Systems and methods for biasing a trajectory of an autonomous vehicle while moving in a lane
Disclosed herein are systems, methods, and computer program products for biasing a trajectory of a vehicle. The methods comprise: identifying a lane biasing interval of a lane in which the vehicle is to travel based on a lane width; obtaining a first reference path of travel for the vehicle that passes through the lane biasing interval; generating a second reference path of travel for the vehicle using a plurality of lateral offsets and the first reference path of travel for the vehicle; and adjusting the trajectory of the vehicle using the second reference path of travel. The lateral offsets are based on a lateral offset from each index point of the first reference path that resides in the lane biasing interval.
1. A method for biasing a trajectory of a vehicle, comprising:
identifying a lane biasing interval of a lane in which the vehicle is to travel based on a lane width;
obtaining a first reference path of travel for the vehicle that passes through the lane biasing interval;
obtaining a plurality of lateral offsets by:
computing a plurality of desired lateral offsets, each said desired lateral offset corresponding to a distance between a respective index point of the first reference path and a buffer object residing a first buffered distance from a boundary line of the lane,
obtaining a final lateral offset for each index point of the first reference path that resides in the lane biasing interval by setting each desired lateral offset associated with the lane biasing interval to a closest one of the plurality of desired lateral offsets or by filtering the desired lateral off sets associated with the lane biasing interval, and
performing a validity check for the final lateral offset and discarding or modifying the final lateral offset based on a result of the validity check;
generating a second reference path of travel for the vehicle using the plurality of lateral offsets and the first reference path of travel for the vehicle;
adjusting the trajectory of the vehicle using the second reference path of travel.
2. The method according to claim 1 , further comprising causing the vehicle to follow the adjusted trajectory.
3. The method according to claim 1 , further comprising modifying the lane biasing interval based on road geometry.
4. The method according to claim 1 , wherein the lane biasing interval is identified when the lane is classified as a wide lane.
5. The method according to claim 1 , wherein the lane biasing interval is part of the lane prior to a turn, a lane split, or a lane width variation.
6. The method according to claim 1 , wherein the lane biasing interval has: a length set to a calibration parameter; an end located a calibratable distance from a start of a turn defined by corner constraints, a start of a width change in the lane, or a start of a lane split; a pre-lane biasing interval in which a lateral offset is set to zero or a small value; and a post-lane biasing interval in which a lateral offset weight is set to zero or a small value.
7. The method according to claim 1 , wherein the first buffered distance is set to a first value when a shoulder or bike lane exists prior to a turn on which the vehicle can drive, and a second different value when no shoulders or bike lanes exit prior to the turn.
8. The method according to claim 1 , further comprising adjusting the desired lateral offset such that the buffered object is a second buffered distance from an obstacle.
9. The method according to claim 8 , wherein the obstacle comprises an obstacle other than an obstacle that is on a side of the lane which is opposite to a side of the lane comprising a turn and an obstacle outside of a temporal horizon.
10. The method according to claim 8 , wherein the adjusting is discontinued when the obstacle is a moving object traveling faster than a threshold speed on a given side of the lane biasing interval.
11. The method according to claim 1 , wherein the lateral offset is computed by:
computing a distance between index points of a portion of the reference path located within a post-lane biasing interval of the lane and points on a boundary line of the lane; and
identifying lateral offset values that ensure the vehicle will remain at the computed distance from the boundary line of the lane throughout the lane biasing interval.
12. The method according to claim 11 , further comprising adjusting the lateral offset values to ensure that the vehicle will remain a buffered distance from any obstacles and that there is a smooth transition between a portion of the reference path with adjusted lateral offset values and a portion of the reference path with unadjusted lateral offset values.
13. The method according to claim 1 , further comprising computing each of the plurality of lateral offsets from each index point of the reference path that resides in the lane biasing interval, and assigning weights to the lateral offsets, wherein the second reference path is generated additionally using the weights.
14. The method according to claim 13 , wherein the weights are assigned to the lateral offsets by:
assigning a first weight value to lateral offset values associated with index points of a portion of the reference path which resides in a pre-lane biasing interval;
assigning a second weight value to lateral offset values associated with index points of a portion of the reference path which resides in the lane biasing interval; and
assigning a third weight value to lateral offset values associated with index points of a portion of the reference path which resides in a post-lane biasing interval.
15. The method according to claim 13 , wherein the weights are assigned to the lateral offsets by:
identifying each sub-interval in the lane biasing interval where the lateral offset is constant;
comparing a length of each identified sub-interval to a threshold value;
assigning a first weight value to the lateral offset when a length of the identified sub-interval exceeds the threshold value; and
assigning a second weight value to the lateral offset when the length of the identified sub-interval does not exceed the threshold value.
16. The method according to claim 13 , wherein the weights are assigned to the lateral offsets by: identifying each sub-interval of the lane biasing interval comprising ramped lateral offsets; and assigning a weight value to the ramped lateral offsets.
17. A system, comprising:
a processor;
a non-transitory computer-readable storage medium comprising programming instructions that are configured to cause the processor to implement a method for biasing a trajectory of a vehicle, wherein the programming instructions comprise instructions to:
identify a lane biasing interval of a lane in which the vehicle is to travel based on a lane width;
obtain a first reference path of travel for the vehicle that passes through the lane biasing interval;
compute each lateral offset among a plurality of lateral offsets from each index point of the first reference path that resides in the lane biasing interval;
assign weights to the lateral offsets by:
identifying each sub-interval in the lane biasing interval where the lateral offset is constant,
comparing a length of each identified sub-interval to a threshold value,
assigning a first weight value to the lateral offset when a length of the identified sub-interval exceeds the threshold value, and
assigning a second weight value to the lateral offset when the length of the identified sub-interval does not exceed the threshold value;
generate a second reference path of travel for the vehicle using the plurality of lateral offsets and the first reference path of travel for the vehicle, the second reference path is generated using the weight; and
adjust the trajectory of the vehicle using the second reference path of travel.
18. The system according to claim 17 , wherein the programming instructions further comprise instructions to cause the vehicle to follow the trajectory which has been adjusted.
19. A non-transitory computer-readable medium that stores instructions that is configured to, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:
identifying a lane biasing interval of a lane in which a vehicle is to travel based on a lane width;
obtaining a first reference path of travel for the vehicle that passes through the lane biasing interval;
computing each lateral offset of a plurality of lateral offsets from each index point of the first reference path that resides in the lane biasing interval;
assigning weights to the lateral offsets by:
identifying each sub-interval of the lane biasing interval comprising ramped lateral offsets, and
assigning a weight value to the ramped lateral offsets;
generating a second reference path of travel for the vehicle using a plurality of lateral offsets and the first reference path of travel for the vehicle, wherein the second reference path is generated using the weights; and
adjusting a trajectory of the vehicle using the refined reference path of travel.