Method, apparatus, and system detecting vehicle operator holding mobile device

Abstract

Systems, methods, apparatuses, and computer program products for allowing detection that a mobile devi ce is held in an operator' s hand while operating a vehicle. One method may include selecting a time window of at least one motion parameter; determining a magnitude ofthe at least one motion parameter; determining a moving variance for the time window; and comparing the moving variance to at least one predetermined threshold.

Description

“METHOD, APPARATUS, AND SYSTEM DETECTING VEHICLE OPERATOR HOLDING MOBILE DEVICE”BACKGROUND

[0001] Certain existing techniques may allow detection that a mobile device is held in a vehicle operator’s hand while operating the vehicle. The mobile device may include at least one inertial sensor (e.g., microelectromechanical system (MEMS) accelerometers, gyroscopes, magnetometers). The inertial sensors may detect signals exceeding predetermined thresholds (e.g., acceleration in meters per second squared (m / s2)). However, these techniques may have disadvantages by being limited to only detecting a particular moment in time that the operator lifts the mobile device, and failing to detect the entire duration of time that the operator holds the mobile device.

[0002] Other existing techniques may use artificial intelligence and / or machine learning models to analyze inertial sensor signals to determine that the operator is currently holding a mobile device. While such techniques may be able to detect the entire duration of time that the operator holds the mobile device, the techniques require a significant amount of training data to train the machine learning model to do the detection.

[0003] SUMMARY OF THE INVENTION

[0004] In order to overcome the above-identified disadvantage of the prior art, the present invention refers to a computer-implemented method comprising: for a mobile device located inside a vehicle; selecting a time window of at least one motion parameter of motion of the mobile device; determining a magnitude of the at least one motion parameter of the motion of the mobile device; and determining a moving variance for the time window; comparing the moving variance to at least a first predetermined threshold; responsive to determining that the moving variance exceeds the at least first predetermined threshold, determining that an operator of the vehicle is holding the mobile device inside the vehicle.

[0005] In a preferred embodiment, the method further comprises, responsiveto the determining that the moving variance exceeds the at least first predetermined threshold, warning the operator to put down the mobile device.

[0006] In a preferred embodiment, the warning comprises generating one or more of a vibration, a sound, a flashing screen, or a flash of a bulb on the mobile device to alert the operator.

[0007] In a preferred embodiment, the method further comprises, responsive to the determining that the moving variance exceeds the at least first predetermined threshold, determining that the operator is engaging in distracted driving, and generating actuarial data to reflect the distracted driving.

[0008] In a preferred embodiment, the vehicle is selected from the group consisting of a car, a truck, a trolley, a train car, a subway car, an airplane, a bicycle, a scooter, a motorcycle, a tractor, construction equipment, and heavy machinery.

[0009] In a preferred embodiment, the at least one motion parameter comprises at least one of: rotation rates of a gyroscope; accelerations of an accelerometer; rotation rates of at least two sequential accelerations of the accelerometer; magnetic field values from a magnetometer; and rotation rates of at least two sequential accelerations of the magnetometer.

[0010] In a preferred embodiment, the method further comprises: identifying at least one time range during which the operator is holding the mobile device.

[0011] In a preferred embodiment, the method further comprises: comparing the moving variance to at least a second threshold; and responsive to determining that the moving variance falls below the at least second predetermined threshold, determining that the operator is no longer holding the mobile device inside the vehicle.

[0012] In a preferred embodiment, the method further comprises: adjusting the at least first predetermined threshold in response to a determination that the vehicle is traveling on a smooth roadway or on a rough roadway.

[0013] In a preferred embodiment, the method further comprises: adjusting the at least second predetermined threshold in response to a determination that the vehicle is traveling on an unimproved roadway.

[0014] In a preferred embodiment, the method further comprises capturing the time window of values of a magnetic field from the magnetometer.

[0015] In a preferred embodiment, a magnitude of the magnetic field values for each time step in the time window is computed using the following equation:

[0016] In a preferred embodiment, the rotation rate from two sequential accelerations of the magnetometer is computed using the following equation:

[0017] In a preferred embodiment, the method further comprises capturing the time window of acceleration values from the accelerometer.

[0018] In a preferred embodiment, a magnitude of the acceleration values for each time step in the time window is computed using the following equation:

[0019] In a preferred embodiment, the rotation rate from two sequential accelerations of the accelerometer is computed using the following equation: k • afc-tfc-i

[0020] In a preferred embodiment, the method further comprises capturing the time window of rotation rates from the gyroscope.

[0021] In a preferred embodiment, a magnitude of the rotation rates for each time step in the time window is computed using the following equation:BRIEF DESCRIPTION OF THE DRAWINGS

[0022] For a proper understanding of example embodiments, reference should be made to the accompanying drawings, wherein:

[0023] Fig. 1 depicts an operator holding a mobile device in their hand;

[0024] Fig. 2 depicts an operator raising their hand while holding a mobile device;

[0025] Fig. 3A depicts a side view of an operator sitting in and operating a vehicle while holding a mobile device in one hand;

[0026] Fig. 3B depicts a front view of an operator sitting in and operating a vehicle while holding a mobile device in one hand;

[0027] Fig. 4 depicts moving variance of rotation rate in radians per second;

[0028] Fig. 5 illustrates a spread of variance of rotation rate;

[0029] Fig. 6 illustrates an example of a flow diagram of a method according to various example embodiments;

[0030] Fig. 7 illustrates an example of a moving variance of rotation rate of a gyroscope over time;

[0031] Fig. 8 illustrates an example of a moving variance of acceleration over time;

[0032] Fig. 9 illustrates an example of a moving variance of rotation rate from an acceleration vector over time;

[0033] Fig. 10 illustrates an example of a moving variance of magnetic field over time; and

[0034] Fig. 11 illustrates an example of a moving variance of rotation rate from a magnetic field vector over time.DETAILED DESCRIPTION

[0035] Certain example embodiments described herein may have various benefits and / or advantages to overcome the disadvantages described above. For example, certain example embodiments may reduce the complexity of existing detection algorithms while improving reliability and accuracy in detection of an operator holding a mobile device. Some example embodiments may prevent false positives by setting a lower-bound threshold higher than the variance caused by an unimproved road type. Various example embodiments may improve the accuracy of the estimation of handling events, and reduce the occurrence of false positives. In addition, certain example embodiments may effectively distinguish between handling events and other motion types, and be applied to a range of road types. Thus, certain example embodiments discussed below are directed to improvements in computer technology.

[0036] Certain example embodiments discussed herein may identify an entire duration of time that an operator holds a mobile device in their hand while in a moving vehicle. The duration of time may be defined as the time that the operator picks up the mobile device to the time that the operator releases the mobiledevice from their hand. While a vehicle is moving, there are vibrations and accelerations, some linear and some rotational, generated along all of the axes (x- axis, y-axis, and z-axis). When the operator holds a mobile device in the moving vehicle, the mobile device experiences these rotational vibrations and accelerations, which are translated to the operator’s hand and arm, and are not immobile during these motions. When the mobile device is in a stationary location in the vehicle (e.g., secured in a phone cradle, on a vehicle seat, in the operator’s pocket, in the operator’s purse, and the like), the mobile device may experience reduced rotational vibrations and accelerations compared to being held in the operator’s hand. Accordingly, monitoring changes in rotational vibrations and accelerations of the mobile device can enable a determination of whether the operator is holding the device.

[0037] There are various sources of vibration and accelerations as a vehicle travels along a road. For example, the road on which the vehicle is traveling may be paved with asphalt and / or may be uneven (e.g. , unpaved, gravel, road reflectors, potholes, speedbumps, debris). Various characteristics of the vehicle itself may be the source of vibration and acceleration (e.g., tire misalignment, underinflation or overinflation in one or more tires, improper balance in one or more tires, defects in the steering mechanism or vehicle suspension, engine acceleration, braking, and changes in direction of travel).

[0038] Whatever the source of vibration and / or acceleration, because an operator would hold the mobile device in their hand, which connects to the operator’s arm, which connects to the operator’s torso, there are various possible movements of the mobile device in the operator’s hand, along different axes. As a result, the vibrations and accelerations may be transferred through the operator’s body to the mobile device as rotational or translational movement. In contrast, the mobile device may experience significantly less road vibration or acceleration while in a more stationary position, such as a phone cradle, pants pocket, purse, or vehicle seat.

[0039] Ordinarily skilled artisans also will appreciate that the vehicle may be of any number of types. By way of non-limiting example, the types may include a car, a truck, a trolley, a train car, a subway car, an airplane, a scooter, a motorcycle, a tractor, construction equipment, and heavy machinery. The just-described vehiclesmay be powered by an internal combustion engine (ICE), an electric motor or motors, or a hybrid propulsion system including both ICE and one or more electric motors.

[0040] Ordinarily skilled artisans will appreciate that, in an embodiment, the mobile device may be running one or more programs, or one or more apps, to perform the detection of rotational or translational movement of the device inside the vehicle. Accordingly, the mobile device will have one or more processors, with associated non-transitory memory connected to the one or more processors and storing instructions which, when executed on the one or more processors, enable the mobile device to perform the detection described herein.

[0041] In an embodiment, the one or more programs or apps may connect the mobile device to computer hardware and software inside the vehicle to enable the mobile device to determine whether the vehicle is in operation (i.e. the engine or motor is running). This connection may be useful, for example, in an emergency situation when an operator is inside the vehicle, which is moving, but not under power, and not under operator control, possibly necessitating the generation of an alert or other type of message to emergency personnel.

[0042] In an embodiment, while the vehicle is in operation, the mobile device may determine that the operator is engaging in distracted driving. In response, in an embodiment, the mobile device may provide a warning to the operator about holding the mobile device while operating the vehicle. By way of non-limiting example, the warning may include one or more of a vibration, a sound, a flashing screen, or a flash of a bulb on the mobile device to alert the operator. The warning may tell the operator to put down the mobile device. In an embodiment, the mobile device may generate actuarial data to reflect the operator’s distracted driving.

[0043] From the foregoing discussion and the following description, ordinarily skilled artisans will appreciate that the method, apparatus, and system of the present invention provides a practical application for the detection of an operator holding a mobile device, from warning the operator to put down the mobile device and stop engaging in distracted driving, to generating actuarial data regarding the operator’s engaging in distracted driving, possibly leading to increased insurance rates. If the operator truly is engaging in distracted driving, it may be impractical for the app or software on the mobile device to instruct the vehicle to cease operation. In an embodiment, the app or software on the mobile device may instruct the vehicle tomove to a location where the vehicle can be turned off.

[0044] Referring now to the various drawing figures, in Fig. 1, an operator holds mobile device 1 with their hand 2 while operating a vehicle and sitting in front of steering wheel 3.

[0045] Similarly to Fig. 1, Fig. 2 depicts the operator moving the position of mobile device 1 while held in hand 2, specifically, raising mobile device 1 in direction 4 (which may be in a direction opposite to the direction of gravity, as shown, or which may be along one or more different axes). Accordingly, while direction 4 is shown as upward, the operator could move mobile device 1 in one or multiple directions simultaneously (e.g., down, left, right, away from the operator, closer to the operator). Direction 4 could be any movement of mobile device 1 in a trajectory that causes a change in acceleration.

[0046] Fig. 3 A illustrates an operator seated in seat 5. The operator’s torso 6 is connected to the operator’s arm 7 and the operator’s wrist 8. The operator is holding mobile device 9 in the operator’s hand 10 which is connected to wrist 8.As described above, changes in acceleration may be transferred from seat 5, to torso 6, to arm 7, to wrist 8, to hand 10, and finally to mobile device 9. These changes may result from road conditions, or from vehicle conditions which affect the vehicle’s interaction with the road. All of these conditions may be referred to as “road noise.” Fig. 3B illustrates a front view perspective of the operator in Fig.3A.

[0047] Fig. 4 displays data detected by inertial sensors (e.g., moving variance of rotation rate in radians per second, CD) in a mobile device in three different locations: on a vehicle seat, secured in a phone cradle, and held in an operator’s hand. Based upon this data, Fig. 5 displays the spread of variance of rotation rate in these three different locations.

[0048] In an embodiment, variance of the rotation and / or translation rate of the mobile device over a current duration of time may be compared to the variance of rotation and / or translation rates from one or more prior durations of time. If the variance of rotation rate is above a predetermined threshold (e.g, 2X magnitude), it may be determined that the operator is holding the mobile device in their hand. Alternatively, when the variance is below or within a predetermined tolerance (e.g, range) of a prior variance value, it may be determined that the operator is notholding the mobile device in their hand.

[0049] Depending on the embodiment, the prior variance value may be a value taken at a particular prior point in time. The prior variance value may be a weighted value taken from several prior points in time. In an embodiment, it may be desirable to measure variance more frequently, for example, when it is determined that the vehicle is traveling over an uneven or otherwise inconsistent road surface. In an embodiment, it may be desirable to measure variance less frequently, for example, when it is determined that the vehicle is traveling over a relatively smooth road surface, for example, when the operator is on a freeway or interstate highway. In an embodiment, GPS measurements may be used to determine and predict the road type on which the operator is traveling.

[0050] Fig. 6 illustrates an example of a flow diagram of a method that may be performed by a mobile device, according to various example embodiments. As an example, the mobile device may include one or more of a mobile phone, smart phone, personal digital assistant (PDA), tablet, or portable media player, digital camera, pocket video camera, video game console, navigation unit, such as a global positioning system (GPS) device, desktop or laptop computer, single-location device, such as a sensor or smart meter, or any combination thereof.

[0051] At step 601, the method may include detecting motion mechanics of the mobile device, such as rotation, vibration, and acceleration. For example, one or more inertial sensors in the mobile device may detect motion, such as rotational vibration, that the mobile device may experience, and output signals indicating the detected motion. The inertial sensors may be configured to detect multi-axis movement, including in any of the x-, y-, and z-axis, as well as rotation in any of these axes.

[0052] In an embodiment, the mobile device may include a variety of inertial sensors, such as MEMS accelerometers, gyroscopes, and magnetometers, that provide signal outputs associated with the detected motion of the mobile device. For example, accelerometers may be configured to detect a change in acceleration of the mobile device. Similarly, gyroscopes may be configured to detect a rotational rate of the mobile device. In addition, magnetometers may be configured to detect magnetic field strength, and thereby provide signal output on a change in direction of the mobile device.

[0053] At step 602, the method may include computing a rotation rate from two sequential accelerations. For example, rotation rate may be derived from a change in direction of an acceleration vector from an acceleration sensor (e.g., accelerometer) by computing the rotation rate from two sequential accelerations according to r = — —(~°tfc otfc~1where a stands for acceleration at time k.

[0054] In certain example embodiments, rotation rate may be derived from a change in direction of a magnetic field vector from a magnetometer sensor by computing the rotation rate from two sequential accelerations according to r =, where m signifies magnetic field at time k. k~ tk-1

[0055] At step 603, the method may include selecting a time window of at least one specific parameter, such as rotation rates from a gyroscope; accelerations from an accelerometer; rotation rates of the two sequential accelerations of the accelerometer sensor in step 602; magnetic field values from the magnetometer sensor; and rotation rates of the two sequential accelerations of the magnetometer sensor in step 602.

[0056] At step 604, the method may include determining a magnitude of rotation rate.

[0057] In some example embodiments, with respect to rotation rates from the gyroscope at step 602, magnitude of rotation rate for each time step may be determined according t

[0058] In various example embodiments, with respect to magnetic field values from the magnetometer in step 602, magnitude of magnetic field for each time step may be determined according to |m| =

[0059] At step 605, the method may include determining moving variance (z.e., moving variance) for the time windows calculated at step 604 (z.e., rotation rates, accelerations, magnetic field values).

[0060] At step 606, the method may include comparing the signal outputs against at least one predetermined threshold. The at least one threshold may be predetermined empirically based on previously measured values. For example, the predetermined threshold may be a value with unit of measurement m / s2. Depending on the embodiment, the predetermined threshold may vary, but as an example, may be twice a baseline value. Also depending on the embodiment, as discussed earlier,the baseline value may be determined in a number of different ways.

[0061] In various example embodiments, the at least one threshold may be computed dynamically. For example, a baseline value of variance may be captured. When the variance falls below the baseline value, the variance may be established as the new baseline value. If the variance exceeds the baseline value by a predetermined amount (in an embodiment, double the baseline value), it may be determined that the operator is holding the mobile device. Alternatively, when the variance drops within a range of the prior magnitude of variance (in an embodiment, within 0 to 20% above the baseline value), it may be determined that the operator no longer is holding the mobile device.

[0062] Certain example embodiments may include two thresholds: an upper threshold to determine if handling has started, and a lower threshold to determine if handling has stopped. These thresholds may be predetermined or computed dynamically, as discussed above. If the variance exceeds an upper threshold value, it may be determined that the operator is holding the mobile device. In an embodiment, this upper threshold value may also be a multiple (for example, 2x) of the value of the baseline value that, when exceeded, indicates that the operator is holding the mobile device. If the variance drops below the lower threshold value, it may be determined that he operator no longer is holding the mobile device.

[0063] In some instances, the variance of an unimproved road type may be multiple times greater than that of a smooth road type. That difference in magnitude may satisfy the threshold conditions and thereby falsely indicate that the operator is holding the mobile device. To prevent this, in an embodiment the lower-bound threshold may be set higher than the variance caused by an unimproved road type. In other embodiments, the lower-bound threshold may be set sufficiently high that it still can be determined that the operator is holding the mobile device because the variance exceeds the upper-bound threshold.

[0064] At step 607, the method may include identifying a time range in which the operator is holding the mobile device. Specifically, the time range may be defined as the moment in time that the operator picks up the mobile device to the moment of time that the operator releases the mobile device from their hand. In an embodiment, such a determination would depend on a duration that the motion exceeds the predetermined threshold at step 606. In various exampleembodiments, the magnitude of the variance may fall below the threshold such that multiple periods may be determined that the mobile device is in hand, as shown in Fig. 1.

[0065] In certain example embodiments, the magnitude of the variance momentarily may fall below a threshold such that multiple periods may be determined that the mobile device is in hand when the mobile device is still in hand, as depicted by two peaks in Fig. 1. These periods may be merged if the end of an event is within a specific time range of a subsequent event, which may capture the entire duration of the handling event, thus resolving the initial issue of measuring only when the operator picks up the mobile device.

[0066] In some example embodiments, a G-graph may be created to determine if multiple handling events are part of one event. For example, nodes or vertices of a G-graph may be the indices of the identified handling events. Edges may be inserted between the nodes to reveal the connected handling events. These connected components may represent the entire duration of each handling event. For each handling event, the nodes with the lowest and highest index, along with the associated timestamps, may indicate the start and the end of the handling event.

[0067] The variance of the rotation rate of the mobile device over a duration of time may be compared to the variance of rotation rates from a prior duration of time. If the variance of rotation rate is above the predetermined threshold (e.g. , 2X magnitude), it may be determined that the operator is holding the mobile device in their hand. Alternatively, when the variance is below or within a predetermined tolerance (e.g. , range) of a prior magnitude of variance, according to the various techniques described earlier, it may be determined that the operator is not holding the mobile device in their hand.

[0068] In various example embodiments, handling may be determined to have started when the threshold of variance in rotation rate is exceeded, and to have stopped when the threshold is not exceeded. In various example embodiments, if the mobile device is not in hand, and the vehicle transitions from a smooth and improved road type (e.g., asphalt) to an unimproved road type (e.g., bumps and gravel), there is the risk that the transition in road types will satisfy the predetermined threshold for handling and the algorithm will falsely classify the mobile device as being in hand.

[0069] Fig. 7 depicts an example of 3-second window of moving variance of rotation rate from the gyroscope over a period of time. Fig. 7 illustrates the data points where the mobile device may be considered to be in an operator’s hand, as well as periods where the road type considered to be unimproved, rather than smooth and improved (e.g., asphalt). The magnitude of variance shown in Fig. 7 may be due to the mobile device in hand is distinct from the magnitude of variance due to an unimproved road type. That is, the peak in variance for mobile device in hand may be an order of magnitude greater than the peak in variance for unimproved road. Therefore, the selection of a threshold between the two peaks may prevent false positives because of a transition in road type.

[0070] Similar to Fig. 7, Fig. 8 illustrates an example of a 3-second window moving variance of acceleration plotted over time. Fig. 8 illustrates an example where the mobile device is in a mount and the road type is smooth and improved (e.g., asphalt). Unlike the moving variance of rotation rate, no threshold value of moving variance of acceleration may distinguish mobile device in hand from unimproved road type.

[0071] In another example, Fig. 9 depicts a 3-second window moving variance of rotation rate from acceleration vector plotted over time. The mobile device may be in a mount and the road type is smooth and improved (e.g, asphalt) unless otherwise labeled. Unlike the moving variance of rotation rate from the gyroscope, no threshold value of moving variance of rotation rate from the accelerometer may distinguish mobile device in hand from unimproved road type.

[0072] In an additional example, Fig. 10 illustrates a 3-second window moving variance of magnetic field plotted over time. Again, the mobile device is in a mount and the road type is smooth and improved (e.g., asphalt) unless otherwise labeled. Unlike the moving variance of rotation rate from the gyroscope, no threshold value of moving variance of magnetic field can distinguish mobile device in hand from unimproved road type. As shown, there is a spike when the mobile device is put in hand, but it is not possible to use a threshold to identify the end of the event, and therefore it is rendered useless to determine the duration of the event.

[0073] Furthermore, Fig. 11 depicts a 3-second window moving variance of rotation rate from magnetic field vector plotted over time. The mobile device is ina mount and the road type is smooth and improved (e.g., asphalt) unless otherwise labeled. Unlike the variance of rotation rate from the gyroscope, no threshold value of variance of rotation rate from the magnetic field can distinguish mobile device in hand from unimproved road type. Fig. 11 depicts a spike when the mobile device is put in hand, but it is not possible to use a threshold to identify the end of the event, and therefore it is rendered useless to determine the duration of the event.

[0074] Partial Glossary

[0075] GPS Global Positioning System

[0076] MEMS Microelectromechanical System

[0077] PDA Personal Digital Assistant

[0078] While the invention has been described in detail above with respect to several embodiments, ordinarily skilled artisans will appreciate that there may be variations within the scope and spirit of the invention. Accordingly, the invention is to be measured by the scope of the following claims.

Claims

[0079] WE CLAIM:

1. A computer-implemented method comprising: for a mobile device (1) located inside a vehicle; selecting a time window of at least one motion parameter of motion of the mobile device (1); determining a magnitude of the at least one motion parameter of the motion of the mobile device; and determining a moving variance for the time window; comparing the moving variance to at least a first predetermined threshold; responsive to determining that the moving variance exceeds the at least first predetermined threshold, determining that an operator of the vehicle is holding the mobile device (1) inside the vehicle.

2. The method of claim 1, further comprising, responsive to the determining that the moving variance exceeds the at least first predetermined threshold, warning the operator to put down the mobile device (1), preferably wherein the warning comprises generating one or more of a vibration, a sound, a flashing screen, or a flash of a bulb on the mobile device (1) to alert the operator.

3. The method of any claim hereinbefore, further comprising, responsive to the determining that the moving variance exceeds the at least first predetermined threshold, determining that the operator is engaging in distracted driving, and generating actuarial data to reflect the distracted driving.

4. The method of any claim hereinbefore, wherein the vehicle is selected from the group consisting of a car, a truck, a trolley, a train car, a subway car, an airplane, a bicycle, a scooter, a motorcycle, a tractor, construction equipment, and heavy machinery.

5. The method of any claim hereinbefore, wherein the at least one motion parameter comprises at least one of: rotation rates of a gyroscope; accelerations of an accelerometer; rotation rates of at least two sequential accelerations of the accelerometer; magnetic field values from a magnetometer; and rotation rates of at least two sequential accelerations of the magnetometer.

6. The method of any claim hereinbefore, further comprising:identifying at least one time range during which the operator is holding the mobile device (1).

7. The method of any claim hereinbefore, further comprising: comparing the moving variance to at least a second threshold; and responsive to determining that the moving variance falls below the at least second predetermined threshold, determining that the operator is no longer holding the mobile device (1) inside the vehicle.

8. The method of claim 1, further comprising: adjusting the at least first predetermined threshold in response to a determination that the vehicle is traveling on a smooth roadway or on a rough roadway or on an unimproved roadway.

9. The method of claim 7 or 8, further comprising: adjusting the at least second predetermined threshold in response to a determination that the vehicle is traveling on a rough roadway or on an unimproved roadway.

10. The method of claim 5, further comprising capturing the time window of values of a magnetic field from the magnetometer or the time window of acceleration values from the accelerometer or the time window of rotation rates from the gyroscope.

11. The method of claim 10, wherein a magnitude of the magnetic field values for each time step in the time window is computed using the following equation:

12. The method of claim 11, wherein the rotation rate from two sequential accelerations of the magnetometer is computed using the following equation: cos- r = -13. The method of claim 10, wherein a magnitude of the acceleration values for each time step in the time window is computed using the following equation:

14. The method of claim 13, wherein the rotation rate from two sequential accelerations of the accelerometer is computed using the following equation: cos-fytk • r =15. The method of claim 10, wherein a magnitude of the rotation rates for each time step in the time window is computed using the following equation:

Images