Vehicular localization systems, methods, and controls
AssigneeClearMotion, Inc.
InventorsYu Jiang, Vijayaraghavan Sridhar, John Parker Eisenmann, William Graves, Michael W. Finnegan
Life of this patent
Estimating the term…
Abstract
Localization systems, including road surface-based localization systems, are disclosed that may work in conjunction with less precise systems, such as GPS, to accurately locate a vehicle on a road segment. Methods and apparatus are disclosed for the identification and selection of landmarks, including road surface-based landmarks.
Claims27
A method for identifying landmarks associated with a road segment for localization, the method comprising:
- while a first vehicle traverses the road segment, collecting data from the first vehicle to obtain a sequence of data;
- dividing the obtained data sequence into a first series of snippets, wherein each snippet in the first series is associated with a portion of the road segment;
- determining a distinctiveness score for each snippet;
- identifying a first snippet, wherein the first snippet has the highest distinctiveness score of the first series of snippets;
- designating the first snippet as a first landmark;
- removing, from the first series of snippets, at least the first snippet and any snippet that overlaps with the first snippet, to form a second series of snippets;
- identifying a second snippet, wherein the second snippet has the highest distinctiveness score of the second series of snippets;
- designating the second snippet as a second landmark;
- storing, in computer memory, data characterizing the first landmark and the second landmark, and information about a location of the first landmark and a location of the second landmark.
The method of any of the preceding claims, wherein the location of the first landmark is stored as a relative location with respect to a location of a previously identified landmark.
A method for identifying landmarks in a road segment for localization, the method comprising:
- while a first vehicle traverses the road segment, collecting data from the first vehicle to obtain a sequence of data;
- dividing the obtained sequence of data into a first series of snippets, wherein each snippet in the first series is associated with a portion of the road segment;
- determining a distinctiveness score for each snippet;
- (i) identifying the snippet, of the series of snippets, having the highest distinctiveness score;
- (ii) designating the snippet identified in (i) as a landmark;
- (iii) removing, from the series of snippets, at least the snippet identified in (i) and any snippet that overlaps the snippet identified in (i);
- (iv) repeating steps (i)-(iii) until no snippets are left in the series of snippets or a predetermined number of snippets are obtained;
- storing, in computer memory:
- multiple landmarks and data about a location of each landmark.
The method of any of the preceding claims, wherein the distinctiveness score of a given snippet characterizes a degree of distinctiveness between the given snippet and any snippet within a predetermined area that includes the given snippet.
A method for identifying landmarks in a road segment for localization, the method comprising:
- while a first vehicle traverses a road segment, collecting data from the first vehicle to obtain a sequence of data;
- dividing the sequence of data into a series of overlapping data snippets, wherein each snippet is associated with a portion of the road segment;
- designating a first snippet of the series of snippets as a first landmark;
- designating a second snippet of the series of snippets as a second landmark;
- and
- storing, in computer memory, the first landmark, the second landmark, and a relative location of the second landmark with respect to the first landmark.
The method of any of the preceding claims, wherein the data is collected in the time domain, and wherein obtaining the sequence of data comprises transforming the collected data from the time domain to a distance domain based on an operating speed of the first vehicle.
The method of any of the preceding claims, wherein at least a portion of the collected data corresponds to vertical motion of a portion of the first vehicle or vertical displacement of a road surface.
The method of any of the preceding claims, wherein the collected data includes data associated with variations in the surface of the road segment.
The method of any of the preceding claims, wherein the collected data includes data associated with variations in subsurface properties of the road segment.
The method of any of the preceding claims, further comprising:
- using the stored landmarks and the relative location to determine a location of a second vehicle traversing the road segment.
The method of any of the preceding claims, comprising controlling a function of the second vehicle based at least in part on the determined location.
The method of any of the preceding claims, wherein at least a portion of the collected data is collected by a motion sensor (e.g., an IMU) that measures motion of a portion of the vehicle.
The method of any of the preceding claims, wherein at least a portion of the collected data is collected by a profilometer attached to the first vehicle and arranged to determine a profile of at least a portion of the surface of the road segment.
The method of any of the preceding claims, wherein at least a portion of the collected data is collected by a surface penetrating radar attached to the first vehicle and arranged to determine subsurface properties of the road segment.
The method of any of the preceding claims, wherein a length of each snippet and/or a spacing between consecutive snippets is dynamically varied based on a type of the road segment and/or an operating speed of the vehicle during collection of the data.
A method of operating a vehicle, the method comprising:
- using a first localization system (e.g., a GPS), identify an approximate location of the vehicle;
- based on the approximate location of the vehicle, identifying a road segment upon which the vehicle is traveling;
- accessing a database containing a plurality of landmarks;
- determining whether any of the landmarks in the database are associated with the road segment upon which the vehicle is traveling;
- upon determining that none of the landmarks in the database are associated with the road segment, initiating a landmark identification mode, wherein in the landmark identification mode, one or more landmarks are identified using any of the methods of the preceding claims.
A method of operating a first vehicle, the method comprising:
- using a first localization system to identify a location of the first vehicle;
- accessing a database that contains microclimate data as a function of location;
- determining, based on the microclimate data contained in the database, a particular microclimate associated with the location of the first vehicle;
- and
- adjusting one or more parameters of a vehicular system based at least in part on the determined particular microclimate.
A method of determining landmarks associated with a road for locating a vehicle's position along a road, the method comprising:
- collecting a continuum of data as a function of time and/or distance as a first vehicle travels along a load;
- identifying at least two portions of the road that are identifiably distinct from each other and all other portions of the road;
- determining the relative location of the at least two portions;
- storing, in computer memory, information related to certain characteristics of the two portions and the distance between them;
- providing data based at least partially on some of the information to a second vehicle.
A method of operating a vehicle, the method comprising:
- using a first localization system (e.g., a GPS), to determine an approximate location of the vehicle;
- based on the approximate location of the vehicle, identifying a road segment upon which the vehicle may be traveling;
- accessing a database containing information about a plurality of road surface-based landmarks associated with the identified road segment;
- receiving information about a first and a second landmark from the database and the relative distance between the first and the second landmarks;
- determining a more precise location of the vehicle on the road by matching vertical displacement experienced by a portion of the vehicle with certain characteristics of the first landmark;
- projecting an anticipated encounter with a second landmark based on information about the relative distance between the first and the second landmarks obtained from the database and on the basis of dead-reckoning between landmarks.
Dependencies are inferred from the claim text.