Systems and methods for using image data to analyze an image
Systems and methods for training and executing machine learning models to generate lane index values are disclosed. A method includes identifying a set of image data captured by at least one autonomous vehicle when the at least autonomous vehicle is positioned in a lane of a roadway and respective ground truth localization data; determining a plurality of lane index values for the set of image data based on the ground truth localization data; labeling the set of image data with the plurality of lane index values, the lane index values representing a number of lanes from a leftmost or rightmost lane to the lane in which the at least one autonomous vehicle was positioned; and training, using the labeled set of image data, a plurality of machine learning models that generate a left lane index value and a right lane index value as output.
1 . A computer-implemented method for using image data to analyze an image via machine learning, comprising:
identifying, by one or more processors coupled to a non-transitory memory, an image from an operating ego vehicle; and
executing, by the one or more processors, a machine learning model to generate lane offset data based on the image, the machine learning model trained based on historical image data and historical lane offset data associated with the historical image data, the historical lane offset data including at least one lane offset of an autonomous vehicle in the historical image data relative to one or more lane lines of a lane along which the autonomous vehicle is traveling, the lane offset data including a distance from the autonomous vehicle to at least one of one or more lane lines of the lane.
2 . The method of claim 1 , wherein the image comprises light-detection and ranging (LiDAR) data.
3 . The method of claim 1 , further comprising:
generating, by the one or more processors, image feature data using the image; and
providing, by the one or more processors, the image feature data as input to the machine learning model.
4 . The method of claim 3 , wherein generating the image feature data using the image comprises detecting, by the one or more processors, a presence of a lane line based on the image.
5 . The method of claim 1 , wherein the lane offset data comprises a unidimensional distance from a longitudinal axis of the ego vehicle to a feature on a roadway on which the ego vehicle is operating.
6 . The method of claim 5 , wherein the feature on the roadway is a right lane marker.
7 . The method of claim 1 , further comprising correlating, by the one or more processors, the lane offset data with a determined position from a global navigation satellite system.
8 . The method of claim 1 , further comprising correlating, by the one or more processors, the lane offset data with data from an inertial measurement unit.
9 . The method of claim 1 , further comprising plotting, by the one or more processors, the lane offset data on a map associated with the ego vehicle.
10 . The method of claim 1 , wherein the historical image data from which the machine learning model is trained was captured from a region in which the ego vehicle is traveling.
11 . A system for using image data to analyze an image via machine learning, comprising:
a display;
a memory storing a machine learning model, the machine learning model trained based on historical image data and historical lane offset data associated with the historical image data, the historical lane offset data including at least one lane offset of an autonomous vehicle in the historical image data relative to one or more lane lines of a lane along which the autonomous vehicle is traveling, the lane offset data including a distance from the autonomous vehicle to at least one of one or more lane lines of the lane;
one or more processors operatively coupled to the display and the memory, the one or more processors configured to:
identify an image from an operating ego vehicle; and
execute the machine learning model to generate lane offset data based on the image.
12 . The system of claim 11 , wherein the image comprises light-detection and ranging (LiDAR) data.
13 . The system of claim 11 , wherein the one or more processors are further configured to:
generate image feature data using the image; and
provide the image feature data as input to the machine learning model.
14 . The system of claim 11 , wherein the one or more processors are further configured to generate image feature data using the image by performing operations comprising detecting a presence of a lane line based on the image.
15 . The system of claim 11 , wherein the lane offset data comprises a unidimensional distance from a longitudinal axis of the ego vehicle to a feature on a roadway on which the ego vehicle is operating.
16 . The system of claim 15 , wherein the feature on the roadway is a right lane marker.
17 . The system of claim 11 , wherein the one or more processors are further configured to correlate the lane offset data with a determined position from a global navigation satellite system.
18 . The system of claim 11 , wherein the one or more processors are further configured to correlate the lane offset data with data from an inertial measurement unit.
19 . The system of claim 11 , wherein the one or more processors are further configured to plot the lane offset data on a map associated with the ego vehicle.
20 . The system of claim 11 , wherein the historical image data from which the machine learning model is trained was captured from a region in which the ego vehicle is traveling.