Apple Inc.

Sunny Kai Pang Chow

Xiaoyuan Tu

Hung A Pham

Gunes Dervisoglu

Ronald Keryuan Huang

Indranil S Sen

Ryan David Shelby

Anil K Kandangath

Umamahesh Srinivas

Lili Cao

Stephen P Jackson

Pejman Lotfali Kazemi

Digvijay Arjunrao Jadhav

Yann Jerome Julien Renard

Jonathan C King

Anh N Phan

Sriram Venkateswaran

Henry Choi

Martin M Menzel

Tejal Bhamre

Omar Aziz

Rebecca L Clarkson

Aniket Aranake

Alexandru Popovici

Yi Wen Liao

Paul D Bryan

Ethan Goolish

Alvaro J Melendez Hasbun

Tianye Sun

Richard G Liu

David Ojeda Avellaneda

Lauren Schutz

Pedro O Varangot

Embodiments are disclosed for crash detection on one or more mobile devices (e.g., smartwatch and/or smartphone). In some embodiments, a method comprises: detecting, with at least one processor, a crash event on a crash device; extracting, with the at least one processor, multimodal features from sensor data generated by multiple sensing modalities of the crash device; computing, with the at least one processor, a plurality of crash decisions based on a plurality of machine learning models applied to the multimodal features; and determining, with the at least one processor, that a severe vehicle crash has occurred involving the crash device based on the plurality of crash decisions and a severity model.

Crash detection on mobile device

In an example method, a mobile device connects a voice call for a user. The voice call causes one or more radio frequency transmitters of the mobile device to transmit radio waves at a first power level. Motion data describing movement of the mobile device is obtained, and the orientation of the mobile device is determined based on the motion data. A determination whether the mobile device is on the user's body or on an inanimate object is made based on the orientation of the mobile device over the period of time. The transmit power level is adjusted based on the determination.

Determination of device placement using pose angle

Systems, methods, devices and computer-readable mediums are disclosed for parking event detection and location estimation. In some implementations, a method comprises: determining, by a processor of a mobile device, a first activity state indicative of a possible parking event; obtaining, by the processor, a speed of the mobile device from a global navigation satellite system (GNSS) of the mobile device; obtaining, by the processor, pedometer data from a digital pedometer of the mobile device; determining, by the processor, a second activity state indicative of a possible parking event based at least in part on the GNSS speed and pedometer data; and responsive to the second activity state, estimating, by the processor, a location of the vehicle.

Parking event detection and location estimation

Motion sensors of a mobile device mounted to a vehicle are used to detect a mount angle of the mobile device. The motion sensors are used to determine whether the vehicle is accelerating or de-accelerating, whether the vehicle is turning and whether the mount angle of the mobile device is rotating. The mount angle of the mobile device is obtained from data output from the motion sensors and can be used to correct a compass heading. Data from the motion sensors that are obtained while the vehicle is turning or the mobile device is rotating are not used to obtain the mount angle.

Detecting mount angle of mobile device in vehicle using motion sensors

In some implementations, a computer-implemented method includes receiving a reading from a magnetometer of a mobile device. A cluster from a plurality of clusters of bias offsets generated from previously-calibrated readings is selected. The selected cluster has a representative bias offset, a mean of magnitudes in the selected cluster, and a magnitude threshold. An external magnetic field is estimated based on the reading and the representative bias offset for the selected cluster. Whether a magnitude of the estimated external field is within a magnitude range defined by the mean magnitude and the mean magnitude plus the magnitude threshold is determined.

Using clustering techniques to improve magnetometer bias estimation

Implementations are disclosed for validating data retrieved from a calibration database. In some implementations, calibrated magnetometer data for a magnetometer of a mobile device is retrieved from a calibration database and validated by data from another positioning system, such as course or heading data provided by a satellite-based positioning system. In some implementations, one or more context keys are used to retrieve magnetometer calibration data from a calibration database that is valid for a particular context of the mobile device, such as when the mobile device is mounted in a vehicle. In some implementations, currently retrieved calibration data is compared with previously retrieved calibration data to determine if the currently retrieved calibration data is valid.

Validating calibrated magnetometer data

Growing community of inventors

San Jose, CA, United States of America

Sunny Kai Pang Chow