Vehicle location determination using synthetic aperture radar

A non-transitory computer-readable storage medium for use in conjunction with an electronic device, the computer-readable storage medium storing program instructions, wherein, when executed by the electronic device, the program instructions cause the electronic device to perform one or more operations comprising:

• transmitting, using a first subset of one or more sensors, radar signals approximately perpendicular to a direction of motion of a vehicle having one or more non-retractable wheels in contact with a surface, wherein the first subset of the one or more sensors has an associated angular resolution, wherein approximately perpendicular is within the angular resolution of perpendicular to the direction of motion, and wherein the one or more sensors comprise multiple antennas;
• receiving, using a second subset of the one or more sensors, reflected radar signals, wherein the first subset of the one or more sensors and the second subset of the one or more sensors are at different locations in the electronic device, and wherein the first subset of the one or more sensors comprises a first subset of the antennas that transmit the radar signals and the second subset of the one or more sensors comprises a second subset of the antennas that receive the reflected radar signals;
• analyzing a time sequence of the reflected radar signals;
• performing an additional electromagnetic or acoustic measurement in a different band of frequencies than the radar signals, wherein the additional electromagnetic or acoustic measurements comprise one of:
• optical imaging in a visible frequency band, an infrared measurement in an infrared frequency band, or a sonar measurement;
• identifying at least a stationary object using the analyzed time sequence of reflected radar signals and the additional electromagnetic or acoustic measurement as inputs to a machine-learning model, wherein at least the identified stationary object is man-made and has a corner or an edge, and wherein the identifying comprises classifying at least the stationary object;
• determining a location of the vehicle based at least in part on at least the identified stationary object and a predefined map of stationary objects and associated predetermined locations, wherein the determining comprises calculating the location based at least on the predetermined location of at least the identified stationary object;
• and
• dynamically adapting the first subset of the antennas and the second subset of the antennas based at least in part on:
• an environmental condition, a size of the stationary object, a distance between the electronic device and the stationary object, or the determined location.

A method for generating a dataset, comprising:

• by an electronic device:
• transmitting, using a first subset of one or more sensors, radar signals approximately perpendicular to a direction of motion of a vehicle having one or more non-retractable wheels in contact with a surface, wherein the first subset of the one or more sensors has an associated angular resolution, wherein approximately perpendicular is within the angular resolution of perpendicular to the direction of motion, and wherein the one or more sensors comprise multiple antennas;
• receiving, using a second subset of the one or more sensors, reflected radar signals, wherein the first subset of the one or more sensors and the second subset of the one or more sensors are at different locations in the electronic device, and wherein the first subset of the one or more sensors comprises a first subset of the antennas that transmit the radar signals and the second subset of the one or more sensors comprises a second subset of the antennas that receive the reflected radar signals;
• analyzing a time sequence of the reflected radar signals;
• performing an additional electromagnetic or acoustic measurement in a different band of frequencies than the radar signals, wherein the additional electromagnetic or acoustic measurements comprise one of:
• optical imaging in a visible frequency band, an infrared measurement in an infrared frequency band, or a sonar measurement;
• identifying at least a stationary object using the analyzed time sequence of reflected radar signals and the additional electromagnetic or acoustic measurement as inputs to a machine-learning model, wherein at least the identified stationary object is man-made and has a corner or an edge, and wherein the identifying comprises classifying at least the stationary object;
• determining a location of the vehicle based at least in part on at least the identified stationary object and a predefined map of stationary objects and associated predetermined locations, wherein the determining comprises calculating the location based at least on the predetermined location of at least the identified stationary object;
• and
• dynamically adapting the first subset of the antennas and the second subset of the antennas based at least in part on:
• an environmental condition, a size of the stationary object, a distance between the electronic device and the stationary object, or the determined location.