Trajectory matching using ambient signals
First Claim
1. A computer implemented method, comprising:
- receiving, by a computer, collected trajectory data from a plurality of mobile devices on a plurality of crowdsourced paths, wherein the trajectory data for each path of the plurality of crowdsourced paths has associated therewith an ambient signal;
processing, by the computer, the collected trajectory data by;
identifying, by the computer, a set of anchor points in the trajectory data for each path of the plurality of crowdsourced paths respectively,generating a set of permutations of each set of anchor points from each path of the plurality of crowdsourced paths respectively, andselecting, by the computer, a best match of paths among the plurality of crowdsourced paths, the best match of paths selected at least partially based on (a) evaluating the set of permutations of each set of anchor points from each path to provide best matching sets of anchor points and (b) determining a signal similarity score between the associated ambient signal for each path of the plurality of crowdsourced paths respectively;
generating, by the computer, a shape of a path in a geographic area based on the best match of paths and providing a linked-node structure corresponding to a path map of the geographical area; and
storing the shape of the path in the geographic area and the linked-node structure in a memory accessible by the computer.
1 Assignment
0 Petitions
Accused Products
Abstract
A technique to automatically determine a shape of a path based on crowdsourced trajectory data is disclosed. Trajectory data is collected for a plurality of crowdsourced paths. The trajectory data for each of the plurality of crowdsourced paths is associated with at least one uniquely identifiable ambient signal. Anchor points may be identified corresponding to a transition in a characteristic of trajectory data of an individual crowdsourced path. Matching may be performed based on analyzing matching patterns of anchor points and evaluating the matches. Various embodiments involve using a linked node structure to represent one or more traces and/or indicate routes in a particular geographical area.
66 Citations
17 Claims
-
1. A computer implemented method, comprising:
-
receiving, by a computer, collected trajectory data from a plurality of mobile devices on a plurality of crowdsourced paths, wherein the trajectory data for each path of the plurality of crowdsourced paths has associated therewith an ambient signal; processing, by the computer, the collected trajectory data by; identifying, by the computer, a set of anchor points in the trajectory data for each path of the plurality of crowdsourced paths respectively, generating a set of permutations of each set of anchor points from each path of the plurality of crowdsourced paths respectively, and selecting, by the computer, a best match of paths among the plurality of crowdsourced paths, the best match of paths selected at least partially based on (a) evaluating the set of permutations of each set of anchor points from each path to provide best matching sets of anchor points and (b) determining a signal similarity score between the associated ambient signal for each path of the plurality of crowdsourced paths respectively; generating, by the computer, a shape of a path in a geographic area based on the best match of paths and providing a linked-node structure corresponding to a path map of the geographical area; and storing the shape of the path in the geographic area and the linked-node structure in a memory accessible by the computer. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A system comprising:
-
a processor;
a memory;a trajectory merging and correction unit; a trajectory matching unit; and a mapping unit; wherein the trajectory merging and correction unit, the trajectory matching unit and the mapping unit are implemented using one or more software modules arranged to be executed by the processor and stored in the memory; the system configured to receive trajectory data from a plurality of mobile devices on a plurality of crowdsourced paths, wherein the trajectory data for each path of the plurality of paths is associated with at least one ambient signal; and the processor configured to; identify, by the trajectory merging and correction unit, a set of anchor points in the trajectory data for each path of the plurality of crowdsourced paths respectively, generate a set of permutations of each set of anchor points from each path of the plurality of crowdsourced paths respectively, select, by the trajectory matching unit, a best match of paths among the plurality of crowdsourced paths, the best match of paths selected at least partially based on (a) an evaluation of the set of permutations of each set of anchor points from each path to provide best matching sets of anchor points and (b) a determination of a signal similarity score between the associated ambient signal for each path of the plurality of crowdsourced paths respectfully, determine a shape of a path in a geographic area based on the best match of paths, and store the shape of the path in the geographic area in the memory.
-
Specification