Self-driving vehicle path adaptation system and method
Provided is a path adaptation system and method for a self-driving vehicle, and a self-driving vehicle including path adaptation technology. The method includes providing a vehicle navigation path through an environment; determining a location of the vehicle relative to the path; generating one or more polygons representing a shape of the vehicle projected from the vehicle location in a vehicle travel direction long the path; combining the one or more projections into a projected motion polygon; acquiring object information representing an object shape from one or more sensors; and determining if the object shape overlaps the projected motion polygon. If so, the method includes adjusting the vehicle navigation. If not, the method includes, continuing navigation along the path. The path adaptation system includes one or more processors configured to execute the path adaptation method.
1 . A path adaptation method for a self-driving vehicle, comprising:
navigating the vehicle on a planned path through an environment; and
while the vehicle moves along the planned path:
determining a location of the vehicle relative to the planned path;
projecting the vehicle's travel, including:
generating a three-dimensional shape of variable dimension representing the vehicle and projecting the three-dimensional shape from the vehicle location along the planned path to form a plurality of projections; and
combining the plurality of projections into a single, continuous, three-dimensional path projection along the planned path;
acquiring object information representing an object shape from one or more sensors; and
determining if the object shape overlaps the three-dimensional path projection and:
if so, adjusting the vehicle navigation; or
if not, continuing the vehicle navigation along the planned path.
2 . The method of claim 1 , wherein determining a location of the vehicle relative to the planned path includes processing forward kinematic projections based on current odometry readings for the vehicle.
3 . The method of claim 1 , wherein the environment is represented by an evidence grid.
4 . The method of claim 3 , further comprising representing the object shape with a point cloud in the evidence grid, wherein points in the point cloud indicate occupancy of a one or more points in space.
5 . The method of claim 4 , wherein determining if the object shape overlaps the three-dimensional path projection includes determining whether at least one point from the point cloud intersects with or is within the three-dimensional path projection.
6 . The method of claim 1 , wherein the evidence grid is a 3-dimensional evidence grid and the object information is 3-dimensional object information.
7 . The method of claim 1 , wherein the one or more sensors include stereo sensors, stereo cameras, lidar, and/or sonar.
8 . The method of claim 1 , further comprising representing the three-dimensional shape as a polyhedron.
9 . The method of claim 1 , further comprising generating the three-dimensional shape to have a first width that corresponds to an approximate width of the vehicle, and is not less than the width of the vehicle.
10 . The method of claim 9 , further comprising generating the three-dimensional shape to have a second width, different from the first width, that represents a width of a load transported by the vehicle.
11 . The method of claim 1 , wherein generating the three-dimensional path projection includes combining the plurality of projections into a 3-dimensional solid projection tunnel representing a volume that would be swept out by the vehicle.
12 . The method of claim 1 , wherein the vehicle generates a right object sensing field and/or a left object sensing field, and the method further includes:
enabling the vehicle to navigate on the planned path when an object is detected in the right and/or left object sensing field, but the object shape does not have a point on or within the three-dimensional path projection.
13 . The method of claim 1 , wherein adjusting the vehicle navigation includes adjusting a vehicle speed.
14 . The method of claim 1 , wherein adjusting the vehicle navigation includes slowing without stopping the vehicle.
15 . The method of claim 1 , wherein adjusting the vehicle navigation includes stopping the vehicle.
16 . The method of claim 1 , wherein adjusting the navigation includes adjusting the planned path to avoid the object.
17 . The method of claim 1 , wherein the vehicle planned path is a trained path.
18 . The method of claim 1 , wherein when the vehicle planned path includes a planned stop, the method includes:
projecting the vehicle motion to a trained stop, but not projecting the vehicle motion beyond the trained stop;
the vehicle stopping at the trained stop; and
resuming projecting the vehicle motion beyond the trained stop when the vehicle is ready to resume navigation.
19 . The method of claim 1 , wherein when the vehicle planned path includes a change in direction, the method includes:
projecting the vehicle motion in the current direction up to a change of direction point;
the vehicle stopping or slowing to change from the current direction to the new direction at the change of direction point; and
then projecting the vehicle motion in the new direction.
20 . The method of claim 1 , wherein the vehicle is an autonomous vehicle in the form of a pallet truck or tugger configured to transport a load.