Field of the invention

The present invention relates to a method and a system for assessing a driver's acceleration behaviour according to the preambles of the independent claims.

Background to the invention

Defensive driving is often synonymous with fuel-efficient driving. It also

contributes to maintaining good traffic flow, greater comfort for any passengers carried and safe carriage of any load carried. What directly contributes to both comfort and safety is how the person driving the vehicle accelerates. The acceleration may be both longitudinal, i.e. in the vehicle's direction of movement, and lateral, i.e. in the vehicle's sideways direction. The vehicle's acceleration may of course also be a combination of longitudinal and lateral acceleration.

High lateral accelerations may not only cause passengers discomfort and risk of falling but also present a safety risk due to uncontrolled movement of unsecured objects. In a truck context, high lateral accelerations may result in cargo shifting or tilting and thereby sustaining damage. In the worst case, high lateral acceleration may also cause a traffic accident. Low lateral acceleration also helps to reduce tyre wear.

US 2008/0015754 describes a system adapted to automatically compensating for sudden changes in a vehicle's lateral acceleration. The lateral acceleration is estimated while the vehicle is in motion, and if it exceeds a predetermined value a control signal adapted to countering the adverse sideways force is generated.

US 2009/0248240 describes a system which measures a vehicle's lateral acceleration and passes the information on to the driver. He/she may then decide whether to reduce, maintain or increase the vehicle's speed in order to be able as soon as possible to drive it in a controlled way. The system is for example used to train racing drivers. The information may also be saved for subsequent analysis.

US2009/0319129 describes a method for assisting a driver during driving. The method calculates the vehicle's maximum permissible lateral acceleration and recommends a longitudinal speed to the driver on the basis of that value.

The object of the invention is to improve a driver's acceleration behaviour and in particular to encourage him/her to improve his/her acceleration behaviour.

Summary of the invention

The object described above is achieved by an assessment method for assessing acceleration behaviour of a driver of a vehicle. The method comprises

determining the vehicle's acceleration a, calculating one or more threshold values for the vehicle's acceleration which depend on at least one situation-specific parameter, comparing the vehicle's acceleration a with said one or more threshold values, and generating on the basis of the comparison a grading signal related to said calculated one or more threshold values. According to another aspect, the object is achieved by an assessment system for assessing acceleration behaviour of a driver of a vehicle. The system comprises at least one sensor unit adapted to delivering an acceleration signal which indicates the vehicle's acceleration a, and a calculation unit adapted to receiving said acceleration signal, calculating one or more threshold values for the vehicle's acceleration which depend on at least one situation-specific parameter, comparing the vehicle's acceleration a with said one or more threshold values, and generating on the basis of the comparison a grading signal related to said calculated one or more threshold values. The system further comprises an output unit adapted to presenting information from the grading signal to the driver. The grading signal may according to an embodiment be sent to an external unit for presentation of information to, for example, the vehicle owner.

The method and the system provide the driver in a constructive way with encouragement to adapt the vehicle's speed in due time to bends or roundabouts ahead. Comfort for any passengers carried may then be improved while at the same time the risk of damage to any cargo may be reduced. The method and the system also help to reduce wear on tyres and other vehicle components, which may result in less time being needed for servicing and hence in more time during which the vehicle can be used. Reducing wear also reduces environmental impact in that vehicle parts need replacing less frequently.

The driver's acceleration behaviour is assessed by being graded on a scale which depends on the situation in which the vehicle is. The vehicle's situation is therefore observed and the driver's acceleration behaviour in response to it is assessed.

The vehicle's acceleration may be divided into lateral acceleration and longitudinal acceleration. In the case of lateral acceleration, the problem comprises according to an embodiment two cases, one of them being a temporary "spike" in lateral acceleration due to a rapid steering wheel movement on a bend, also called "jerk", and the other being continuously high lateral acceleration due to taking a bend at too high a speed. In the case of longitudinal acceleration, the problem according to an embodiment is too high acceleration related to a certain speed.

Assessment of the driver's acceleration behaviour may be part of a set of categories for assessing his/her driving behaviour.

Preferred embodiments are described in the dependent claims and the detailed description. Brief description of the attached drawings

The invention is described below with reference to the attached drawings, in which:

Figure 1 depicts an assessment system according to an embodiment of the invention.

Figure 2 is a flowchart for the assessment method according to an embodiment of the invention.

Figure 3 is a flowchart for the assessment method according to another embodiment of the invention.

Figure 4 is a diagram illustrating an example of how a rapid steering wheel movement on a bend may affect the vehicle's lateral acceleration (curve A), and what the lateral acceleration may be like in cases where the vehicle takes the bend at too high a speed (curve B).

Figure 5 illustrates in diagram form an example of the vehicle's maximum and mean lateral acceleration on a bend.

Figure 6 is a flowchart for the assessment method according to a further embodiment of the invention.

Figure 7 illustrates how the threshold values for assessment of the vehicle's longitudinal acceleration are arrived at according to an embodiment of the invention.

Figure 8 depicts examples of permissible reference acceleration values for assessing the driver's acceleration behaviour in the case of longitudinal acceleration. Detailed description of preferred embodiments of the invention

Figure 1 depicts an assessment system according to the invention which will now be explained with reference to the invention. The system comprises at least one sensor unit 1 ... n adapted to delivering an acceleration signal which indicates the vehicle's acceleration a. The sensor unit may for example be an accelerometer which measures the vehicle's acceleration, or a unit which measures the difference in wheel speed between the sides of the vehicle and then calculates the vehicle's lateral acceleration. The system further comprises a calculation unit adapted to receiving said acceleration signal. The calculation unit is further adapted to calculating one or more threshold values for the vehicle's acceleration which depend on at least situation-specific parameter, comparing the vehicle's acceleration with said one or more threshold values, and generating a grading signal based on the comparison. The grading signal is related to said calculated one or more threshold values.

The system comprises according to an embodiment an output unit adapted to presenting the driver with information from the grading signal. The grading signal may according to another embodiment be sent via a telematic unit in the vehicle to an external unit for presentation of the information from the grading signal to, for example, the vehicle owner. The driver may then subsequently be provided with information about his/her acceleration behaviour, e.g. via a report. The calculation unit comprises preferably a processor unit adapted to doing calculations, comparisons etc., and one or more memories. The output unit may for example be a display in the instrument panel in the vehicle, or an audio unit which conveys hints, feedback and/or grading in the form of voice messages. Figure 2 is a flowchart for the method according to an embodiment of the invention and comprises a first step S21 which determines the vehicle's acceleration a. A second step S22 calculates one or more threshold values for the vehicle's acceleration which depend on at least one situation-specific parameter. The threshold values preferably indicate limits for what is regarded as safe and comfortable cornering or safe and comfortable longitudinal acceleration. According to an embodiment, a situation-specific parameter is any from among the vehicle's speed, the vehicle's acceleration, the radius of curvature of the road or the arc angle of the road. The vehicle's speed is a parameter which is usually accessible via the vehicle's internal communication network, and may be received in the calculation unit in the form of a speed signal. The radius of curvature of a road may be determined by the calculation unit having access to map data concerning the vehicle's route and information about the vehicle's location, which may be obtained via a positioning unit in the vehicle, e.g. via GPS (global positioning system). This makes it possible to ascertain the radius of curvature in advance. Alternatively, the radius of curvature may be determined while the vehicle is negotiating, or after it has negotiated, the bend from, for example, the alignments of the wheels during the bend. The arc angle or edge angle of the road may for example be determined by measuring the vehicle's amount of time spent or distance travelled on a bend, i.e. the time or distance over which the vehicle's sensor unit 1 . n delivers an acceleration signal which indicates lateral acceleration. If the lateral acceleration is zero, the vehicle is travelling straight ahead.

Said threshold value or values are therefore calculated on the basis of the situation in which the vehicle is. The way they change depends on one or more situation-specific parameters, e.g. the threshold values at a vehicle speed of 50 km/h may be different from those at 70 km/h.

A step S23 then compares the vehicle's acceleration a with said one or more threshold values, followed by a step S24 which generates a grading signal based on the comparison and therefore related to said calculated one or more threshold values. The grading signal therefore adapts itself to the threshold values determined for the specific situation and hence to the scopes which follow from that situation.

According to an embodiment, the information from the grading signal is presented to the driver via, for example, an output unit. The grading signal may instead or in addition be sent to an external unit for presentation of information to, for example, the vehicle owner. The information presented may for example be in the form of a numeral on a scale from 1 to 10, in which 1 is not good acceleration behaviour and 10 is very good acceleration behaviour for the situation. Another example of how the grading may be expressed is in the form of a number of symbols, e.g. stars, in which case the number of symbols which light up depends on the grading signal. If only one symbol lights up, this indicates less good acceleration behaviour, and if all of the symbols light up this indicates good acceleration behaviour. The driver thus receives continuous feedback about how well he/she accelerated in the situation and is motivated to improve his/her results. According to an embodiment, the information in the grading signals is accumulated and saved in a memory in, for example, the calculation unit to allow it to be analysed subsequently. The information may also be sent to a central unit situated externally from the vehicle for further analysis and follow-up. According to an embodiment, the method comprises, at the time when a grading signal is generated, feeding back to the driver via an output unit how he/she can improve his/her acceleration behaviour, or praising his/her acceleration behaviour. The calculation unit is then adapted to, for example, generating via the output unit a text message related to the information from the grading signal. In the event of a low grading the driver may be urged to "Reduce speed before the bend", or in the event of a high grading he/she may be congratulated by, for example, "Good cornering!". The calculation unit according to an embodiment is adapted to taking account of whether the driver brakes or accelerates on a bend, accelerates or brakes before the bend etc., in cases where there is to be feedback to the driver.

According to an embodiment, said acceleration is the vehicle's lateral acceleration aiat- Lateral acceleration occurs mainly when the vehicle is cornering. The problem with regard to lateral acceleration comprises mainly two cases, 1 and 2. Case 1 is a temporary "spike" in lateral acceleration caused by a rapid steering wheel movement. This may for example be due to the driver not having realised that a bend was coming and consequently having had to turn the steering wheel quickly to avoid running off the road or into an incorrect lane. Case 2 is continuously high lateral acceleration due to taking a bend at too high a speed.

Assessment of cornering involves according to an embodiment the fulfilment of at least one situation criterion. If at least one situation criterion is fulfilled, the calculation unit is then adapted to allowing the vehicle's lateral acceleration to be assessed. According to an embodiment, said situation criterion is any from among a lowest vehicle speed, a lowest acceleration, a largest radius of curvature or a smallest arc angle. It is thus possible to ensure that very slight bends are not assessed and also that the driver is driving in normal running conditions. Lowest vehicle speed entails the vehicle travelling at above a certain speed, e.g. 20 km/h, before and after the bend, for the driver's acceleration behaviour on the bend to have to be assessed. Largest radius of curvature entails according to an embodiment the radius being below a value which depends on the vehicle's speed before the bend. Suitable values appear for example in the Swedish National Road Administration's highway construction recommendations (VV publication 2004:80), which state limits so that the radius of curvature is below what is classified as "less good" to "good". Smallest arc angle entails according to an embodiment the arc angle exceeding a value which depends on the vehicle's speed. For example, high values are set for the arc angle at low speeds (about 70 degrees) in order to assess only bends and no slight curves. At high speeds this value may be very low or even zero. This limitation is therefore used mostly as a filter in order to disregard slight curves in urban surroundings.

We now go on to explain how case 1 and case 2 described above may be detected and assessed. Case 1

According to an embodiment, the calculation unit is adapted to determining the absolute amount for a _\at during a predetermined period of time t _maXj erk and comparing this value with a threshold value y2, and generating a grading signal based on the comparison. Figure 3 illustrates a method for deciding whether a driver makes a rapid steering wheel movement, and curve A in the diagram in Figure 4 illustrates how a rapid steering wheel movement on a bend may affect the vehicle's lateral acceleration. Curve B illustrates the possible pattern of lateral acceleration when the speed at which the vehicle takes the bend is too high.

According to Figure 3, the absolute amount for is compared with the

acceleration threshold value yi at step S31 , and if |ai _a t| is greater than yi the driver's acceleration behaviour has to be assessed. The situation criterion here is therefore that a lowest acceleration has been reached. The threshold value yi (and also y ₂ ) may here depend on, for example, vehicle speed, radius of curvature, arc angle etc. If for example the vehicle is travelling at high speed, yi is a lower value than when it is travelling at a lower speed. If |ai _a t| is greater than yi, the method moves on to step S32 and |ai _a t| is then compared with a further acceleration threshold value y2. The calculation unit is adapted to measuring the period of time during which |ai _a t| exceeds y2, a period here referred to as t _je rk- If |aiat| is greater than y ₂ and t _jer k is shorter than a predetermined period t _maX jerk, a grading signal which indicates a low grading is generated at step S33 on the basis of the comparison. The driver is then regarded as having for some reason made a large turn, i.e. a large steering wheel movement. For example, at step S34 a hint will be displayed to the driver via the output unit to encourage him/her to take the next bend more gently. If either or both of the conditions at step S32 are not fulfilled, the driver's acceleration behaviour is assessed according to the assessment criteria for said case 2 as regards lateral acceleration, as indicated by step S35. The result is a way of detecting whether the driver makes a rapid steering wheel movement, and of assessing his/her subsequent acceleration behaviour.

According to another embodiment, the calculation unit is adapted to calculating the derivative da _lat /dt for the vehicle's lateral acceleration ai _at and determining the absolute value for the derivative, comparing this value with one or more threshold values and generating a grading signal based on the comparison. A high value of the derivative thus means a large jerk. The result is another way of detecting whether the driver makes a rapid steering wheel movement, and of assessing his/her subsequent acceleration behaviour.

According to a further embodiment, the calculation unit is adapted to determining the absolute amount for the vehicle's maximum lateral acceleration ai _at _ _m ax and mean lateral acceleration ai _at _ _m e _a n during a period of time t and determining whether the difference between |ai _at _ _max | and |ai _a t_ _mean | is greater than a threshold value, and generating a grading signal based on the comparison. This embodiment is illustrated in the diagram in Figure 5, which shows the absolute amount of the vehicle's maximum and mean lateral acceleration during a period from ti to t ₂ . This difference between |ai _a t_max| and |ai _a t_mean | is illustrated as adiff, and if it is greater than a threshold value a grading signal which indicates a jerky bend is generated. The result is a further way of detecting whether the driver makes a rapid steering wheel movement, and of assessing his/her subsequent acceleration behaviour.

High vehicle speed results in less tolerance for jerking and acceleration, so the threshold values in the above situations depend preferably on the situation- specific parameter of vehicle speed.

Case 2

To assess whether the driver has driven too fast on a bend, the calculation unit according to an embodiment is adapted to determining the vehicle's maximum lateral acceleration ai _a t_max and comparing it with two threshold values a _t hi and a _t h2 for the vehicle's maximum lateral acceleration, and generating a grading signal based on the comparison. The comparison is done according to an embodiment by using the formula:

8 = •10, g e [0,...,10] (1 ) atk2 ^a tkl

The grading signal then indicates g, which in this example is a value between 0 and 10. 10 is the maximum score and 0 the lowest possible. The flowchart in Figure 6 illustrates a method for assessing whether the driver drove too fast on a bend. According to an embodiment, a first step S61 checks whether the vehicle is cornering, e.g. if its acceleration is greater than 0. If such is the case, this is followed according to an embodiment by checking at step S62 that the bend is significant enough, e.g. long and/or sharp enough. If such is the case, an assessment may be done and the vehicle's maximum lateral acceleration ai _at _ _m ax be determined. Step S63 compares ai _at _ _max with the threshold values according to formula (1 ), and at step S64 a grading signal is generated. The information in the grading signal is then presented at step S65 via the output unit. In the event of a low grading, there is also the possibility of the driver being given a hint via the output unit to encourage him/her to reduce speed before the bend next time. According to an embodiment, the threshold values a _th i and a _t h2 are the same as the threshold values yi and y2 in Figure 4.

However, assessment according to case 2 is not necessarily tied to the vehicle's maximum acceleration. Instead of using the vehicle's maximum lateral acceleration ai _a t_max as in the above embodiment, what it used instead according to an embodiment is the vehicle's mean lateral acceleration ai _a t_ _m ean through the bend. In this case the calculation unit is adapted to determining the vehicle's mean lateral acceleration ai _a t_ _m ean and comparing it with two threshold values a and a _t h4 for the vehicle's mean lateral acceleration, and to generating a grading signal based on the comparison. The flowchart in Figure 6 is therefore the same for this embodiment except that step S63 and step S64 use instead the vehicle's mean lateral acceleration, as also indicated in the flowchart.

According to an embodiment, the road ahead is known, in which case the calculation unit may be adapted to generating hints to the driver which are displayed via the input unit if he/she is about to enter a curve at too high a speed in reference to highest score.

On slippery road surfaces the vehicle may lose grip in the longitudinal direction. According to an embodiment the system takes account of factors such as slippery road surface or rain and adapts the threshold values accordingly. Hints may then also be given to the driver before or after the bend.

Various hints and forms of feedback may be given to the driver via the input unit. There follow a number of examples of conceivable feedback to the driver in an identified situation. Where a situation with very high lateral acceleration is detected, e.g. if g=1 according to formula (1 ), the driver is urged to maintain a lower cornering speed. In the case of medium to high lateral acceleration, e.g. g=5 according to formula (1 ), if the driver braked the vehicle on the bend, he/she is urged to maintain a lower cornering speed.

In the case of low lateral acceleration, e.g. g=9 according to formula (1 ), if the driver released the accelerator pedal in good time before the bend and no braking occurred on the bend, he/she is praised for good driving.

In a situation with high lateral acceleration where the ratio between maximum lateral acceleration and mean lateral acceleration exceeds a threshold value, the driver is urged to avoid "jerking" during a bend.

According to an embodiment, said acceleration is the vehicle's longitudinal acceleration ai _ong - This makes it possible to assess how well the driver accelerates in the longitudinal direction.

Assessment of acceleration in the longitudinal direction requires according to an embodiment the fulfilment of at least one situation criterion. If at least one situation criterion is fulfilled, the calculation unit is then adapted to allowing the vehicle's longitudinal acceleration to be assessed. According to an embodiment, said situation criterion is either a longitudinal acceleration ai _ong greater than a predetermined threshold value or a vehicle speed increase by more than a defined number of km/h. This makes it possible to ensure that very small changes in the vehicle's acceleration are not assessed, and also that the driver is driving in normal running conditions.

As in case 2 above, it is also possible to assess the acceleration in the

longitudinal direction, referred to here as case 3.

Case 3

Assessment here is based on how the driver accelerates on the basis of a permissible reference acceleration which depends on the vehicle's speed. The higher the speed maintained by the vehicle, the less the driver is expected to need to accelerate. According to an embodiment, the calculation unit is adapted to determining reference values a _ma x_ion _g i and a _ma x_iong2 based on threshold values for the vehicle's permissible instantaneous acceleration ai _ong at current vehicle speed and comparing the vehicle's instantaneous acceleration ai _ong with a _ma x_ion _g i and a _m ax_iong2, and generating a grading signal based on the comparison.

The comparison is done according to an embodiment with the formula:

^long ^max_ longl

8 10, g e [θ,.,.ΐθ], ^ _h . > a _{max long2} , a _lons > 0 (2) ^max_ longl ^max_ longl The grading signal then indicates the assessment of g. Formula (2) is based on instantaneous acceleration or maximum acceleration. This may of course be converted to being calculated on mean acceleration over a certain period of time, e.g. the whole or parts of the acceleration process. The calculation unit is then adapted to determining reference values a _mea n_iongi and a _mea n_iong2 based on threshold values for the vehicle's permissible mean acceleration a _mean _iong at current mean speed and comparing its mean acceleration a _mean _iong with a _mea n_iongi and a _mean _iong2, and generating a grading signal based on the comparison.

Formula (2) may be used for comparison, but with changed parameters. Figure 7 is a diagram illustrating how the reference values a _max jongi and a _max jong2 and the reference values a _mean _iongi and a _mean _iong2 are calculated for the

instantaneous acceleration ai _ong - The y-axis denotes acceleration values, the x- axis speed values. The following are given: the vehicle's longitudinal acceleration aiong, current speed v _cur r and maximum speed v _max , a minimum acceleration a _m i _n and two threshold values a _max i and a _maX2 for the vehicle's permissible reference acceleration. From these it is possible to arrive at limits for the vehicle's

permissible longitudinal acceleration. Depending on the values of ai _ong and v _CU rr, the reference values are then arrived at. The diagram illustrates clearly that the higher the vehicle's speed, the smaller the permissible acceleration. Another embodiment uses a table of speed ranges and permissible acceleration levels to assess the driver's acceleration behaviour. An example of such a table appears in Figure 8. If for example the vehicle's speed changes from 10 km/h to 20 km/h, the permissible acceleration according to the table is X1 m/s ² . The calculation unit in this embodiment is adapted to monitoring the vehicle's speed and acceleration, and when an increase in its speed within any of the ranges in the diagram is observed, its acceleration is compared with the permissible acceleration in the table, and a grading signal is generated. If for example its acceleration is greater than that in the table, the grading signal will indicate a low score. According to an embodiment, the table values X1 -Xn are variable and depend for example on the vehicle's weight. For example, higher accelerations are permissible at low vehicle weight.

Said acceleration a comprises according to an embodiment both lateral and longitudinal acceleration, which makes it possible to assess both the lateral and the longitudinal acceleration as described above.

The invention relates also to a computer programme product which comprises computer programme instructions to enable a computer system in a vehicle to perform the steps according to the method when the computer programme instructions are run on said computer system. The invention comprises also a computer programme product which has the computer programme instructions stored on a medium which can be read by a computer system. The present invention is not limited to the embodiments described above. Various alternatives, modifications or equivalents may be used. The aforesaid

embodiments therefore do not limit the scope of the invention, which is defined by the attached claims.