Patent application number | Description | Published |
20090248605 | NATURAL LANGUAGE PARSERS TO NORMALIZE ADDRESSES FOR GEOCODING - The present invention provides a technique for building natural language parsers by implementing a country and/or jurisdiction specific set of training data that is automatically converted during a build phase to a respective predictive model, i.e., an automated country specific natural language parser. The predictive model can be used without the training data to quantify any input address. This model may be included as part of a larger Geographic Information System (GIS) data-set or as a stand alone quantifier. The build phase may also be run on demand and the resultant predictive model kept in temporary storage for immediate use. | 10-01-2009 |
20100153005 | SYSTEM AND METHOD FOR EFFICIENT ROUTING ON A NETWORK IN THE PRESENCE OF MULTIPLE-EDGE RESTRICTIONS AND OTHER CONSTRAINTS - Embodiments provide systems and methods that find the quickest route between two locations on a graph with multi-edge constraints in a time and space efficient manner. In some embodiments, Dijkstra's algorithm is split into separate universes when a) a multiple-edge constraint is reached, and b) along each edge of a multi-edge constraint. In some embodiments, the split is performed for the purpose of finding the quickest (i.e. lowest weighted) route to the intersection(s) at the end of the constraints. These universes, in some embodiments, are merged or discarded when the intersection at the end of the constraint is found. Using these systems and methods, in some embodiments, the shortest path between two locations of a multi-edge constrained road network can be efficiently determined. | 06-17-2010 |
20120158299 | SYSTEM AND METHOD FOR EFFICIENT ROUTING ON A NETWORK IN THE PRESENCE OF MULTIPLE-EDGE RESTRICTIONS AND OTHER CONSTRAINTS - Embodiments provide systems and methods that find the quickest route between two locations on a graph with multi-edge constraints in a time and space efficient manner. In some embodiments, Dijkstra's algorithm is split into separate universes when a) a multiple-edge constraint is reached, and b) along each edge of a multi-edge constraint. In some embodiments, the split is performed for the purpose of finding the quickest (i.e. lowest weighted) route to the intersect ion(s) at the end of the constraints. These universes, in some embodiments, are merged or discarded when the intersection at the end of the constraint is found. Using these systems and methods, in some embodiments, the shortest path between two locations of a multi-edge constrained road network can be efficiently determined. | 06-21-2012 |
20120226391 | VEHICLE ROUTE CALCULATION - This disclosure describes vehicle route calculation methods including “fitting” route segments between locations reported by a vehicle at different times along a route to create an overall “fitted” route. In certain embodiments, a vehicle management system generates a vehicle management user interface that displays the “fitted” route overlaid on a map. The “fitted” route can provide more useful tracking information to an administrator compared to simply displaying direct routes between known route stops. | 09-06-2012 |
20130297207 | VEHICLE FLEET ROUTING SYSTEM - A vehicle management system is described herein that can perform a variety of features related to managing fleets of vehicles. For example, the vehicle management system includes a routing module that can perform a variety of functions. Some of these functions can include determining a first route between locations using a simplified routing network and then determining a second route between the locations using a more exhaustive routing network. Further, the routing module can store information about the natural flow through intersections, improving control of routes through intersections. Some or all of these features may be used together, or any of these features may be omitted in any given embodiment. | 11-07-2013 |
20130338855 | SYSTEM FOR PROCESSING FLEET VEHICLE OPERATION INFORMATION - In one embodiment, a system for presenting fleet vehicle operation information in standardized forms includes a telematics module and a data standardizing module. The telematics module receives measurements related to operation of multiple vehicles in a fleet. The data standardizing module, using a first technique, estimates a first value for a parameter for at least one vehicle of the multiple vehicles based at least on the measurements. Further, the data standardizing module, using a second technique, estimates a second value for the parameter for at least one vehicle of the multiple vehicles based at least on the measurements. The second technique including using some measurement to estimate the second value different from the measurements used to estimate the first value according to the first technique. The data standardizing module outputs one or both of the first value and the second value for presentation to a user. | 12-19-2013 |
20150088414 | VEHICLE ROUTE SELECTION BASED ON ENERGY USAGE - Vehicle management systems and associated processes can consider energy consumption when selecting routes for fleet vehicles. Vehicle management systems and associated processes are described that, in certain embodiments, evaluate vehicle energy usage based on factors such as terrain or elevation, vehicle characteristics, driver characteristics, road conditions, traffic, speed limits, stop time, turn information, traffic information, and weather information, and the like. The features described herein may also be implemented for non-fleet vehicles, such as in personal vehicle navigation systems. | 03-26-2015 |
20150127324 | NATURAL LANGUAGE PARSERS TO NORMALIZE ADDRESSES FOR GEOCODING - The present invention provides a technique for building natural language parsers by implementing a country and/or jurisdiction specific set of training data that is automatically converted during a build phase to a respective predictive model, i.e., an automated country specific natural language parser. The predictive model can be used without the training data to quantify any input address. This model may be included as part of a larger Geographic Information System (GIS) data-set or as a stand alone quantifier. The build phase may also be run on demand and the resultant predictive model kept in temporary storage for immediate use. | 05-07-2015 |