Vehicle trajectory modification for following
Techniques for determining to modify a trajectory based on an object are discussed herein. A vehicle can determine a drivable area of an environment, capture sensor data representing an object in the environment, and perform a spot check to determine whether or not to modify a trajectory. Such a spot check may include processing to incorporate an actual or predicted extent of the object into the drivable area to modify the drivable area. A distance between a reference trajectory and the object can be determined at discrete points along the reference trajectory, and based on a cost, distance, or intersection associated with the trajectory and the modified area, the vehicle can modify its trajectory. One trajectory modification includes following, which may include varying a longitudinal control of the vehicle, for example, to maintain a relative distance and velocity between the vehicle and the object.
1. A method comprising:
determining a drivable area through an environment for a vehicle, wherein the drivable area represents a region in the environment on which the vehicle is permitted to traverse;
determining a trajectory within the drivable area for the vehicle to follow;
receiving sensor data from a sensor on the vehicle;
detecting, based at least in part on the sensor data, an object in the environment;
determining, based at least in part on the sensor data, object parameters associated with the object, wherein the object parameters comprise one or more of a velocity, a position, a classification, or an extent;
determining, as a modified drivable area, based at least in part on the object parameters, and independent of the trajectory, a portion of the drivable area that excludes a space that the object is predicted to occupy based on the object parameters;
determining, based at least in part on the modified drivable area, that the trajectory is invalid;
determining, based at least in part on determining the trajectory is invalid, a modified trajectory; and
controlling the vehicle according to the modified trajectory.
2. The method of claim 1 ,
wherein determining the trajectory is invalid comprises determining, based at least in part on the modified drivable area and the trajectory, a cost, and
wherein the cost is based at least in part on a minimum distance to maintain between the vehicle and the object.
3. The method of claim 2 , further comprising:
determining, based at least in part on the object parameters, a predicted trajectory of the object,
wherein the cost is based at least in part on the predicted trajectory.
4. The method of claim 1 , wherein the object parameters indicate a lane change or merge by the object into a lane occupied by the vehicle.
5. The method of claim 1 , wherein controlling the vehicle according to the modified trajectory further comprises:
controlling the vehicle to follow the object based at least in part on one or more of a relative velocity of the object or a relative distance to the object.
6. The method of claim 1 , wherein determining that the trajectory is invalid comprises:
determining that the trajectory intersects or is within a threshold distance of a boundary of the modified drivable area.
7. The method of claim 1 , wherein the modified drivable area is based at least in part on at least one of a static object or a dynamic object in the environment.
8. A vehicle comprising:
a sensor disposed on the vehicle;
a processor; and
a computer-readable media storing instructions executable by the processor, wherein the instructions, when executed, cause the processor to perform operations comprising:
determining a drivable area for the vehicle;
determining, based at least in part on the drivable area, a reference trajectory for the vehicle to follow;
receiving sensor data from the sensor;
determining, based at least in part on the sensor data, object parameters of an object in an environment through which the vehicle is traversing;
modifying the drivable area, based at least in part on the sensor data and independent of whether the reference trajectory is valid in view of the object, to generate a modified drivable area that excludes a space that is either occupied by the object or that the object is predicted to occupy based on the object parameters;
determining, based at least in part on the modified drivable area and the reference trajectory, whether to modify the reference trajectory; and
controlling the vehicle based at least in part on whether to modify the reference trajectory.
9. The vehicle of claim 8 , wherein the object parameters comprise one or more of a velocity, a position, a classification, or an extent.
10. The vehicle of claim 9 , wherein the classification comprises a car, a bicycle, a motorcycle, or a pedestrian.
11. The vehicle of claim 8 , wherein, modifying the reference trajectory comprises:
maintaining a relative distance and a relative velocity between the vehicle and the object.
12. The vehicle of claim 8 , wherein the operations further comprise:
determining a predicted trajectory of the object; and
determining a probability associated with the predicted trajectory,
wherein controlling the vehicle is further based at least in part on the probability associated with the predicted trajectory.
13. The vehicle of claim 8 , wherein the object parameters of the object indicate a lane change or a merge.
14. The vehicle of claim 8 , wherein the object parameters of the object indicate a turn at an intersection.
15. The vehicle of claim 8 , wherein the modified drivable area is a first modified drivable area, and wherein the operations further comprise:
receiving additional sensor data based on a second object;
modifying the drivable area, based at least in part on the additional sensor data, to generate a second modified drivable area;
determining, based at least in part on the second modified drivable area and the reference trajectory, a cost; and
controlling the vehicle based at least in part on the cost and the first modified drivable area.
16. A non-transitory computer-readable medium storing instruction that, when executed, cause one or more processors to perform operations comprising:
receiving sensor data from a sensor on a vehicle;
determining a drivable area that represents a region in which the vehicle is permitted to traverse;
determining a trajectory within the drivable area for the vehicle to follow;
determining, based at least in part on the sensor data, an object parameter associated with an object;
modifying the drivable area, based at least in part on the object parameter and independent of the trajectory, to generate a modified drivable area that excludes a space that is either occupied by the object or that the object is predicted to occupy based on the object parameter;
determining, based at least in part on the modified drivable area and the trajectory, a cost;
determining, based at least in part on the cost, whether to modify the trajectory; and
controlling the vehicle based at least in part on the modified drivable area.
17. The non-transitory computer-readable medium of claim 16 , the operations further comprising:
determining a modified trajectory, the modified trajectory comprising a different longitudinal acceleration profile than the trajectory; and
controlling the vehicle according to the modified trajectory.
18. The non-transitory computer-readable medium of claim 16 , wherein the object parameter comprises one or more of a velocity, a position, a classification, or an extent.
19. The non-transitory computer-readable medium of claim 16 , wherein the cost is based at least in part on a distance between the vehicle and a boundary of the modified drivable area, and
wherein controlling the vehicle comprises controlling the vehicle based on an acceleration determined based, at least in part, on a relative distance or a relative velocity between the vehicle and the object.
20. The non-transitory computer-readable medium of claim 16 , the operations further comprising:
determining that the cost is above a threshold cost; and
controlling the vehicle to follow the object.