cost to enter the target lane at the merge point, and
determining that entering the target lane at the merge point does not violate a plurality of constraints;
determining a motion plan for entering the target lane at the merge point; and
controlling the autonomous vehicle to navigate in accordance with the motion plan.
2. The computer-implemented method of claim 1 , wherein identifying the merge point in the target lane comprises:
identifying a lane change region in the target lane;
identifying one or more potential merge points in the target lane based on the lane change region; and
selecting the merge point from the one or more potential merge points.
3. The computer-implemented method of claim 1 , wherein the plurality of features comprises at least one of a respective acceleration, a respective speed, or a respective turning rate associated with entering the target lane at the merge point.
4. The computer-implemented method of claim 3 , wherein the respective acceleration comprises a maximum acceleration, wherein the respective speed comprises a maximum speed, and wherein the respective turning rate comprises a maximum turning rate.
5. The computer-implemented method of claim 4 , wherein determining whether the lane change criterium is satisfied comprises at least one of determining whether the maximum acceleration exceeds a predetermined acceleration threshold; determining whether the maximum speed exceeds a predetermined speed threshold; or determining whether the maximum turning rate exceeds a predetermined turning rate threshold.
6. The computer-implemented method of claim 1 , wherein the target lane is adjacent to the current lane.
7. The computer-implemented method of claim 1 , wherein the learned lane change model is configured to evaluate the cost associated with one or more vehicle actions associated with entering the target lane at the merge point.
8. The computer-implemented method of claim 1 , wherein the plurality of constraints comprises one or more of a stopping location, a velocity target, or a safe buffer distance.
9. The computer-implemented method of claim 1 , wherein controlling the autonomous vehicle to navigate in accordance with the motion plan comprises converting the motion plan into one or more vehicle controls for implementation by the autonomous vehicle.
10. An autonomous vehicle control system, comprising:
one or more processors; and
one or more non-transitory, computer-readable media storing instructions that cause the one or more processors to perform operations comprising:
determining a target lane for an autonomous vehicle that is different from a current lane in which the autonomous vehicle is operating;
identifying a merge point in the target lane, the merge point describing a location between two vehicles in the target lane;
determining that the autonomous vehicle should enter the target lane at the merge point, wherein determining that the autonomous vehicle should enter the target lane comprises:
processing a plurality of features with a learned lane change model to determine that a lane change criterium is satisfied, wherein the lane change criterium comprises a limit on one or more of the plurality of features, and wherein the lane change criterium is indicative of a cost to enter the target lane at the merge point, and
determining that entering the target lane at the merge point does not violate a plurality of constraints;
determining a motion plan for entering the target lane at the merge point; and
controlling the autonomous vehicle to navigate in accordance with the motion plan.
11. The autonomous vehicle control system of claim 10 , wherein identifying the merge point in the target lane comprises:
identifying a lane change region in the target lane;
identifying one or more potential merge points in the target lane based on the lane change region; and
selecting the merge point from the one or more potential merge points.
12. The autonomous vehicle control system of claim 10 , wherein the plurality of features comprises at least one of a respective acceleration, a respective speed, or a respective turning rate associated with entering the target lane at the merge point.
13. The autonomous vehicle control system of claim 12 , wherein the respective acceleration comprises a maximum acceleration, wherein the respective speed comprises a maximum speed, and wherein the respective turning rate comprises a maximum turning rate.
14. The autonomous vehicle control system of claim 13 , wherein determining whether the lane change criterium is satisfied comprises at least one of determining whether the maximum acceleration exceeds a predetermined acceleration threshold; determining whether the maximum speed exceeds a predetermined speed threshold; or determining whether the maximum turning rate exceeds a predetermined turning rate threshold.
15. The autonomous vehicle control system of claim 10 , wherein the learned lane change model is configured to evaluate the cost associated with one or more vehicle actions associated with entering the target lane at the merge point.
16. The autonomous vehicle control system of claim 10 , wherein the plurality of constraints comprises one or more of a stopping location, a velocity target, or a safe buffer distance.
17. The autonomous vehicle control system of claim 10 , wherein controlling the autonomous vehicle to navigate in accordance with the motion plan comprises converting the motion plan into one or more vehicle controls for implementation by the autonomous vehicle.
18. One or more non-transitory, computer-readable media storing instructions for implementation to cause one or more processors to perform operations comprising:
determining a target lane for an autonomous vehicle that is different from a current lane in which the autonomous vehicle is operating;
identifying a merge point in the target lane, the merge point describing a location between two vehicles in the target lane;
determining that the autonomous vehicle should enter the target lane at the merge point, wherein determining that the autonomous vehicle should enter the target lane comprises:
processing a plurality of features with a learned lane change model to determine that a lane change criterium is satisfied, wherein the lane change criterium comprises a limit on one or more of the plurality of features, and wherein the lane change criterium is indicative of a cost to enter the target lane at the merge point, and
determining that entering the target lane at the merge point does not violate a plurality of constraints;
determining a motion plan for entering the target lane at the merge point; and
controlling the autonomous vehicle to navigate in accordance with the motion plan.