GB2485971A - Transmitting recorded data in the event of a road vehicle accident - Google Patents

Abstract

A method and system comprises measuring vehicle data, including vehicle acceleration data, and recording the measured data. The measured data is analysed to detect a crash event and in the event of a crash being detected, sending the recorded data to a data storage device separate from the vehicle. The data at the remote storage device can then be analysed to determine accident causes and faults for insurance and prosecution purposes. A plurality of accelerometers each having the same alignment direction but different sampling frequencies may be used to detect the crash with a threshold acceleration of 1.5G being used to trigger the crash detection. A GPS receiver may alternatively be used to calculate acceleration and used to send a location to the remote data storage device. The data transmitted to the separate data storage device can include a time period before and after the detected crash. The method and system also includes sending vehicle data to the data storage device at predetermined intervals even if a crash is not detected.

Description

A METHOD OF ANALYSING A ROAD VEHICLE ACCIDENT EVENT

The present invention relates to a method of analysing a road vehicle accident event and relates particularly, but not exclusively, to a method and apparatus for analysing vehicle crash data and reducing fraudulent insurance claims.

When reviewing an existing road network, with a view to improving driver safety, it is common place to identify accident black-spots, that is, places with an unusually high incidence of accidents. Databases of road traffic accident fatalities are well known and are used to form the basis of a review of a road network to identify potential improvements. Similarly, databases of accidents, for example that have led to insurance claims, can be used to identify areas where accidents have occurred. However, such data only identifies places where accidents have happened and the volume of data, in particular with fatalities, is so small as to make it extremely difficult to draw statistically significant conclusions from the data. Furthermore, the data can only be analysed retrospectively after a significant period of time and therefore a temporary change to a road, such as the introduction of road works which creates a potential accident black-spot, cannot be identified.

Techniques for identifying poor driving habits, and informing the car driver, are well-known. However, the basis on which this analysis takes place is very simple and cannot be regarded as universally applicable to all driving situations. As a result, a driver may begin to ignore the information sent to them making it ineffective. Alternatively, the driver may overly rely on a driver safety device and engage in inappropriate driving when it is not safe to do so due to other factors not considered by the device.

Vehicle "black-box" recorders are well-known for use in vehicles as well as in aeroplanes.

Such devices allow partial reconstruction of the events surrounding a crash to be analysed. However, this data may be lost or interfered with making the presence of the black-box in the vehicle pointless. For example the box may be damaged by fire or immersed in water may not be admissible in a court if the device can be removed as it could be tampered with.

Preferred embodiments oI the present invention seek to overcome the above described

disadvantages of the prior art.

According to an aspect of the present invention, there is provided a method of analysing a road vehicle accident event, comprising the steps of:-measuring vehicle data including at least vehicle acceleration data; recording at least some of said data; analysing said vehicle data to detect a crash event; and if a crash event is detected, sending at least some of said recorded data to a data storage device separate from the vehicle.

By sending data gathered immediately prior to a crash event to a data storage device the advantage is provided that the data is safe and easily available for later analysis of the accident. As a result: even if the data recorder were to be destroyed, or rendered unreadable, the data that provides a lot of information about the crash will have already been transmitted back to the data storage device. Furthermore, even if the data on the recorder is not rendered unreadable, the second copy of the data would make tampering with the data, which may be used as evidence, significantly more difficult. This will therefore assist in allowing the data to be used in evidence in court. It is also the case that the transferred data can be received nearly instantly at the storage device allowing near instantaneous analysis of the data. This analysis can then be used to immediately alert the emergency services if the accident appears sufficiently serious from the analysed data, which can in turn significantly reduce the time taken for help to arrive at the scene of an accident.

In a preferred embodiment the acceleration data is measured by a plurality of accelerometers.

* A plurality of accelerometers allows a more detailed analysis of the data recorded.

In another preferred embodiment a plurality of said accelerometers operate in the same direction.

By having more than one accelerometer working in the same direction the advantage is provided that the different accelerometers can work in different ranges of acceleration providing different information.

In a further preferred embodiment the accelerometers measure acceleration at different sampling frequencies.

The vehicle data that is analysed to detect a crash event may include said vehicle acceleration data.

In a preferred embodiment a crash event is detected if the acceleration is less than -1.5 G (14.7ms2).

In another preferred embodiment at least some vehicle data is sent to a data storage device prior to the detection of a crash event.

By sending data (generally at a lower rate than in the event of a crash) to the data storage device prior to a crash event, the advantage is provided that the behaviour of the driver throughout the journey before the accident can be assessed. It is therefore possible to identify whether the driver was driving in a manner that was likely to lead to a crash or had been engaging in safe driving techniques through the earlier part of the journey.

In a further preferred embodiment if a crash event is detected, substantially all of said measured data over a predetermined period of time prior to and during the crash is sent to said data storage device.

According to another aspect of the present invention, there is provided an apparatus for analysing a road vehicle accident event, comprising the steps of:-data measuring means for measuring vehicle data including at least vehicle acceleration data; data recording means for recording at least some of said data; processing means for analysing said vehicle data to detect a crash event; and data transfer means for sending at least some of said recorded data to a data storage device separate from the vehicle in the event that a crash event is detected.

In a preferred embodiment the data measuring means comprise a plurality of accelerometers.

in another preferred embodiment a plurality of said accelerometers operate in the same direction.

In a further preferred embodiment the accelerometers measure acceleration at different sampling frequencies.

The processor may analyse said acceleration data to detect a crash event.

In a preferred embodiment a crash event is detected if the acceleration is less than -2 G (1 9.6ms2).

In another preferred embodiment the data transfer means sends at least some vehicle data to a data storage device prior to the detection of a crash event.

In a further preferred embodiment, ft a crash event is detected, said data transfer means transfers substantially all of said measured data over a predetermined period of time prior to and during the crash is sent to said data storage device.

Preferred embodiments of the present invention will now be described by way of example only, and not in any limitative sense, with reference to the accompanying drawings in which: Figure 1 is a schematic representation of the apparatus of the present invention; Figure 2 is a schematic representation of the data measuring device shown in Figure 1; Figure 3 is a flow chart showing the steps undertaken in the method of the present invention; Figure 4 is a flow chart showing the steps in measuring, categorising and transferring the data used in the present invention; Figure 5 is a flow chart showing another aspect of the present invention; and Figure 6 is a flow chart showing a further aspect of the present invention.

Referring to Figures 1 and 2, a vehicle 10 carries within it a data measuring device 12.

This device 12 includes a location determining device in the form of GFS receiver device 14 that receives OPS signals from a plurality of GFS satellites 16. The data measuring device also has at least one vehicle data measuring device, in the form of an acceleration measuring device or accelerometers 18, that measure acceleration and deceleration along at least one, preferably two and ideally three perpendicularly separated axes (i.e. along the X, V and 7 axes, with the X axis being the forward direction of travel of the vehicle 10). It should be noted that throughout this application the term acceleration refers to both acceleration (where the velocity of a vehicle increases in the direction of travel, which is a positive acceleration) and deceleration (where the velocity of the vehicle decreases in the direction of travel, which is a negative acceleration). The data measuring device 12 also has a processor 20 that processes the data produced by the GPS receiver 14 and accelerometer 18. The processor 20 has an output to a first transmitter 22. The first transmitter 22 typically is a GPRS data transmitter although any wireless data transmission means would be suitable, including but not limited to Satellite, Paknet, 3G, 4G, TDMS, Selective Tone signalling, Mobitex, WiFi.

The output from the first transmitter is received by a first receiving aerial 24 that is connected into a network containing one or more data storage and processing servers 26.

A second transmitter 28 is also connected to server 26 and can transmit GPRS data. A second receiving aerial 3D in data measuring device 12 in vehicle 10 receives the GPRS data from the second transmitter 28. Server 26 is also connected to computer 32 and this connection is typically via the internet 34 although there may be a direct connection or by other networking means. The apparatus described above transmits data using a circuit switching network technique to transmit between the apparatus 10 on vehicle 12 and the data storage and processing server 26 and back again. However, greater data efficiency can be obtained using a packet switching system using an internet connection from the data processing server and a mobile telephone I mobile internet connection. Furthermore, clearly any means for transferring data from device to server and back again will be sufficient to fulfil the needs of the present invention.

The apparatus set out above, with reference to Figures 1 and 2, is used in a method of improving driver safety which will now be described with additional reference to Figures 3 and 4. Data measuring device 12 measures first data from a plurality of vehicles 10, step 50. This first data includes acceleration of the vehicle (including a positive acceleration, an increase in velocity and a negative acceleration or deceleration as a decrease in velocity). This acceleration is measured by accelerometer 18 measuring the acceleration and deceleration of the vehicle in the X direction of travel. The OPS receiver 14 receives GPS data from a plurality of satellites 16 and this data is processed to provide vehicle location data.

It should be noted that the acceleration data could alternatively be provided by accelerometers already forming part of the vehicle 10 and the data measuring device being connected to onboard diagnostics either via the OBD connector, a non intrusive CANBus measuring device or direct into the vehicle Engine Management System of vehicle 10 to provide this acceleration data to the processor 20. In addition to the acceleration and location data, time data is also measured. This time measurement can either be using time received from the GF'S satellite or from another onboard time recording device. Further data of distance is also measured and this can either be measured by using the location data or by using other distance measuring means on the vehicle 10. Similarly, velocity measurements can be taken either by interpretation of the GPS data or from velocity data measured by vehicle 10.

The first data and second data can be used to calculate celebration which uses acceleration data to rate a drivers behaviours. An example of how celeration is calculated can be found in Af Wâhlberg, A. E.(2008) Driver celeration behaviour and accidents -an analysis', Theoretical Issues in Ergonomics Science, 9:5, 383 -403.

At least some of the first data or second derived data is used to identify incidents where the data exceeds predetermined levels. Typically this is either where the acceleration or celeration data exceeds a predetermined threshold level. The speed of travel of the vehicle is ideally also incorporated in this categorisation since the significance of a vehicle's acceleration or deceleration varies depending on the speed of travel. These determination steps are therefore completed using a lookup table including velocity against either celeration or acceleration.

A further example of the second data derived from the first data is the angle of turn of the vehicle which is calculated using the sideways acceleration of the vehicle. This is done by using measurements taken by the accelerometers 18 in the Y plane, perpendicular to the direction of travel of the vehicle, which is the sideways movement of the vehicle. Angle of turn cal also be calculated from the GE'S location data.

At least some of the first data and/or second data are transferred from the data measuring device 12 via the first transmitter 22 (and the first aerial receiver 24 to the data storage and processing server 26. Sampling of the first and second data can take place at 4Hz and all of this data may be transferred to server 26. However, the majority of this data is generally of no significance and therefore ideally the data is categorised into incident data, step 52, before the data transfer step, 54, takes place.

The operation of steps 50, 52 and 54 are set out in more detail in Figure 4. At the start of a cycle, at step 56, counters for time t, distance d, angle of turn 6, velocity v, celeration a, maximum velocity Vmax, maximum angle of turn 6max and maximum Cm are all reset. The time t, distance d, angle of turn 6, velocity v, celeration a and GPS data are sampled at a frequency of 4Hz, step 58. As set out above, the time, distance, angle of turn, velocity and celeration may be calculated or derived from a combination of the GE'S data (together with, if necessary, accelerometer measurements). At step 60 it is determined whether celeration a has exceeded the maximum recorded value for celeration since the last counter reset Cm, whether the current velocity has exceed the maximum velocity since the counter reset v3 or whether the current angle of turn U has exceeded the maximum angle of turn since the counter reset. If c is greater than max then, at step 62, c is recorded as equal to the current celeration a. Similarly if v is greater than Vmax then Vmax is set to the current velocity v and if 8 is greater than 8max then 6max is set to the current angle of turn 8.

At step 64 it is determined whether Crnax has exceeded the threshold level for celeration Cpjj. If Cmax is greater than Cthresn, (in other words, if the vehicle is accelerating, braking or swerving too violently) then a data report is sent at step 66. This data report includes distance travelled since last data report sent, the maximum velocity reached, Vmax since the last data report was sent, the current velocity v, the current celeration c, the maximum celeration since the last data report was sent, Cmag the current angle of turn 0, the maximum angle of turn 0max since the last data report was sent, and GPS data including the current location, time and current heading and speed (other data may also be S included). When the data report is sent at step 66, the counters are reset at step 56 and the cycle restarts. This data report is referred to as an incident data report since the celeration (closely related to acceleration or deceleration) has exceeded a predetermined level requiring data to be sent to the server 26.

Returning to step 64, if Cmax is not greater than CmrQ$h, then a series of further tests are applied to determine whether to send a data report. These tests relate to the time since the last data report was sent, the distance travelled since the last data report was sent and the angle of turn of the vehicle or the speed of the vehicle. These data transfers are referred to as event data reports and do not indicate necessarily an incident of excessive braking, acceleration or swerving. At step 68, it is determined whether the time since the last data report was sent has exceeded the threshold time tjhrh. If t is greater than t(hrh, a data report is sent at step 66 and the counter reset at step 56. It is generafly, although not necessarily, the case that the data sent in an event data report is the same as in an incident data report (although the trigger for sending the report is different). Similarly at step 70, if the distance travelled since the last data report ci was sent has exceed the threshold distance tfhre, then a data report is sent at step 66 and the counter reset at step 56. It is clear that when a data report is sent at step 66, as a result of a positive outcome from the tests supplied at steps 68 and 70 (time and distance), that these are simply periodic reporting of the data and do not indicate any dangerous manoeuvring of the vehicle.

If the distance travelled since the last data report was sent ci does not exceed the threshold distance dre$h at step 70 then a further test is applied at step 72. In this test the maximum angle of turn of the vehicle 0max is compared to the threshold value °resh and if O,, is greater than O(esh then an event data report is sent at step 66 and the counter reset at step 56. If umax was not greater than 9thrcsh then a tinal test at step 74 is applied determining whether the maximum velocity of travel since the last reset has exceeded the threshold value V8 is greater than Vresn. If Vmax is greater than VEIJreSh then an event data report is sent at step 66 and the counter reset at step 56. Since the angle of turn of the vehicle i9 and velocity could be regarded as indicators of dangerous driving behaviour, then the data reports sent at step 66 as a result of positive results from tests 72 and 74 could be regarded as incident data rather than event data. If Vm does not exceed VtJiresh then further 4Hz sampling takes place at step 56.

Once significant quantities of data have been gathered from a plurality or multiplicity of vehicles, the incident data can be used to identify potential accident black spots at step 76. The incident data for a plurality of vehicles is analysed to identify areas with high occurrences of incidents in approximately the same Location along a road. If a sufficient number of incidents occur, this suggests that drivers are taking evasive manoeuvres, such as hard braking, to avoid having accidents. This location data, suggesting potential accident black spots, is can then used to improve driver safety. For example, at step 78 this data can be used by road planners to identify stretches of road where modifications to the road could improve driver safety.

Alternatively, or in addition, as indicated at step 80, the location of a potential accident black spot can be fed back to vehicle drivers. In the case of the vehicle 10, the second receiving aerial 30 in data measuring device 12 can receive a GPRS, or other data, signal as the vehicle approaches a section of road where a high occurrence of incident reports has occurred indicating a potential accident black spot. This third data can be used to alert the driver to this danger to encourage them to drive more cautiously. As previously mentioned the transfer of data between the server and apparatus in the vehicle occurs using IP packet switching data transfer.

The transmission of this data can be time dependent, where sufficient data of incidents has been reported to indicate a statistically significant time element to the incident reporting. For example, a section of road may be more dangerous at a particular time of day as a result of parked cars. Such cars are typically present first thing in the morning, after normal working hours and at the weekend. As a result, the vast majority of reported incidents may occur when cars are parked along this road and therefore a driver need only be informed of the potential hazard ahead outside of weekday normal working hours.

The apparatus of the present invention also has the potential for real-time reporting and driver feedback. For example, if a vehicle is travelling along a road and a statistically significant number of vehicles ahead of the vehicle in question suddenly brake, this suggests a road hazard such as an accident. This incident can be instantly reported to vehicles travelling along the same road some distance behind the area where high levels of incidents were reported. Reporting potential black spots could occur through a device that does not also transmit data. A simple receiver and display device would be sufficient to indicate to a driver that he is approaching an incident black spot. The data gathering and transmitting aspects of the device need not necessarily be included in order to simply report to the driver.

The celeration and in particular speed data can be used to provide road traffic planners with information about traffic flows at different times of day. This data can be used to identify under utilised routes or set certain roads to prioritise traffic in one direction at certain times of day and change that priority direction later in the day. For example, in a town or city various routes are available to access the town / city centre. Many routes are under utilised and traffic flow on other routes could be reduced by directing traffic to certain routes and prioritising the flow on those route in a single direction, normally toward the town / city centre in the morning and away from the centre in the evening.

The apparatus set out above, with reference to Figures 1 and 2, is also used in a method of improving driver safety which will now be described with additional reference to Figures 4 and 5. From starting step 82, the data measuring device 12 measures GPS location and acceleration data at step 84. This data is categorised by processor 20 at step 86 and transferred to data storage and processing server 26 at step 88. Steps 84, 86 and 88 are equivalent to steps 50, 52 and 54 in Figure 3 thereby utilising the method set out in connection with Figure 4 and described in detail above. It is again noted that the minimum data measurement required is the data location which, if measured at insufficiently high frequency, can be used to derive all of the other information. However, there is significant processing efficiency in direct measurement of at least acceleration.

A further categorising step 90 is undertaken on the data processing and storage server 26 in which the transferred data is re-categorised on the basis of additional information provided in the form of driving condition data. This driving condition data is in a number of different forms reflecting different potential driving hazards and is either static data (that is data that changes very infrequently) or dynamic data (that is data that changes regularly during the course of a day, is time dependent and must be updated frequently during each day). An example of dynamic data is weather data, which can be gathered from a series of weather stations. The data from these weather stations is regularly updated and can be applied as a grid across the area through which the vehicle is travelling. This can either be done by forming a grid of regularly shaped polygons, each with a weather station within it, for example a grid of squares, and assuming that the weather is consistent within that square. Alternatively, a calculation can be made as to which is the nearest weather station to the vehicle at the time the data is transmitted and the assumption made that the weather at the vehicle is consistent with that at the weather station at the time the weather data was last gathered.

As a further alternative, or ideally to complement the weather station data, some weather data can be gathered from the vehicle. Many vehicles have external temperature sensors, as well as sensors that detect rain, used to automatically turn on windscreen wipers, and sensors to detect low levels of light. This data can be retrieved from the vehicle and sent with other data during the transfer of data to the server at step 8& Further sensors for detecting other weather conditions could also be added to the vehicle.

In step 90 the weather data can be used in conjunction with the transferred data to re-examine the already categorised data. For example, a vehicle may be travelling at 50 mph in an area that has a 50 mph speed limit and under normal driving conditions this is likely to be regarded as safe or appropriate driving. However, if the weather data indicates thick fog, heavy rain or very low temperatures potentially leading to ice on the road, this speed may be inappropriate and dangerous. As a result, this data, which was transferred as an event data report (therefore not indicating dangerous driving), may be upgraded to an incident due to the weather creating less safe driving conditions.

It is not only velocity data that is re-evaluated in the light of the weather data. Similarly, data identifying excessive acceleration, whether indicating hard braking or swerving, is inappropriate in wet or icy conditions. As a result, the angle of turn threshold °fhresh reduced below that which would trigger, at 72, a data report being sent at step 66. With event data report being sent on a regular basis, for example at least every mile or every minute (steps 70 and 68) and the data including the maximum angle of turn 6 or velocity v that has been reached since the last transfer, it is possible to see the most significant turn and speed that occurred within the last short period of time or distance.

Examples of static data include the locations of speed and safety cameras. In these cases, the re-evaluation of the data is to check for examples of excessive deceleration when approaching a safety camera and acceleration just after the camera. The triggers for unacceptable braking and acceleration in the vicinity of a speed or safety camera are Cower than for normal driving. In the vicinity of a camera these acceleration and breaking events are an indicator that a driver is possibly too close to a vehicle in front which is braking quickly on seeing a speed camera, or not paying sufficient attention to his own speed relative to that of the speed limit and assuming that he is over the speed limit and therefore braking sharply to avoid receiving a speeding fine. Another example of static data is sunrise and sunset data which, potentially in combination with ambient light levels measured on a vehicle, give an indication of whether the driver is adjusting his driving technique at dusk and at night.

A further example of driving condition data, which can be used in combination with the invention described in connection with Figures 3 and 4, is where the driver is being provided with instant feedback information relating to an accident black-spot. If an accident black-spot is indicated to the driver, it would be expected that the driver would adjust his driving behaviour appropriately. As a result, the threshoFd for an incident event for speed, celeration and angle of turn would be reduced dependent upon the severity of the accident black spot.

The re-categorised data can be transmitted back to the vehicle and instantly displayed to the driver, for example as red for very dangerous driving, amber for dangerous driving and green for acceptable driving behaviour.

However, in order to prevent the driver from being distracted whilst driving, an alternative way of alerting the driver is to send a message to a user registered to the vehicle when the journey has ended. As a result, at step 92 a check is undertaken to determine whether the engine of the vehicle has been switched off. If it has not, then the loop of data gathering continues again from step 84. However, if at step 92 the engine has stopped, data indicating this is transferred to the data server. Once this message is received at step 94, the data server checks to see whether any events were either categorised as incident events prior to transfer, or re-categorised as incident events in combination with driving condition data, to indicate that a potentially dangerous driving event has occurred. If no such dangerous driving event has occurred, then no further action is taken and the cycle ends at step 96.

However, if a dangerous driving event did occur, then a message, typically sent as an SMS. message and/or email, is sent to a person who is registered to that vehicle. The message indicates that a potentially dangerous driving event occurred in the last journey and requests that the registered person reviews the driving events in the last journey by visiting a website connected to the data storage and processing server 26, at step 98.

After a period of time, for example twelve hours, it is determined at step 100 whether the events have been reviewed. If the events have been reviewed then the cycle ends at step 102. However, if no review has taken place then a further message is sent at step 98. Typically messages would not be sent during the night. It should be noted that driving incident summary data is also sent daily to the company supervisor, fleet manager, HR Department etc. The review of dangerous driving events can require that the person reviewing their data acknowledges the driving as inappropriate or provides a reason. For example, if a vehicle unexpectedly pulls out of a side road it is necessary to brake very hard which will result in an incident being logged. However, such an incident cannot be anticipated and therefore does not reflect badly on the driver's performance. Additionally, should the driving be so extreme as to indicate a potential serious breach of safety or an extremely high level of risk, the driver will be invited to undertake an On-Line' driving assessment which will make them more aware of the mistakes which they have made and will make recommendations as to how they can improve their driving. A result of taking the on-line test will enable them to receive bonus points which will help with their standing in the driver behavioural database which is a list or table comparing the ratings of drivers within a group, for example within a company. These courses are run through the web site and drivers are invited to complete the relevant driving module (identified by the system) for the problems they are encountering. For example, a driver who uses excessive corner speed and braking will be invited to undertake this particular interactive module. The points earned by undertaking the module are then added back onto the driving score. It should be noted that the driving behaviour modules may be taken from the STEC Driving scheme and form part of the BTEC Diploma in Driving Science (an industry recognised accreditation).

The apparatus set out above, with reference to Figures 1 and 2, is further used in a method of analysing a road vehicle accident event which will now be described with additional reference to Figure 6. The only modification to the data measuring device 12 in this instance is that a plurality of second accelerometers are included within the data measuring device. It should be noted that these additional accelerometers are not essential to the operation of the data gathering device in the context of this invention, but provide significant advantages. The first accelerometers, as used in the methods described in Figures 3, 4 and 5, measures acceleration between zero and 20, which is comfortably within all normal manoeuvring of normal vehicles (it should be noted that air bags deploy at I.50). The additional accelerometers are able to measure acceleration and deceleration up to 20G which would only normally be experienced in crash situations.

These secondary accelerometers produce data that is recorded at a rate of 10Hz.

Following the start of a journey at step 104, acceleration data is measured at step 106 by the secondary accelerometer and continually recorded to a memory within processor 20 at step 108. At least two minutes of data is recorded and for each datum recorded, a test to determine whether the acceleration of the vehicle has exceed 1.50 (14.7ms2). The main criterion for determining that a crash has occurred is if the vehicle decelerates at 1.50, which is acceleration in the direction of travel of the vehicle of -1.5G. This would indicate a head-on collision of the vehicle either with a stationary object or another vehicle or a high speed glancing accident can produce accelerometer readings in excess of 1.5 0.

Such an impact of 1.50 is sufficient to deploy an airbag. Other accelerations, with other acceleration thresholds, indicate other crash conditions such as side impact on the vehicle or the vehicle being hit from behind.

If the deceleration exceeded 1.50, or there is any other indication of a crash or impact, the last two minutes of 10Hz accelerometer data is retained in the memory at step 112.

The acceleration data continues to be measured and recorded at step 114 until the vehicle has stopped moving for a predetermined period of time as tested at step 116.

When it is determined that the vehicle has stopped moving, the two minutes of retained data plus any further recorded acceleration data is all transferred, as previously described, to data processing and storage server 26, thereby ending the cycle at step 120.

Once the data has been received, the data is safely stored and archived and an immediate assessment of the crash can be made. This data can be used later to determine the severity of the accident and the likelihood of injuries such as whiplash being suffered by people in the vehicle or another vehicle involved in the accident. Criteria can be immediately applied to the data to determine the likely severity of the accident by for example looking at maximum deceleration or identifying a signature consistent with a certain type of crash. In the event of an accident that appears to be very serious, the emergency services can be immediately contacted and despatched to the GPS identified location of the accident.

In addition to the acceleration data being transferred at step 118, other data may also be transferred that may assist in identifying the circumstances that led to the accident.

Furthermore, the driver's data which will already have been being transferred to server 26 as part of the methods described in connection with the methods set out in Figures 3 and 5, can be used to put together a picture of the journey prior to the accident and the driving behaviour of the driver.

it will be appreciated by person skilled in the art that the above embodiments have been described by way of example only and not in any limitative sense, and that the various alterations and modifications are possible without departure from the scope of the invention as defined by the appended claims.

For example, the acceleration data can be gathered by interpretation of GPS data. If GPS data is gathered at a sufficiently high rate, acceleration as well as velocity data can be derived from the position and time data. However, the use of accelerometers provides more accurate data. It should be noted that the means for determining the vehicle location, described above as GPS receiver 14, could be any suitable device for determining the location of the vehicle including, but not limited to mobile phone or elI location, differential, LFDF, RFDF, Loran and Decca. It should also be noted that, in connection with the method set out in Figure 5, the step of re-categorising the data could take place before any data is sent simply by changing the threshold levels for incident data. For example, if data is being read on the vehicle such as temperature data or whether it is raining, or data is being sent back to the vehicle, such an accident black-spot data, the threshold levels can be reduced. So that if it is raining the maximum acceptable speed for driving can be reduced, either to an absolute value or be a percentage of the road speed limit. In a further example, where data is not gathered on the vehicle, the driving condition data can be transferred to the data measuring device 12 to be analysed by processor 20 this would then allow instant and direct feedback was given to the driver.

In this instance the data does not need to be transferred back to the data storage server, although in the exemplary embodiments described above this is done in order to record and retain the driver behaviour data. It is also the case that the above described apparatus can be used to assist insurance companies in assessing drivers for underwriting a policy for a driver or to modify their insurance premium in response to their driving behaviour. Drivers who show good driving behaviour, even after reassessment of their driving can be rewarded with reduced premiums.

Claims (18)

1. Claims 1. A method of analysing a road vehicle accident event, comprising the steps of:-S measuring vehicle data including at least vehicle acceleration data; recording at least some of said data; analysing said vehicle data to detect a crash event; and if a crash event is detected, sending at least some of said recorded data to a data storage device separate from the vehicle.
2. 2. A method according to claim 1, wherein said acceleration data is measured by a pluralityofaccelerometers.
3. 3. A method according to claim 2, wherein a plurality of said accelerometers operate in the same direction.
4. 4. A method according to claim 2 or 3, wherein said accelerometers measure acceleration at different sampling frequencies.
5. 5. A method according to any of the preceding claims, wherein the vehicle data that is analysed to detect a crash event includes said vehicle acceleration data.
6. 6. A method according to claim 5, wherein a crash event is detected if the acceleration is less than -1.5 0 (147ms2).
7. 7. A method according to any of the preceding claims, wherein at least some vehicle data is sent to a data storage device prior to the detection of a crash event.
8. 8. A method according to any of the preceding claims, wherein, if a crash event is detected, substantially all of said measured data over a predetermined period of time prior to and during the crash is sent to said data storage device.
9. 9. A method of analysing a road vehicle accident event substantially as hereinbefore described with reference to the accompanying drawings.
10. 10. An apparatus for analysing a road vehicle accident event, comprising the steps of:-data measuring means for measuring vehicle data including at least vehicle acceleration data; data recording means for recording at least some of said data; processing means for analysing said vehicle data to detect a crash event; and data transfer means for sending at least some of said recorded data to a data storage device separate from the vehicle in the event that a crash event is detected.
11. 11. An apparatus according to claim 10, wherein said data measuring means comprise a plurality of accelerometers.
12. 12. An apparatus according to claim 11, wherein a plurality of said accelerometers operate in the same direction.
13. 13. An apparatus according to claim 11 or 12, wherein said accelerometers measure acceleration at different sampling frequencies.
14. 14. An apparatus according to any of claims 10 to 13, wherein said processor analyses said acceleration data to detect a crash event.
15. 15. An apparatus according to claim 14, wherein a crash event is detected if the acceleration is less than -1.5 G (14.7ms).
16. 16. An apparatus according to any of claims 10 to 15, wherein said data transfer means sends at Least some vehicle data to a data storage device prior to the detection of a crash event.
17. 17. An apparatus according to any of claims 10 to 15, wherein, if a crash event is detected, said data transfer means transfers substantially all of said measured data over a predetermined period of time prior to and during the crash is sent to said data storage device.
18. 18. An apparatus of analysing a road vehicle accident event substantially as hereinbefore described with reference to the accompanying drawings.