Motion planning with variable grid resolution for graph-search-based planning
This document describes motion planning with variable grid resolution for graph-search-based planning. An example system includes a processor that obtains an initial pose, a goal pose, and an obstacle map for an environment. The processor uses a motion-planning algorithm to determine a path or trajectory using two or more grid resolutions for a graph-based search. The path includes a series of waypoints, including two-dimensional positional coordinates (and time coordinates if a trajectory), to navigate from the initial pose towards the goal pose. Operation of the host vehicle is then controlled to maneuver along the path using an assisted-driving or autonomous-driving system. In this way, motion planning is performed for the entire path but uses a coarser grid resolution for the portion nearer the goal pose. This allows motion planning for autonomous parking, especially in environments that include static and dynamic objects, to be handled in a more computationally-efficient manner.
1 . A method comprising:
obtaining an initial pose and a goal pose of a host vehicle;
obtaining an obstacle map for an environment that includes the initial pose and the goal pose;
determining, using the obstacle map and two or more grid resolutions for a graph-based search by a motion-planning algorithm, a path, the path including a series of waypoints that includes two-dimensional (2D) positional coordinates for the host vehicle to navigate from the initial pose toward the goal pose; and
controlling, using an assisted-driving or an autonomous-driving system, operation of the host vehicle to maneuver along the path toward the goal pose,
wherein the graph-based search for the path is performed using three grid resolutions; and
wherein the three grid resolutions include a first grid resolution, a second grid resolution, and a third grid resolution along each of at least two of a longitudinal dimension, a lateral dimension that is perpendicular to the longitudinal dimension, and a time dimension, the first grid resolution being smaller than the second grid resolution that is smaller than the third grid resolution,
the first grid resolution being adjacent to the initial pose, the third grid resolution being adjacent to the goal pose, and the second grid resolution being in between the first grid resolution and the third grid resolution.
2 . The method of claim 1 , wherein:
the graph-based search is performed in a two-dimensional search space that includes a longitudinal dimension and a lateral dimension that is perpendicular to the longitudinal dimension; and
the two or more grid resolutions include a first grid resolution and a second grid resolution along each of the longitudinal dimension and the lateral dimension, the first grid resolution being smaller than the second grid resolution, the first grid resolution being adjacent to the initial pose.
3 . The method of claim 2 , wherein:
the second grid resolution is at least twice as large as the first grid resolution along each of the longitudinal dimension and the lateral dimension.
4 . The method of claim 1 , wherein:
the graph-based search is performed in a three-dimensional search space that include a longitudinal dimension, a lateral dimension that is perpendicular to the longitudinal dimension, and a time dimension;
the two or more grid resolutions include a first grid resolution and a second grid resolution along each of the longitudinal dimension, the lateral dimension, and the time dimension, the first grid resolution being smaller than the second grid resolution, the first grid resolution being adjacent to the initial pose; and
the waypoints further include time coordinates of the host vehicle.
5 . The method of claim 4 , wherein:
the second grid resolution is at least twice as large as the first grid resolution along each of the longitudinal dimension, the lateral dimension, and the time dimension.
6 . The method of claim 1 , wherein the method further comprises:
selecting first waypoints of the path, the first waypoints comprising a subset of the series of waypoints for the host vehicle to navigate from the initial pose toward the goal pose;
in response to the assisted-driving or the autonomous-driving system completing operation of the host vehicle along the first waypoints, identifying an intermediate pose from among the first waypoints that is at a positional end of the first waypoints;
determining, using the two or more grid resolutions for the graph-based search by the motion-planning algorithm, second waypoints of the path for the host vehicle to navigate from the intermediate pose toward the goal pose; and
controlling, using the assisted-driving or the autonomous-driving system, operation of the host vehicle to maneuver along the second waypoints toward the goal pose.
7 . The method of claim 6 , wherein the subset of the series of waypoints comprises positional coordinates for a predetermined operation time of the host vehicle along the path.
8 . The method of claim 1 , wherein:
the environment includes one or more stationary objects or one or more moving objects; and
the series of waypoints avoids collisions between the host vehicle and the one or more stationary objects or the one or more moving objects.
9 . The method of claim 8 , wherein:
the motion-planning algorithm comprises a graph-search based algorithm;
the graph-search based algorithm uses space or space-time artificial potential fields for each of the one or more stationary objects and the one or more moving objects to avoid collisions with the one or more stationary objects or the one or more moving objects in the environment; and
the artificial potential fields include repulsive potential fields and at least one attractive potential field, a respective repulsive potential field being a function of a distance between the host vehicle and a respective stationary object or a respective moving object, the at least one attractive potential field including a goal potential field that is a function of a distance between the host vehicle and the goal pose.
10 . The method of claim 9 , wherein the at least one attractive potential field further includes a reference path potential field that is a function of a lateral distance between the host vehicle and a reference path, the lateral distance being perpendicular to the reference path.
11 . The method of claim 9 , wherein the graph-search based algorithm comprises a variant of a Hybrid A star (A*), A*, Dijkstra algorithm for finding an optimal path from the initial pose to the goal pose.
12 . The method of claim 1 , wherein:
the goal pose comprises a selected parking space, a position near the selected parking space, a position along a roadway in the environment, or an exit from the environment;
the initial pose is generated by a vehicle state estimator using location data;
the goal pose is generated by a parking space selector using the location data and other sensor data or map data; and
the other sensor data includes data from at least one of a camera system, a radar system, a lidar system, or an ultrasonic sensor system.
13 . A system comprising one or more processors configured to:
obtain an initial pose and a goal pose of a host vehicle;
obtain an obstacle map for an environment that includes the initial pose and the goal pose;
determine, using the obstacle map and two or more grid resolutions for a graph-based search by a motion-planning algorithm, a path, the path including a series of waypoints that includes two-dimensional (2D) positional coordinates for the host vehicle to navigate from the initial pose toward the goal pose; and
control, using an assisted-driving or an autonomous-driving system, operation of the host vehicle to maneuver along the path toward the goal pose,
wherein the graph-based search for the path is performed using three grid resolutions; and
wherein the three grid resolutions include a first grid resolution, a second grid resolution, and a third grid resolution along each of at least two of a longitudinal dimension, a lateral dimension that is perpendicular to the longitudinal dimension, and a time dimension, the first grid resolution being smaller than the second grid resolution that is smaller than the third grid resolution,
the first grid resolution being adjacent to the initial pose, the third grid resolution being adjacent to the goal pose, and the second grid resolution being in between the first grid resolution and the third grid resolution.
14 . The system of claim 13 , wherein:
the graph-based search is performed in a two-dimensional search space that includes a longitudinal dimension and a lateral dimension that is perpendicular to the longitudinal dimension; and
the two or more grid resolutions include a first grid resolution and a second grid resolution along each of the longitudinal dimension and the lateral dimension, the first grid resolution being smaller than the second grid resolution, the first grid resolution being adjacent to the initial pose.
15 . The system of claim 13 , wherein:
the graph-based search is performed in a three-dimensional search space that include a longitudinal dimension, a lateral dimension that is perpendicular to the longitudinal dimension, and a time dimension;
the two or more grid resolutions include a first grid resolution and a second grid resolution along each of the longitudinal dimension, the lateral dimension, and the time dimension, the first grid resolution being smaller than the second grid resolution, the first grid resolution being adjacent to the initial pose; and
the waypoints further include time coordinates of the host vehicle.
16 . The system of claim 13 , wherein the one or more processors are further configured to:
select first waypoints of the path, the first waypoints comprising a subset of the series of waypoints for the host vehicle to navigate from the initial pose toward the goal pose;
in response to the assisted-driving or the autonomous-driving system completing operation of the host vehicle along the first waypoints, identify an intermediate pose from among the first waypoints that is at a positional end of the first waypoints;
determine, using the two or more grid resolutions for the graph-based search by the motion-planning algorithm, second waypoints of the path for the host vehicle to navigate from the intermediate pose toward the goal pose; and
control, using the assisted-driving or the autonomous-driving system, operation of the host vehicle to maneuver along the second waypoints toward the goal pose.
17 . The system of claim 16 , wherein the subset of the series of waypoints comprises positional coordinates for a predetermined operation time of the host vehicle along the path.
18 . Computer-readable storage media comprising computer-executable instructions that, when executed, cause a processor to:
obtain an initial pose and a goal pose of a host vehicle;
obtain an obstacle map for an environment that includes the initial pose and the goal pose;
determine, using the obstacle map and two or more grid resolutions for a graph-based search by a motion-planning algorithm, a path, the path including a series of waypoints that includes two-dimensional (2D) positional coordinates for the host vehicle to navigate from the initial pose toward the goal pose; and
control, using an assisted-driving or an autonomous-driving system, operation of the host vehicle to maneuver along the path toward the goal pose,
wherein the graph-based search for the path is performed using three grid resolutions; and
wherein the three grid resolutions include a first grid resolution, a second grid resolution, and a third grid resolution along each of at least two of a longitudinal dimension, a lateral dimension that is perpendicular to the longitudinal dimension, and a time dimension, the first grid resolution being smaller than the second grid resolution that is smaller than the third grid resolution,
the first grid resolution being adjacent to the initial pose, the third grid resolution being adjacent to the goal pose, and the second grid resolution being in between the first grid resolution and the third grid resolution.