# US 11,441,918 B2 · Machine learning model for predicting speed based on vehicle type > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,441,918 B2 - Document ID: US11441918B2 - Status: Granted - Kind: B2 - Application number: 16/328,930 - Filed: 2017-12-05 - Published: 2021-06-10 - Granted: 2022-09-13 - Earliest claimed: 2017-12-05 - Assignees: Google Inc. - Applicants: Google LLC - Inventors: Marc Nunkesser; Brendan Wallace; Stella Stylianidou - Web page: https://patentgrep.com/patent/US11441918B2 ## CPC - G01C 21/3492: {employing speed data or traffic data, e.g. real-time or historical (traffic control systems for road vehicles involving transmission of navigation instructions to the vehicle G08G1/0968)} - G01C 21/3484: {Personalized, e.g. from learned user behaviour or user-defined profiles} - G01C 21/3691: {Retrieval, searching and output of information related to real-time traffic, weather, or environmental conditions (arrangements for giving variable traffic instructions G08G1/09)} - G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting - G06N 3/04: Architecture, e.g. interconnection topology - G06N 3/08: Learning methods - G06N 3/09: Supervised learning - G06N 20/00: Machine learning - G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A computer-implemented method for predicting speeds for a particular vehicle type includes receiving first tracking data indicative of individual speeds of first vehicles while traveling on road segments at various times, and second tracking data indicative of individual speeds of second vehicles while traveling on the same road segments at the same times. The second vehicles correspond to the particular vehicle type. The method also includes training a machine learning model to predict speeds for the particular vehicle type using a feature set based on the individual speeds indicated by the first tracking data and labels based on the individual speeds indicated by the second tracking data. The method further includes using the model to predict a speed of a vehicle (of the particular type) on a road segment, at least by processing a real-time speed estimate corresponding to other vehicles traveling on the same road segment. ## Claims ### Claim 1 A computer-implemented method for predicting speeds for a particular vehicle type, the method comprising: receiving first tracking data indicative of individual speeds of a first plurality of vehicles while traveling on a plurality of road segments at a plurality of times; receiving second tracking data indicative of individual speeds of a second plurality of vehicles while traveling on the plurality of road segments at the plurality of times, the second plurality of vehicles corresponding to the particular vehicle type; training a machine learning model to predict speeds for the particular vehicle type using a feature set based on the individual speeds indicated by the first tracking data, and labels based on the individual speeds indicated by the second tracking data; using the trained machine learning model to predict a speed of a first vehicle on a first road segment, wherein the first vehicle corresponds to the particular vehicle type, and wherein the trained machine learning model predicts the speed of the first vehicle at least by processing a real-time speed estimate corresponding to one or more other vehicles of at least one different vehicle type traveling on the first road segment; and one or both of (i) using the predicted speed of the first vehicle to provide a predicted travel time to a user, and (ii) using the predicted speed of the first vehicle to provide a recommended route to the user. ### Claim 2 (depends on 1) The method of claim 1, wherein the first tracking data is further indicative of individual speeds of vehicles traveling on the plurality of road segments at times other than the plurality of times. ### Claim 3 (depends on 2) The method of claim 2, further comprising: using the first tracking data to determine free-flow speeds for the plurality of road segments, wherein the feature set is based on (i) the individual speeds of the first plurality of vehicles while traveling on the plurality of road segments at the plurality of respective times, and (ii) the free-flow speeds. ### Claim 4 (depends on 1) The method of claim 1, further comprising: receiving road type data indicative of road types associated with the plurality of road segments, wherein the feature set is based on (i) the individual speeds indicated by the first tracking data and (ii) the road types indicated by the road type data, wherein the first road segment corresponds to a first road type, and wherein using the trained machine learning model to predict the speed of the first vehicle on the first road segment corresponding to the first road type includes processing (i) the real-time speed estimate corresponding to the one or more other vehicles traveling on the first road segment and (ii) an indication of the first road type. ### Claim 5 (depends on 1) The method of claim 1, further comprising: receiving country data indicative of countries corresponding to the plurality of road segments, wherein the feature set is based on (i) the individual speeds indicated by the first tracking data, and (ii) the countries indicated by the country data. ### Claim 6 (depends on 1) The method of claim 1, wherein: receiving first tracking data includes receiving the first tracking data from a first plurality of mobile computing devices; and receiving second tracking data includes receiving the second tracking data from a second plurality of mobile computing devices. ### Claim 7 (depends on 1) The method of claim 1, further comprising: receiving a navigation request from a first mobile computing device, wherein using the trained machine learning model to predict the speed of the first vehicle on the first road segment is in response to receiving the navigation request from the first mobile computing device. ### Claim 8 (depends on 7) The method of claim 7, wherein one or both of (i) using the predicted speed of the first vehicle to provide the estimated travel time to the user, and (ii) using the predicted speed of the first vehicle to provide the recommended route to the user, includes transmitting one or both of the predicted travel time and the recommended route to the first mobile computing device via a network. ### Claim 9 (depends on 7) The method of claim 7, wherein receiving the navigation request from the first mobile computing device includes receiving an indication that the navigation request is associated with a first travel mode, the first travel mode corresponding to the particular vehicle type. ### Claim 10 (depends on 1) The method of claim 1, wherein the particular vehicle type is a motorized two-wheel vehicle. ### Claim 11 (depends on 1) The method of claim 1, wherein the second tracking data includes a plurality of speeds obtained using a satellite positioning technology. ### Claim 12 A computing system adapted to execute a method comprising: receiving first tracking data indicative of individual speeds of a first plurality of vehicles while traveling on a plurality of road segments at a plurality of times; receiving second tracking data indicative of individual speeds of a second plurality of vehicles while traveling on the plurality of road segments at the plurality of times, the second plurality of vehicles corresponding to a particular vehicle type; training a machine learning model to predict speeds for the particular vehicle type using a feature set based on the individual speeds indicated by the first tracking data, and labels based on the individual speeds indicated by the second tracking data; using the trained machine learning model to predict a speed of a first vehicle on a first road segment, wherein the first vehicle corresponds to the particular vehicle type, and wherein the trained machine learning model predicts the speed of the first vehicle at least by processing a real-time speed estimate corresponding to one or more other vehicles of at least one different vehicle type traveling on the first road segment; and one or both of (i) using the predicted speed of the first vehicle to provide a predicted travel time to a user, and (ii) using the predicted speed of the first vehicle to provide a recommended route to the user. ### Claim 13 (depends on 12) The computing system of claim 12, including: a first computing system that receives the first tracking data, receives the second tracking data, and trains the machine learning model; and a second computing system that uses the trained machine learning model to predict the speed of the first vehicle on the first road segment, and one or both of (i) uses the predicted speed of the first vehicle to provide the predicted travel time to the user, and (ii) uses the predicted speed of the first vehicle to provide the recommended route to the user. ### Claim 14 (depends on 12) The computing system of claim 12, wherein the method further comprises: receiving a navigation request from a first mobile computing device, wherein using the trained machine learning model to predict the speed of the first vehicle on the first road segment is in response to receiving the navigation request from the first mobile computing device. ### Claim 15 (depends on 14) The computing system of claim 14, wherein one or both of (i) using the predicted speed of the first vehicle to provide the estimated travel time to the user, and (ii) using the predicted speed of the first vehicle to provide the recommended route to the user, includes transmitting one or both of the predicted travel time and the recommended route to the first mobile computing device via a network. ### Claim 16 (depends on 12) The computing system of claim 12, wherein the first tracking data is further indicative of individual speeds of vehicles traveling on the plurality of road segments at times other than the plurality of times. ### Claim 17 (depends on 16) The computing system of claim 16, wherein the method further comprises: using the first tracking data to determine free-flow speeds for the plurality of road segments, wherein the feature set is based on (i) the individual speeds of the first plurality of vehicles while traveling on the plurality of road segments at the plurality of respective times, and (ii) the free-flow speeds. ### Claim 18 (depends on 12) The computing system of claim 12, wherein the method further comprises: receiving road type data indicative of road types associated with the plurality of road segments, wherein the feature set is based on (i) the individual speeds indicated by the first tracking data and (ii) the road types indicated by the road type data, wherein the first road segment corresponds to a first road type, and wherein using the trained machine learning model to predict the speed of the first vehicle on the first road segment corresponding to the first road type includes processing (i) the real-time speed estimate corresponding to the one or more other vehicles traveling on the first road segment and (ii) an indication of the first road type. ### Claim 19 (depends on 12) The computing system of claim 12, wherein the method further comprises: receiving country data indicative of countries corresponding to the plurality of road segments, wherein the feature set is based on (i) the individual speeds indicated by the first tracking data, and (ii) the countries indicated by the country data. ### Claim 20 (depends on 12) The computing system of claim 12, wherein: receiving first tracking data includes receiving the first tracking data from a first plurality of mobile computing devices; and receiving second tracking data includes receiving the second tracking data from a second plurality of mobile computing devices.