Location-based activity tracking
Inventors
Singleton, Alec • Kukka, Janne • Partanen, Jukka • SERGEEV, Dmitry • Akhter, Azeem
Assignees
Publication Number
US-12254971-B2
Publication Date
2025-03-18
Expiration Date
2041-08-24
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Methods, systems, and devices for activity tracking are described. A method for automatic activity detection may include receiving physiological data associated with a user from a wearable device and identifying an activity segment during which the user is engaged in a physical activity based on the physiological data. The method may further include identifying location data associated with the user for at least a portion of the activity segment and identifying one or more parameters associated with the physical activity based on the physiological data and the location data. The method may further include causing a user device to display the one or more parameters associated with the physical activity.
Core Innovation
The invention presents methods, systems, and devices for improved activity tracking by using physiological data collected from a user via a wearable device along with location data to identify activity segments during which the user is physically active. The system identifies parameters associated with the physical activity such as speed, pace, distance, route, and elevation gain, then displays these parameters on a user device.
The problem being addressed is the deficiency in conventional wearable activity tracking techniques, which often rely solely on physiological or motion data and may require user input to confirm activity segments. These drawbacks lead to inaccurate tracking because some physical activity occurring before user confirmation is ignored, and activity completion is sometimes not detected automatically, resulting in erroneous inclusion of data after activity ends.
The innovation leverages both physiological data and location data to automatically detect activity start and stop points, characterize activities more accurately, and retroactively calculate parameters from the true start time to improve classification and tracking. Location data enhances the accuracy of parameter identification such as distance, pace, and route mapping, and enables better differentiation between activity types and conditions (e.g., indoor vs outdoor). Leveraging location data also allows for continuous location monitoring, enabling activity detection with greater temporal precision compared to prior solutions.
Claims Coverage
The claims include one independent method claim and one independent apparatus claim focused on improved automatic activity detection using physiological and location data, including recognition of prior routes for more accurate activity segment completion determination and parameter identification.
Incorporation of previous routes for activity completion detection
Identifying one or more previous routes associated with previous activity segments of a user, and automatically identifying completion of a current activity segment without user input by comparing the ending location of the current route to the ending locations of the previous routes.
Use of physiological data and location data for accurate activity segmentation
Receiving physiological data from a wearable device, identifying an activity segment based on that data, and identifying location data for at least a portion of the activity segment to determine one or more parameters of the physical activity.
Automatic completion detection without user input
Automatically identifying a completion of the activity segment based at least in part on identifying a route end point that matches previously identified routes, without requiring user input.
Activity parameter determination based on physiological and location data
Determining parameters such as activity type, duration, distance, elevation change, calories burned, pace, speed, route map, split time, and elevation-adjusted pace, based on physiological data and location data after identifying activity segment completion.
User interface interaction for activity confirmation and parameter adjustment
Displaying an indication of the activity segment on a graphical user interface, receiving a user confirmation or input modifying parameters, and basing display and calculation of parameters at least partly on received user interactions.
Use of a machine learning model for completion detection
Inputting location data, physiological data, previous routes, and past physiological data into a machine learning model to aid automatic identification of activity segment completion.
Activity classification using confidence values influenced by location data
Receiving activity classification data including multiple classified activity types with confidence values, determined based on physiological and location data, and using that data to identify parameters.
Applicability to wearable ring devices
The wearable device collecting physiological data is specified as a wearable ring device.
The claims cover a system and method for activity detection that automatically identifies an activity segment and its completion by leveraging prior user routes and combining physiological data with location data. The inventive features emphasize automatic completion detection without user input, parameter determination enhanced with location information, user interface integration for confirmation and editing, and the use of machine learning to improve detection accuracy, specifically implemented with wearable ring devices.
Stated Advantages
Improved efficiency and accuracy of activity tracking by leveraging both physiological and location data.
More accurate determination of activity start and stop times, reducing errors from delayed user confirmation.
Enhanced differentiation between activity types and indoor versus outdoor activities through location data.
Ability to retroactively calculate activity parameters over the entire activity segment, including periods before user confirmation.
Continuous location tracking enables activity detection with greater temporal precision compared to conventional solutions.
Automated detection of activity segment completion reduces reliance on user input and prevents overestimation of activity duration and calories burned.
Improved user engagement via graphical user interface displaying detailed activity segments and allowing confirmation and editing.
Personalized machine learning classifiers trained with user confirmations and location data improve activity detection and classification over time.
Documented Applications
Tracking physical activities such as running, walking, cycling, hiking, swimming, skiing, and indoor activities like treadmill running or stationary cycling.
Generating activity segments with parameters including duration, calories burned, distance, pace, elevation, route mapping, and intensity.
Displaying detected and confirmed activity segments and their detailed parameters via a user interface on wearable-associated devices.
Training machine learning classifiers for activity detection personalized to users based on physiological data, location data, and user confirmations or edits.
Differentiating indoor versus outdoor activities and specific activities by analyzing location patterns and physiological parameters.
Integrating location information to provide additional health related insights like effects of elevation changes on physiological metrics and readiness scores.
Providing controlled location tracking feature opt-in and permission request flows to ensure user privacy.
Interested in licensing this patent?