Patent classifications
G01C21/04
Precision Localization and Geofencing Governance System and Method for Light Electric Vehicles
A location and governance system and method for light electric vehicles that includes on-board sensors and receivers for providing readings used to compute absolute and relative vehicle position information, and combining the absolute and relative position information to compute a determined vehicle position, and a current surface type being traveled on by the vehicle. Governance commands for the vehicle can be generated based on the current surface type. Positioning system receivers, inertial measuring units, cameras and other sensor can be used. Vibration analysis, image processing, transition detection and other methods can be used to determine vehicle position and surface type, and spatial databases and other resources can be used. Determining a current surface type the vehicle is travelling on can include determining whether the vehicle is traveling on a sidewalk.
LANDMARK CONFIGURATION MATCHER
Systems and techniques for determining a list of geographic location candidates from an image of an environment are described. Open source data indicative of the earth's surface may be obtained and combined into grids to create region data to gain a representation of the earth's surface within each grid cell. An image of an environment may be obtained from an unknown camera. Image characteristics may be compared to the region data to determine error between the image characteristics and projections of the region data. A list of lowest error geographic location candidates may be determined and provided to the user.
LANDMARK CONFIGURATION MATCHER
Systems and techniques for determining a list of geographic location candidates from an image of an environment are described. Open source data indicative of the earth's surface may be obtained and combined into grids to create region data to gain a representation of the earth's surface within each grid cell. An image of an environment may be obtained from an unknown camera. Image characteristics may be compared to the region data to determine error between the image characteristics and projections of the region data. A list of lowest error geographic location candidates may be determined and provided to the user.
Integrated Visual Geo-Referencing Target Unit And Method Of Operation
Integrated Visual Geo-referencing Target Unit ABSTRACT A georeferencing target unit including: a generally planar top surface including a visual marker structure on the top surface, dimensioned to be observable at a distance by a remote visual capture device; an internal GPS tracking unit tracking the current position of the target unit; a microcontroller and storage means for storing GPS tracking data; and wireless network interconnection unit for interconnecting wirelessly with an external network for the downloading of stored GPS tracking data; a power supply for driving the GPS tracking unit, microcontroller, storage and wireless network interconnection unit, a user interface including an activation mechanism for activating the internal GPS tracking unit to track the current position of the target unit over an extended time frame and store the tracked GPS tracking data in the storage means.
Landmark configuration matcher
Systems and techniques for determining a list of geographic location candidates from an image of an environment are described. Open source data indicative of the earth's surface may be obtained and combined into grids to create region data to gain a representation of the earth's surface within each grid cell. An image of an environment may be obtained from an unknown camera. Image characteristics may be compared to the region data to determine error between the image characteristics and projections of the region data. A list of lowest error geographic location candidates may be determined and provided to the user.
Landmark configuration matcher
Systems and techniques for determining a list of geographic location candidates from an image of an environment are described. Open source data indicative of the earth's surface may be obtained and combined into grids to create region data to gain a representation of the earth's surface within each grid cell. An image of an environment may be obtained from an unknown camera. Image characteristics may be compared to the region data to determine error between the image characteristics and projections of the region data. A list of lowest error geographic location candidates may be determined and provided to the user.
Integrated visual geo-referencing target unit and method of operation
Integrated Visual Geo-referencing Target Unit ABSTRACT A georeferencing target unit including: a generally planar top surface including a visual marker structure on the top surface, dimensioned to be observable at a distance by a remote visual capture device; an internal GPS tracking unit tracking the current position of the target unit; a microcontroller and storage means for storing GPS tracking data; and wireless network interconnection unit for interconnecting wirelessly with an external network for the downloading of stored GPS tracking data; a power supply for driving the GPS tracking unit, microcontroller, storage and wireless network interconnection unit, a user interface including an activation mechanism for activating the internal GPS tracking unit to track the current position of the target unit over an extended time frame and store the tracked GPS tracking data in the storage means.
Multi-camera system for altitude estimation
Techniques are provided for vision-based altitude estimation using one or more platform mounted cameras. An embodiment includes generating projected ground imagery of imagery provided by cameras of the platform, the projection based on a hypothesized altitude. The method also includes obtaining reference ground imagery based on the location of the platform, the location based on platform navigation data. The method further includes registering the projected ground imagery to the reference ground imagery and generating a match score associated with the registration. The method further includes selecting the hypothesized altitude as the estimated altitude based on the match score (e.g., if the match score exceeds a threshold value or is maximized over a set of hypothesized altitudes. The method may further include otherwise adjusting the hypothesized altitude and repeating the altitude estimation process based on the adjusted hypothesized altitude to search for an improved estimated altitude based on the match score.
Multi-camera system for altitude estimation
Techniques are provided for vision-based altitude estimation using one or more platform mounted cameras. An embodiment includes generating projected ground imagery of imagery provided by cameras of the platform, the projection based on a hypothesized altitude. The method also includes obtaining reference ground imagery based on the location of the platform, the location based on platform navigation data. The method further includes registering the projected ground imagery to the reference ground imagery and generating a match score associated with the registration. The method further includes selecting the hypothesized altitude as the estimated altitude based on the match score (e.g., if the match score exceeds a threshold value or is maximized over a set of hypothesized altitudes. The method may further include otherwise adjusting the hypothesized altitude and repeating the altitude estimation process based on the adjusted hypothesized altitude to search for an improved estimated altitude based on the match score.
VEHICLE LOCALIZATION USING THE GROUND SURFACE WITH AN EVENT CAMERA
A method for estimating vehicle location by obtaining change events from an event camera's observations of a ground surface moving relative to the vehicle, determining a signature of the ground surface from the change events; and estimating the location using the signature. The change events may be processed to produce an 1st invariant representation of a ground surface patch for use as the signature. Alternatively, range measurements representing a patch may be used as the signature. A map is constructed having the representations of the ground surface patches including the locations of the patches. The same patch of ground surface is subsequently measured thereby obtaining a sequence of change events which are processed to produce a 2nd representation. The 2nd representation is matched to the map of 1st invariant representations. The location of the vehicle on the ground is determined based on the match.