Abstract:
A method and system may identify vehicle collisions in real-time or at least near real-time based on statistical data collected from previous vehicle collisions. The statistical data may be used to train a machine learning model for identifying whether a portable computing device is in a vehicle collision based on sensor data from the portable computing device. The machine learning model may be trained based on a first subset of sensor data collected from vehicle trips involved in vehicle collisions and a second subset of sensor data collected from vehicle trips not involved in vehicle collisions. When a current set of sensor data is obtained from a portable computing device in a vehicle, the current set of sensor data is compared to the machine learning model to determine whether the portable computing device is in a vehicle involved in a vehicle collision.
Abstract:
A method based on separating ambient gravitational acceleration from a moving three-axis accelerometer data for determining a driving pattern is presented. A server may receive telematics data originating from a client computing device and combine the telematics data. The server may estimate a gravitational constant to the combined telematics data and generate a function for pitch and a roll angle from the combined telematics data. The server may further determine a driving pattern using at least the pitch and the roll angle.
Abstract:
A method for determining a yaw angle estimate or vehicle heading direction is presented. A potential range of yaw angles is generated based on a plurality of primary telematics data. One or more yaw angle estimates are generated from the potential range of yaw angles. A driving pattern is determined based on at least one of the yaw angle estimates. The primary telematics data is a plurality of telematics data originated from a client computing device. The effects of gravity have been removed from the plurality of telematics data in a first primary movement window.
Abstract:
A method and system may identify vehicle collisions in real-time or at least near real-time based on statistical data collected from previous vehicle collisions. A user's portable computing device may obtain sensor data from sensors in the portable computing device and compare the sensor data to a statistical model indicative of a vehicle collision, where the statistical model includes sensor data characteristics which correspond to the vehicle collision. Each sensor data characteristic may have a threshold value, the portable computing device may compare a value for the sensor data characteristic to the threshold value. If the portable computing device identifies a vehicle collision based on the comparison, the portable communication device may display collision information. Further, notifications may be sent to emergency contacts and/or emergency personnel to provide assistance to the user.
Abstract:
A method based on separating ambient gravitational acceleration from a moving three-axis accelerometer data for determining a driving pattern is presented. A server may receive telematics data originating from a client computing device and combine the telematics data. The server may estimate a gravitational constant to the combined telematics data and generate a function for pitch and a roll angle from the combined telematics data. The server may further determine a driving pattern using at least the pitch and the roll angle.
Abstract:
A computer implemented method for determining a yaw angle estimate or vehicle heading direction is presented. A data server may receive a plurality of telematics data originating from a client computing device and determine a first primary movement window from the telematics data. The data server may also determine a potential range of yaw angles from the plurality of telematics data from the first primary movement window and generate an equality that evaluates the potential range of yaw angles. The data server may further maximize the count of acceleration events of the telematics data from the first primary movement window to further generate one or more refined yaw angle estimates. The data server stores the one or more yaw angle estimates on a memory.
Abstract:
A computer implemented method for determining one or more idling time windows from a vehicle trip is presented. A data server may receive, via a computer network, a plurality of telematics data originating from a client computing device and identify primary movement data from the plurality of telematics data. The data server may also measure a total variance from the plurality of telematics data at one or more time stamps and determine an average total variance for an entire trip from the plurality of telematics data. The data server may further normalize total variance at the one or more time stamps using the generated average and determine one or more idling time windows from the normalized total variance.
Abstract:
A computer implemented method for determining a yaw angle estimate or vehicle heading direction is presented. A data server may receive a plurality of telematics data originating from a client computing device and determine a first primary movement window from the telematics data. The data server may also determine a potential range of yaw angles from the plurality of telematics data from the first primary movement window and generate an equality that evaluates the potential range of yaw angles. The data server may further maximize the count of acceleration events of the telematics data from the first primary movement window to further generate one or more refined yaw angle estimates. The data server stores the one or more yaw angle estimates on a memory.