WO2008062257A2

WO2008062257A2 - Driving safety logging system and method for vehicles

Info

Publication number: WO2008062257A2
Application number: PCT/IB2006/054401
Authority: WO
Inventors: Sreekanth Thumrugoti
Original assignee: Dcrypt Consultancy Services Limited; Abramowitz, Philip B.
Priority date: 2006-11-23
Filing date: 2006-11-23
Publication date: 2008-05-29
Also published as: WO2008062257A3

Abstract

A system (10) for logging instances in which a road vehicle is driven in a manner exceeding safe performance criteria includes a wheel vibratory sensor arrangement (14) for each wheel (12) configured to measure vibrational motion of the wheel hub along at least two, and preferably three, perpendicular axes. The system processes the vibrational motion measurements to determine whether the current use of the vehicle exceeds safe performance criteria for the vehicle under current driving conditions, and logs instances of such exceedance. The logged data may be used in applications such as calculation of an insurance premium adjustment or by law enforcement agencies for analysis of driving behavior prior to occurrence of a road accident.

Description

Driving Safety Logging System and Method for Vehicles

FIELD AND BACKGROUND OF THE INVENTION

The present invention relates to vehicle safety monitoring and, in particular, it concerns a system for logging instances in which a road vehicle is driven in a manner exceeding safe performance criteria.

Speed of vehicles is increasing along with development of automobile technology and speeding is one of the most prevalent factors contributing to traffic crashes. According to National Highway Traffic Safety Administration (USA) the economic cost to society of speeding related crashes is estimated to be $40.2 billion per year in USA alone. In the year 2004, speeding was identified as a contributory factor in 30% of all fatal crashes and 13,192 lives were lost. Speeding crashes occur when an automobile is driven too fast for the prevalent conditions, more so during winters and on wet/snowy/icy road conditions.

Rollovers are another area of concern, more so due to the rapid growth of sport-utility vehicles (SUVs) in the passenger vehicle segment. According to National Highway Traffic Safety Administration (USA) research note issued in November 2005, 10,553 fatalities occurred due to passenger vehicle rollovers during the year 2004 in USA alone. SUVs, a rapidly growing segment in passenger vehicles, have the highest rate of rollovers. In some of the vehicles, even when they are equipped with ABS and stability control systems, the percentage chance of rollover is as high as 17%.

Two technologies which have been widely adopted in an attempt to improve vehicle controllability are the "anti-locking braking system" ("ABS") and the "electronic stability control" ("ESC") system. ABS (anti-locking braking system) is a computerized system that keeps wheels from locking up during hard brake applications. ABS does not decrease or increase normal braking capability. It also does not necessarily shorten stopping distance, but it helps keeping vehicle under control during hard braking. ESC (Electronic stability control system) compares the driver's intended direction in steering and braking inputs, to the vehicle's response, via lateral acceleration, rotation (yaw) and individual wheel speeds. ESC then brakes individual front or rear wheels and/or reduces excess engine power as needed to help correct understeer (plowing) or oversteer (fishtailing). ESC also integrates all-speed traction control, which senses drive-wheel slip under acceleration and individually brakes the slipping wheel or wheels, and/or reduces excess engine power, until control is regained. ESC cannot override a car's physical limits. If a driver pushes the possibilities of the car's chassis and ESC too far, ESC cannot prevent a crash. It is a tool to help the driver maintain control.

ESC combines ABS, traction control and yaw control (yaw is spin around a vertical axis). To grasp how it works, think of steering a canoe. If you want the canoe to turn or rotate to the right, you plant the paddle in the water on the right to provide a braking moment on the right side. The canoe pivots or rotates to the right. ESC fundamentally does the same to assist the driver.

Numerous international studies have confirmed the effectiveness of ESC in helping the driver maintain control of the car, help save lives and reduce the severity of crashes. In the fall of 2004 in the U.S., the NHTSA confirmed the international studies, releasing results of a field study in the U.S. of ESC effectiveness. NHTSA concluded that ESC reduces crashes by 35%.

Despite ABS, ESC and traction control systems, an unacceptable number of speeding accidents occur as the drivers push the car too far beyond a threshold point. This threshold point is dynamic as it is defined by the automobile, weather and road conditions and very few experienced drivers can judge this accurately. It has been proposed to measure vibrations of wheels during operation of a vehicle in order to provide diagnostic information. For example, vibrations at the rate of rotation of the wheel may indicate an imbalance of the wheel. Vibrations at twice that frequency may indicate wear in the CV joint. Vibrations at other frequencies may indicate misfiring of an engine cylinder or the like. A sudden reduction in vibrations may indicate aquaplaning.

Although in principle, the idea of measuring vibrations is promising, the simplistic approach mentioned above is ineffective for deriving reliable information about the interaction of a vehicle with the road. Many different factors may cause variations in the frequency and amplitude of wheel vibrations, including variations in road surface, loading of the vehicle, shift of center of gravity during acceleration or braking. Simple one-dimensional vibration measurements are unable to distinguish between these different causes. In the field of vector calculus, various parameters may be used to quantify properties of a motion. Of particular relevance to the present invention are vector curvature and vector torsion. Expressed in terms of a point moving with velocity v and acceleration a in three dimensions, the vector curvature is defined by: v x a λ' =

and gives a measure of how sharply the path of the point turns in three dimensional space. The vector curvature becomes zero if the acceleration is along the line of the velocity. The vector torsion is defined by:

(y ^y~ a).a |α ^χ α'| where a ' is the time derivative of the acceleration. The vector torsion is a measure of the helical progression of the motion in three dimensions, and is zero for any path contained in a plane. Parenthetically, it will be noted that both the curvature and the torsion are scalar quantities. They are referred to herein as "vector curvature" and

"vector torsion" to identify them as the quantities of curvature and torsion defined in vector calculus, as distinct from other common uses of the terms "curvature" and "torsion". There is therefore a need for a system and method based on wheel vibration measurements which would reliably indicate and record instances of a road vehicle being driven in a manner exceeding safe performance criteria.

SUMMARY OF THE INVENTION

The present invention provides a system and method which measure and interpret wheel vibrations in two or three dimensions to identify when a road vehicle is driven in a manner exceeding safe performance criteria, i.e., when the vehicle is at risk of imminent loss of road holding capability. The system logs instances of such situations. In certain preferred embodiments, the logged data is then used by insurance companies to make "driving quality" adjustments to insurance premiums. In other preferred embodiments, the system may operate as a "black box" recorder conceptually similar to the "black box" used on aircraft, recording recent instances of unsafe vehicle operation for use by law enforcement authorities for determining culpability in case of an accident, or for assessing whether a driver took evasive action moments before an impact. Optionally, the system may also provide a warning to the driver when the vehicle is approaching or exceeds safe performance criteria for the current driving conditions, and may be further combined with an automated intervention system for reducing vehicle speed by selective individual wheel braking, thereby helping to ensure stability of an automobile at all times under all automobile, road and weather conditions. The system may further include a simulator/diagnostic program that assists manufacturers and testing agencies in enhancing automobile stability through better design and evaluating vehicle dynamics. The system may detect and store faults in tires, wheel alignment, castor camber, brakes or steering linkage at their onset, prompting service center personnel to take necessary action. Thus, according to the teachings of the present invention there is provided, a method for logging instances in which a road vehicle is driven in a manner exceeding safe performance criteria, the vehicle having at least four wheels, the method comprising: (a) repeatedly measuring wheel-hub vibrational motion at the hub of each of at least four of the wheels; (b) processing the vibrational motion measurements to determine whether the current use of the vehicle exceeds safe performance criteria for the vehicle under current driving conditions; and (c) storing data indicative at least of instances in which safe performance criteria have been exceeded, wherein the measuring and the processing measure and process the vibrational motion for each wheel along at least two perpendicular axes, and preferably along three perpendicular axes.

According to a further feature of the present invention, the processing includes: (a) processing the vibrational motion measurements along each of the at least two axes for each wheel to derive, substantially in real time, at least one motion-related parameter for each wheel; and (b) applying classification criteria to values of the at least one motion-related parameter for all of the wheels to identify whether the current use of the vehicle exceeds safe performance criteria for the vehicle under current driving conditions. According to a further feature of the present invention, the at least one motion- related parameter for each wheel includes a parameter which varies as a function of a balance between vertical vibrations and horizontal vibrations of the wheel hub.

According to a further feature of the present invention, the at least one motion- related parameter for each wheel includes a proportional deviation of the vector curvature of the wheel-hub vibrational motion.

According to a further feature of the present invention, the proportional change is evaluated relative to an average value of vector curvature.

According to a further feature of the present invention, the at least one motion- related parameter for each wheel includes a proportional deviation of the vector torsion of the wheel-hub vibrational motion.

According to a further feature of the present invention, the proportional change is evaluated relative to an average value of vector torsion.

According to a further feature of the present invention, the at least one motion- related parameter for each wheel includes: (a) a proportional deviation of the vector curvature of the wheel-hub vibrational motion; and (b) a proportional deviation of the vector torsion of the wheel-hub vibrational motion.

According to a further feature of the present invention, the proportional changes are evaluated relative to an average value for each of the vector curvature and the vector torsion.

According to a further feature of the present invention, the classification criteria include applying both an upper threshold and a lower threshold to the at least one motion-related parameter for each wheel.

According to a further feature of the present invention, the data is employed to determine a driving quality adjustment to an insurance premium.

According to a further feature of the present invention, the data is employed to derive information about how the vehicle was driven prior to a road accident.

There is also provided according to the teachings of the present invention, a system for logging instances in which a road vehicle is driven in a manner exceeding safe performance criteria, the vehicle having at least four wheels, the system comprising: (a) a wheel vibratory sensor arrangement associated with each of at least four of the wheels, the wheel vibratory sensor arrangement being configured to measure vibrational motion of the hub of each wheel along at least two perpendicular axes; (b) a processing system in data communication with each of the wheel vibratory sensor arrangements, the processing system being configured to process the vibrational motion measurements to determine whether the current use of the vehicle exceeds safe performance criteria for the vehicle under current driving conditions; and

(c) a data storage system associated with the processing system and configured to store data indicative at least of instances in which safe performance criteria have been exceeded.

According to a further feature of the present invention, the wheel vibratory motion sensor arrangement includes a tri-axial accelerometer associated with each of the at least four wheels and deployed for measuring vibrational motion of the hub of each wheel along three perpendicular axes. According to a further feature of the present invention, the data storage system is configured to store the data together with a time at which each instance of safe performance criteria exceedance occurs.

According to a further feature of the present invention, the data storage system is configured to store the data together with data relating to a duration during which safe performance criteria were exceeded.

According to a further feature of the present invention, the data storage system is configured to selectively retain data indicative of most recent instances in which safe performance criteria have been exceeded. According to a further feature of the present invention, there is also provided a vehicle disabling arrangement associated with the processing system and configured to disable operation of the vehicle if the data storage is prevented.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is herein described, by way of example only, with reference to the accompanying drawings, wherein:

FIG. 1 is a schematic representation of a system, constructed and operative according to the teachings of the present invention, for logging instances in which a road vehicle is driven in a manner exceeding safe performance criteria, the system being shown in the context of a plan view of a vehicle; FIG. 2 is a schematic side view of a wheel illustrating conventions used herein for describing axes of motion;

FIG. 3 is a schematic illustration of a part of a vibrational path of motion of a wheel hub as sampled by the system of the present invention;

FIG. 4 is a flow chart illustrating schematically the operation of an exemplary implementation of the present invention;

FIG. 5 is a table presenting a range of different safe-performance-criteria exceedance conditions which can be identified according to an exemplary implementation of the present invention; and FIG. 6 is a table illustrating a simple example of use of the logged data of the present invention to determine a driving quality adjustment to an insurance premium.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention is a system and method for logging instances in which a road vehicle is driven in a manner exceeding safe performance criteria.

The principles and operation of systems and methods according to the present invention may be better understood with reference to the drawings and the accompanying description.

Before referring to the drawings, by way of introduction, the underlying concept of the present invention relates to monitoring a balance between vibrational motion of the vehicle wheels in at least two, and preferably three, dimensions. Most, if not all, of the friction between a tire and a typical road has nothing to do with attractive intermolecular forces between the tire rubber and the road material. Instead, by a mechanism dubbed bulk friction, the rubber of a moving tire briefly presses down into small pits in the road surface. As those ups and downs of road topography jostle the tire, they cause vibrations. Reduction in road friction results in corresponding reduction in the vibrations.

Many factors influence the dynamics of wheel vibrations. These include mechanical factors such as tire loading, tire balance, defective or worn tire construction, bent wheels or axles, warped brake-drum or rotor, worn or loose universal joints, and worn shock absorbers. Other factors relate to the driving conditions, such as variations in the road surface, or shift in loading during acceleration, braking or cornering.

The suspension system of a road vehicle accommodates vertical vibrations, with a result that the amplitude of vertical vibrations tends to be much larger than horizontal (forward/back and sideways) vibrations. For this reason, horizontal vibrations are often dismissed as insignificant. To the best of the inventor's knowledge, three-dimensional wheel vibration measurements have previously been used only for assessing passenger comfort with the focus on ways and means to reduce vibrations and enhance passenger comfort.

In contrast to this approach, the present invention teaches that the measurement of wheel hub vibrations along at least two, and preferably three, axes, provides additional information vital for distinguishing between different scenarios during operation of the vehicle and to measure, indirectly, how the road holding capability of the vehicle is changing. Although the vibrations in the forward/back and sideways directions are of much smaller amplitude than those in the vertical direction, their velocity and acceleration are not necessarily smaller, and may often be of roughly the same order of magnitude as the vertical vibrations.

There follows an attempt to describe the balance between the vertical and horizontal vibrations of a vehicle wheel in simple intuitive terms. This explanation is presented as an aid to the reader and is believed to be helpful to an understanding of the operation of the present invention. However, it should be noted that the present invention as defined in the appended claims has been found experimentally to be effective, and that its patentability is not in any way conditional upon the accuracy of the theory set out below, which may well be overly simplistic.

When there is normal pressure on the wheels, vertical vibrations will be more pronounced than lateral. As the load increases, there will be much less leeway for the tires to undergo vertical vibrations. The heavier load thus reduces the vertical displacement length. It also varies the tire geometry to make the contact region larger. The extra force generated by the pits and projections of the road hitting the tire with more impact due to increase in the tire load results not only increased heat but also an increase in lateral vibrations. Hence we can conclude that load increases are followed by decrease in vertical vibrations and increase in horizontal vibrations. Vice versa is true on the wheels where there is a decrease in load.

During cornering, the load of the vehicle shifts onto the outside wheels, and load is reduced on the inside wheels. During braking, load moves onto the front wheels and is reduced on the rear wheels. Conversely, during acceleration, more load is transferred to the rear wheels. If follows that, by monitoring the varying balance of horizontal and vertical vibrations for each wheel, these different scenarios can be identified and monitored to ensure that they remain within safe limits.

When either the road or the tires become too smooth, the road holding capabilities will be significantly reduced. When vehicles travel on an icy road, ice melts and forms a watery barrier between wheels and road and stops pits and projections of the road from hitting tires. Hence the road friction is reduced. At the same time vibrations are also reduced. Hence, monitoring of the balance of vibrations can also provide an indication of the sufficiency of traction during driving. It follows that monitoring of the relative balance between horizontal and vertical vibrations for each of the wheels of a road vehicle can provide information regarding the current level of general traction, and about whether vehicle maneuvers are bringing the vehicle close to its limits of safe operation under current conditions. By dynamically calculating a threshold point which inherently takes into consideration automobile, road and weather conditions and by reducing speed to keep the vehicle under this point all the times, the present invention enables the average driver to drive a vehicle safely under almost any circumstances, without requiring exceptional driving skills. Additionally, certain variations of parameters for individual or groups of wheels not matching the aforementioned situations of acceleration, braking or cornering may be used to indicate various malfunctions or servicing requirements.

Turning now to the drawings, Figure 1 shows a schematic representation of a system, generally designated 10, constructed and operative according to the teachings of the present invention, for logging instances in which a road vehicle is driven in a manner exceeding safe performance criteria. Generally speaking, system 10 includes a wheel vibratory sensor arrangement 14 associated with each of at least four wheels 12. Each wheel vibratory sensor arrangement 14 is configured to measure vibrational motion of the hub of the corresponding wheel 12 along at least two, and preferably three, perpendicular axes. System 10 further includes a processing system 16, in data communication with each sensor arrangement 14, configured to process the vibrational motion measurements to determine whether the current use of the vehicle exceeds safe performance criteria for the vehicle under current driving conditions. A data storage system or "logging" system 18, associated with processing system 16, includes a data storage medium 20 and is configured to store data indicative at least of instances in which safe performance criteria have been exceeded. According to certain particularly preferred implementations, the present invention further includes a user display or warning module 22 configured to generate a visible and/or audible indication to the driver when safe performance criteria are being approached, or have been exceeded, thereby alerting the driver to take corrective action. At this point, it will be useful to define certain terminology as used in the description and claims. Firstly, with regard to the conventions used herein for defining directions of motion, reference is made to a set of orthogonal axes (x, y, z) defined individually for each wheel, wherein the x-axis corresponds to the "forward" direction of the wheel, the z-axis is upwards, and the y-axis completes the set of right-handed coordinates, pointing leftwards along the wheel axis of rotation. The "JC" and "_y" directions for two wheels are visible in Figure 1 while the "JC" and "z" directions are visible in Figure 2. Most preferably, these directions are chosen to be the axes of measurement of the vibratory sensor arrangement 14 associated with each wheel, although the preferred implementations of the data processing described below are essentially unchanged if the measurement axes are chosen differently.

The phrase "measuring wheel-hub vibrational motion" and similar phrases are used herein in the description and claims to refer to measurement of any one or more of displacement, velocity, acceleration and rate-of-change of acceleration. As is well known in the art, in principle, measurement of any one of the aforementioned parameters allows derivation of all of the other motion parameters listed. In fact, many commercially available accelerometers are configured or configurable to provide direct outputs indicative of most or all of the aforementioned parameters.

The terms "vibration", "vibrational motion" and "vibratory" are all used herein in the description and claims to refer to perturbations from a mean position after disregarding any uniform (mean) velocity component. In other words, all measurements for each wheel are taken in a frame of reference moving with that wheel and centered at the mean position of a reference point on the wheel hub at which the sensor(s) is attached. Turning now to the structural details of wheel vibratory sensor arrangements

14, these may be any type of motion sensor capable of measuring one or more of the parameters of vibrational motion of the wheels during operation of the vehicle. The most practical implementation currently known to the inventor is the use of one or more accelerometer rigidly mounted on the central hub of the wheel via a bracket which stabilizes the sensor in a constant orientation. Ideally, the accelerometers are mounted close to the center points of the wheel hubs, although this is not essential, and may not be possible due to structural considerations in many vehicles. In some cases, it is possible to achieve useful results by mounting the sensor on a non-rotating part of the wheel suspension adjacent to the wheel, thereby simplifying the mounting. While two or three separate uniaxial accelerometers may be used, the implementation is simplified by the use of a single triaxial sensor unit, i.e., which generates data regarding three orthogonal axes. Various types of accelerometer technology are suitable for implementing the present invention. By way of a non- limiting example, successful trials were performed using a miniature triaxial accelerometer based on piezoelectric sensor technology. Examples of suitable sensors are commercially available from Bruel & Kjser (Denmark) under the trademark DELTATRON ^R . Alternative implementation may use commercially available inertial sensor based on MEMS technology.

In the trials, a broad-band response accelerometer was used which was sensitive to vibrations in a frequency range from 0.3 Hz through to 3 kHz, although a significantly narrower range may be used. The frequency of data sampling is preferably at least 10 per second, and more preferably at least 100 data sampling points per second. A particularly preferred range is between about 200 to about 800 Hz sampling frequency, depending upon the technical capabilities of the sensor and of the data processing system. This high-frequency repeated sampling of data provides highly effective continuous real-time monitoring of the vehicle performance.

Optionally, if the quantity of data sampled is beyond the capabilities of processing system 16 for processing in real-time, data culling may be implemented. For example, according to one preferred option, since the negative parts of the x, y and z oscillations tend to closely mirror their corresponding positive vibration paths, the system may disregard sampled data with negative values of x, y and z, instead processing only the readings with positive "JC", positive "_y" and positive "z".

In order to normalize the data sampling, a predefined data capturing point ("DCP") defined within the cyclic motion may be used. For example, in trials conducted to-date, a DCP defined at 315° through the peak-to-peak oscillatory cycle along the major displacement axis (typically vertical "z") was used. This corresponds to 135° in the zero-to-peak cycle. The selection of the DCP, as in this example, preferably ensures that the displacement, velocity, acceleration and time-derivative of acceleration are all non-zero at the sampling point, at least along the primary displacement axis. Depending upon the sensor capabilities, synchronization of the DCP may be achieved automatically by appropriate configuration of the sensor, or via data pre-processing subsequent to sampling at a predetermined sampling frequency falling at arbitrary points in the vibration cycle. It should be noted that the present invention is applicable to a wide range of types of vehicle, ranging from small cars through to large trucks, and correspondingly having various different numbers of wheels. For four-wheeled vehicles, all four wheels are preferably provided with their own vibratory sensor arrangements 14. In the case of vehicles having more than four wheels, road-handling performance monitoring can typically be achieved by monitoring only four wheels, preferably those nearest the outer corners of the vehicle. Even in such cases, deployment of sensor arrangements on each wheel may be preferred in order to offer the additional functions of detecting wheel-related faults, such as wheel imbalances, worn tread or wheel misalignment. Data sampled by wheel vibratory sensor arrangements 14 is then passed to processing system 16. The data connection may be via hard-wired connection or via wireless telemetry (e.g., IR or RF communications links), depending upon practical engineering considerations and economic considerations for any given implementation.

Turning now to processing system 16 itself, this may be any suitable type of processing system including, but not limited to, a general-purpose processing system operating suitable software, purpose-built hardware, and any combination of hardware and software or "firmware". The use of onboard computer systems is already highly prevalent in automobile manufacture, and systems designed for implementing systems such as ESC may in some cases already have processing capacity sufficient for implementing the present invention.

Turning now to the processing architecture of processing system 16, which may be reflected either in hardware modules or in software modules, as is known in the art, this processing architecture may be implemented in numerous ways. Underlying the various options, is a common inventive concept, namely, that deteriorating driving conditions are identified by processing data from vibration measurements in at least two dimensions from each measured wheel. Three different approaches for implementing this common concept will now be addressed briefly. According to a first approach, the vibrational data for each measured wheel is translated to the center of gravity of the vehicle, and the data is then treated as a single space curve in a multidimensional vector space, for example of 13 dimensions (3 dimensions of displacement, velocity and acceleration for each of four wheels plus time). It has been found empirically that the vector curvature of this multidimensional space curve is indicative of a current driving state of the vehicle, thereby allowing preparation of a look-up table or algorithm to translate the derived curvature into a current driving state classification.

Although the aforementioned first approach is particularly elegant in that it performs a single calculation in the multidimensional space of all measurements, this approach has been found computationally intensive, and is therefore difficult to implement in real-time and with high data sampling rates using compact and economical processing components. As a result, it has been found preferable to perform the computational analysis of sampled data as a two stage process. First, the vibrational motion measurements along each of the measured axes are processed for each wheel individually so as to derive, substantially in real time, at least one motion- related parameter for each wheel. Then, classification criteria are applied to the motion-related parameters for all of the wheels to identify a current status of vehicle function. As a particularly simple implementation of this two-stage computational analysis, effective under a considerable range of circumstances, the vertical and horizontal vibration amplitudes may be compared against thresholds defined proportionally relative to the average amplitudes for each. This approach follows the intuitive explanation of the balance of vibrations as described above. Thus, during overly sharp high-speed cornering, thresholds for intervention might be set at a 30% decrease in vertical vibration amplitude relative to the mean value and a 60% increase in horizontal vibration amplitude on the outside wheels, and/or corresponding variations in the reverse direction on the inside wheels. The specific threshold values are empirically determined. While this amplitude-based approach ties in well with the proposed intuitive explanation of the balance of vibrations described above, it suffers from a number of shortcomings. Most notably, the reliance on amplitude alone inherently disregards a large proportion of the information available from the vibratory sensor arrangements. To avoid this shortcoming, certain most preferred implementations of the present invention employ one, or more preferably two, parameters which quantify certain properties of the instantaneous motion of a reference point on each measured wheel. In particular, two properties which have been found highly effective for reliably determining the current status of the interaction between a vehicle wheel and the underlying surface are the vector curvature and the vector torsion of the space curve currently traveled by a reference point on each wheel.

Referring briefly to Figure 3, this illustrates schematically an example of a possible path of motion of the reference point on one wheel during a fraction of a second as measured by the vibratory sensor arrangement. At each point along the path shown, the reference point has an instantaneous velocity v which is a tangent to the path. In the case of a circuitous path, the reference point also has an acceleration directed towards the center of the instantaneous curvature. In the case of a path as shown which exhibits momentary helical progression, the derivative of the acceleration with time has a component in the direction of helical progression. Each of these parameters is sampled by, or can readily be derived from, the vibratory sensor assembly associated with the wheel, and is typically provided in terms of components parallel to the x, y, z coordinate frame defined above.

General definitions of the vector curvature and vector torsion were given above in the "Background" section. These definitions may be rewritten in a standard manner using the components of the vectors, expressing the vector products and triple products using matrix determinant notation as follows:

wherein x, y, z are unit vectors parallel to the corresponding axes; the sampled velocity v has components (v_x, v_y, v_z); the sampled acceleration a has components (a_x, a_y, a_z); and the time derivative of acceleration α' has components (a_x\ a/, a/).

Since all of these components are either directly output from the vibratory sensor arrangements 14 or may be simply derived from those outputs, these formulae allow straightforward calculation of values for curvature and torsion from the sensor outputs with a low computational overhead readily implemented in real-time.

It has been found empirically that monitoring proportional changes from average values for each of curvature and torsion provide a highly reliable and robust tool for classifying the interaction between a wheel and the underlying surface. For example, increased load on a wheel (e.g., outside wheel during cornering or front wheel during braking) leads to an increase in curvature and a decrease in torsion, and conversely a reduced load leads to decreased curvature and increased torsion. When these deviations from normal/average values reach certain values, they become indicative of impending loss of vehicle control. Preferably, a less extreme threshold value for each parameter is used for classification purposes in order to provide a predictive warning that a potentially hazardous scenario is developing and to allow timely corrective action.

More specifically, it has been found useful for purposes of classifying different scenarios to define both an upper (increase) threshold and a lower (decrease) threshold for each parameter, and most preferably two levels of threshold for both increase and decrease of each parameter. Thus, each parameter may be classified as: normal, high, very high, low, or very low. For convenience of notation, these classifications may be identified as levels or "quantum changes" as follows: normal (0), high (+1), very high (+2), low (-1), and very low (-2). From experimental evidence to-date, it has been found that first thresholds in the range of 40%-60% increase or decrease, and particularly around 50%, are particularly useful. Similarly, it has been found that second (more extreme) thresholds in the range of 70%-90% increase or decrease, and particularly around 80%, are particularly useful. It follows that the parameters generated by processing data from each DCP according to this exemplary implementation and in a case of a four-wheeled vehicle may be represented as eight integers between -2 and +2 corresponding to the "relative levels" of curvature and torsion for each of the four wheels. As long as all values are "0", i.e., within normal operating ranges, no intervention is required. When values outside the first upper or lower threshold ranges are obtained, these eight parameters provide a basis for classification of the current scenario, typically according to either preprogrammed algorithms or a look-up table, and implementation of logging, as well as optionally suitable warning or corrective action. By way of one non-limiting example, Figure 5 illustrates schematically a lookup table which may be used to identify exceedance of safe driving criteria by a vehicle during operation. In each case, the state is identified by subgroup of wheels exhibiting the given abnormal values, the combination of values obtained, and in some cases, the time variance or constancy of the values. The values illustrated for curvature K and torsion τ are the "level" identifiers as defined above, in the range of -2 to +2. Each case specifies the suspected cause of the abnormal measurements.

Turning now to data storage system 18, it should be noted that this unit is functionally defined and may, in practice, be an integrated part of a computer system which also includes processing system 16. Data storage system 18 is responsive to data received from processing system 16 to store data in data storage medium 20 indicative at least of instances in which safe performance criteria have been exceeded. For example, in the case of the integer indices for changes in curvature and torsion in the range from -2 to +2 described above, data logging may be triggered by any index value of ±2, or in certain cases by any value of ±1, as will be mentioned below. Alternatively, particularly in the case where values of parameters output by processing system 16 during "normal" vehicle operation are consistently zero, suitable data compression techniques may make it attractive to store a continuous record of the parameters output by processing system 16.

Data storage medium 20 may be any suitable data storage medium including, but not limited to, fixed solid state memory, removable solid state memory such as a "smart card" or "flash disk", magnetic media or optical media. The storage medium may be removable, for subsequent transfer to an external computer system for data analysis, or may be fixed. In the latter case, a wired or wireless communications interface is provided for uploading the data to a separate computer system for data analysis. In certain cases in which a vehicle is provided with networked communications capabilities, such as for telemetry or vehicle fleet management, the entire storage medium may be located at a location remote from the vehicle.

User display or warning system 22 may be implemented as a warning light or other visual display on the dashboard of the vehicle. Optionally, an already existing multi-function display may be used. In this case, the system may interrupt the normal display content to display a specific suggested corrective action, such as "REDUCE SPEED", in the event of a safe performance criteria exceedance. Alternatively, or additionally, an audio feedback generator may be provided. In this case, a feature of the sound generated, typically pitch, volume, tone duration and/or repetition rate, preferably changes as a function of the measured parameters. For example, a first warning indication may be generated on exceedance of a lower threshold corresponding to the ±1 indices described above, and a second more "urgent" indication may be generated on exceedance of higher thresholds corresponding to the ±2 indices. The first level warning signals may be produced independent of whether the logging system records such events.

At this stage, the operation of system 10, corresponding to the method of the present invention, will be clear. Referring to Figure 4, at step 30 the system samples the vibratory motion for each monitored wheel, and then processes the sampled data. In the preferred case shown here, the processing includes calculating two parameters, specifically the curvature and torsion (step 32), comparing these values to average or pre-stored expected values and calculating a proportional change (step 34). These proportional changes are then used to determine whether the current use of the vehicle exceeds safe performance criteria for the vehicle under current driving conditions, thereby classifying a current status of the vehicle function (step 36), most preferably by applying threshold criteria to the changes and then applying algorithms or look-up tables to determine the current status. When the values are within normal ranges, the average values are optionally updated (step 38) and the monitoring process repeats from step 30. If abnormal conditions are identified, the system performs data logging (step 40), saving data indicative of the exceedance of the threshold criteria. According to certain preferred options, as mentioned above, the system also provides an indication to the driver of exceedance of the thresholds, or of impending exceedance thereof. The exact content and format of the stored data may take many forms. In the particularly preferred classification scheme described above based on percentage changes in vector curvature and vector torsion, non-limiting examples of data stored by the system of the present invention may be any one or combination of the following: L A count of the number of instances in which exceedance of one or both of the parameter thresholds has occurred.

2. An identifier (such as a number indicative of one of the cases from the table of Figure 4, or the curvature and torsion indices themselves) indicative of the type of each exceedance event which has occurred. 3. Driving style indicators other than safe performance criteria exceedance, such as instances of exceeding a warning threshold (e.g., the ±1 thresholds mentioned above) prior to reaching dangerous conditions.

4. A continuous record of vibration-related parameter classification values (such as curvature and torsion indices) for a continuous period, even when no exceedance event has occurred.

5. A record of the time of occurrence of each exceedance event.

6. A record of the duration of each exceedance event.

7. A record of actual calculated vibration-related parameter values (such as curvature and torsion) for the purpose of optional additional post processing.

The selection of parameters stored, and the data handling architecture are preferably chosen depending upon the intended use of the stored data. In a first set of preferred implementations, the data is employed to determine a driving quality adjustment to an insurance premium. In other words, data recorded according to the present invention is used to give an indication of the "driving quality" with which a vehicle is used. Where a driver is found to push his or her vehicle frequently to or beyond its performance limitations, an insurance company may choose to increase the insurance premiums or, in extreme cases, even to refuse to offer insurance to an applicant. Alternatively, or additionally, drivers whose vehicle use is consistently within the safe limits as defined by the criteria applied by the system may be rewarded with an insurance premium reduction. It should be noted that a "normal" or "acceptable" driving style may include occasional exceedance events, such as may occur during emergency braking or other normal evasive action required by events outside the driver's control.

For the purpose of insurance assessment, the safe driving exceedance events are typically monitored continuously over an extended period of vehicle operation, preferably in excess of one month, and most preferably in the range of 6-12 months. For this application, there is typically no need for precise detail regarding the values or timing of the exceedance events. Accordingly, either a simple count of events or a list of exceedance types, optionally together with low-resolution time and/or duration data, is typically sufficient. One simple exemplary implementation of insurance premium adjustment based on a count of exceedance events over a 12 month period is indicated in Figure 6. Clearly, a more sophisticated analysis may be preferred, for example, performing filtering of events based on type, duration and/or closeness in time, or taking into account other parameters, such as the aforementioned non- dangerous driving style indicators.

In a further set of preferred implementations, the present invention may provide a "black box" functionality, recording data about vehicle operation which may be used to derive information about how the vehicle was driven prior to a road accident. Various commercial automobiles are already sold with a rudimentary "black box" recorder which typically maintains a record of vehicle operation parameters in a rolling time window of about 5 seconds for crash analysis. The data is used for research into improving vehicle safety, as well as for investigation of culpability for traffic accidents. The vibration-based analysis of the present invention provides a valuable additional layer of information which gives a direct indication of the tire-road surface interaction during the moments prior to a crash, as well as an indication of the vehicle traction and the efficacy of any evasive action taken by the driver. Optionally, where a driver might have an interest in disabling the logging functions of the system, or of tampering with the data recording, the system may include various precautions for ensuring integrity and completeness of the recorded data. According to one optional feature, the system includes a vehicle disabling arrangement associated with processing system 16 and configured to disable operation of the vehicle if data storage is prevented. Details of this, and other possible precautions to prevent tampering with data, will be clear to one ordinarily skilled in the art by analogy to well know vehicle immobilizing and data protection technology.

It should be noted that, while there exist many systems for monitoring of vehicle position, speed and acceleration, the logging system of the present invention and the corresponding method are distinguished, among other features, by the use of wheel vibration measurements in at least two dimensions as described above, thereby providing an accurate indication of the tire-road interface performance under current driving conditions. Thus, unlike simple acceleration-based systems, the system of the present invention can distinguish between cautious driving on a good road surface and driving with identical acceleration and braking on a wet or icy road which may exceed safe performance limitations. As a result, the present invention provides a much more reliable indication of the care and skill of the driver than is offered by conventional systems.

In addition to the logging function described hereinabove, the system of the present invention may optionally also include components for direct corrective action, such as automatic application of brakes at one or more wheel, in order to return vehicle operation to within safe operating parameters. The relevant features of a system and mode of operation for performing such corrective action, as well as additional diagnostic functions which can be performed by the system of the present invention, are described in US Patent Application Publication No. US 2006/0265114 Al which is here by incorporated by reference. The aforementioned application is co-assigned with the present application, has a common inventive entity with this application, and was unpublished prior to the date of filing this application, such that it does not constitute prior art against the present application.

Example

At DCP, the values of velocity and acceleration of vibrations along longitudinal, transverse and vertical for are

FRONT RIGHT WHEEL Velocity (x) = 0.0361299

Velocity (y) = 0.029545

Velocity (z) = 0.0742197

Acceleration (x) = 0.00094975

Acceleration (y) = 0.00143623 Acceleration (z) = 0.00138715

Derivative of acceleration (x) = 0.000712

Derivative of acceleration (y) = 0.00158232

Derivative of acceleration (z) = 0.000592 x 0.0361299 0.029545 0.0742197

0.00094975 0.00143623 0.00138715 Curvature =

0.0361299 ^Λ2 + 0.029545 ^Λ2 + 0.0742197 ^Λ2

;

Curvature = 0.1079133620 0.0361299 0.029545 0.0742197

0.00094975 0.00143623 0.00138715

0.000712 0.00158232 0.000592

Torsion = x y z

0.0361299 0.029545 0.0742197 0.00094975 0.00143623 0.00138715

Torsion = 0.0793359665

Average Curvature = 0.2180802397

Average Torsion = 0.1746612365

The percentage difference between curvature and average curvature is given by

The percentage difference between torsion and average torsion is given by

(T - AT)/ * I QQ

/AT ^{1 U} = -54.57723 i.e. Low

FRONTLEFT WHEEL

Velocity (x) = 0.1531569 Velocity (y) = 0.06180523 Velocity (z) = 0.1798801 Acceleration (x) = 0.000301166 Acceleration (y) = 0.00701622 Acceleration (z) = 0.00305532 Derivative of acceleration (x) = 0.0002112 Derivative of acceleration (y) = 0.00198112 Derivative of acceleration (z) = 0.0008916944 Calculating the matrix formulae as before gives: Curvature = 0.1072234463 Torsion = 0.0792064760 Average Curvature = 0.2180802397 Average Torsion = 0.1746612365 Percentage change in curvature = -50.83302987 i.e. Low Percentage change in torsion = -54.65137107 i.e. Low

REAR RIGHT WHEEL Velocity (x) = 0.2207165

Velocity (y) = 0.0669311

Velocity (z) = 0.17

Acceleration (x) = 0.000515644

Acceleration (y) = 0.00653633 Acceleration (z) = 0.00968345

Derivative of acceleration (x) = 0.000800672

Derivative of acceleration (y) = 0.00076896

Derivative of acceleration (z) = 0.001632

Calculating the matrix formulae as before gives: Curvature = 0.1075362292

Torsion = 0.0795573662

Average Curvature = 0.2180802397

Average Torsion = 0.1746612365

Percentage change in curvature = -50.68960426 i.e. Low Percentage change in torsion = -54.45047353i.e. Low

REAR LEFT WHEEL

Velocity (x) = 0.0485634 Velocity (y) = 0.105931 Velocity (z) = 0.0793662

Acceleration (x) = 0.00027004 Acceleration (y) = 0.00305029 Acceleration (z) = 0.0042401999 Derivative of acceleration (x) = 0.0003648672 Derivative of acceleration (y) = 0.000785304

Derivative of acceleration (z) = 0.0006312

Calculating the matrix formulae as before gives:

Curvature = 0.1077510154 Torsion = 0.0796132520

Average Curvature = 0.2180802397

Average Torsion = 0.1746612365

Percentage change in curvature = -50.59111476 i.e. Low

Percentage change in torsion = -54.41847683 i.e. Low Thus, in this example, all four wheels are currently exhibiting a first level decrease (-1) in both curvature and torsion. This indicates an onset of deterioration in the road holding capability of a specific tire of automobile due to slippery road conditions like wet/icy roads. In this example, road holding capability of all the four tires is at the onset of deterioration. In this situation, the system of the present invention would preferably generate a warning to the driver, and would optionally log a non-dangerous event approaching performance limits. If this situation worsens, the curvature and torsion would change to (-2), corresponding to case #1 in Figure 5, indicating immediate risk of automobile skid on braking or hydroplaning at higher speeds. Such an exceedance would be recorded as a dangerous performance limitation exceedance.

It will be appreciated that the above descriptions are intended only to serve as examples, and that many other embodiments are possible within the scope of the present invention as defined in the appended claims.

Claims

WHAT IS CLAIMED IS:

1. A method for logging instances in which a road vehicle is driven in a manner exceeding safe performance criteria, the vehicle having at least four wheels, the method comprising:

(a) repeatedly measuring wheel-hub vibrational motion at the hub of each of at least four of the wheels;

(b) processing the vibrational motion measurements to determine whether the current use of the vehicle exceeds safe performance criteria for the vehicle under current driving conditions; and

(c) storing data indicative at least of instances in which safe performance criteria have been exceeded, wherein said measuring and said processing measure and process the vibrational motion for each wheel along at least two perpendicular axes.

2. The method of claim 1, wherein the vibrational motion is measured along three perpendicular axes.

3. The method of claim 2, wherein said processing includes:

(a) processing the vibrational motion measurements along each of said at least two axes for each wheel to derive, substantially in real time, at least one motion-related parameter for each wheel; and

(b) applying classification criteria to values of said at least one motion- related parameter for all of the wheels to identify whether the current use of the vehicle exceeds safe performance criteria for the vehicle under current driving conditions.

4. The method of claim 3, wherein said at least one motion-related parameter for each wheel includes a parameter which varies as a function of a balance between vertical vibrations and horizontal vibrations of the wheel hub.

5. The method of claim 3, wherein said at least one motion-related parameter for each wheel includes a proportional deviation of the vector curvature of the wheel-hub vibrational motion.

6. The method of claim 5, wherein said proportional change is evaluated relative to an average value of vector curvature.

7. The method of claim 3, wherein said at least one motion-related parameter for each wheel includes a proportional deviation of the vector torsion of the wheel-hub vibrational motion.

8. The method of claim 7, wherein said proportional change is evaluated relative to an average value of vector torsion.

9. The method of claim 3, wherein said at least one motion-related parameter for each wheel includes:

(a) a proportional deviation of the vector curvature of the wheel-hub vibrational motion; and

(b) a proportional deviation of the vector torsion of the wheel-hub vibrational motion.

10. The method of claim 9, wherein said proportional changes are evaluated relative to an average value for each of said vector curvature and said vector torsion.

11. The method of claim 3, wherein said classification criteria include applying both an upper threshold and a lower threshold to said at least one motion- related parameter for each wheel.

12. The method of claim 1, further comprising employing said data to determine a driving quality adjustment to an insurance premium.

13. The method of claim 1, further comprising employing said data to derive information about how the vehicle was driven prior to a road accident.

14. A system for logging instances in which a road vehicle is driven in a manner exceeding safe performance criteria, the vehicle having at least four wheels, the system comprising:

(a) a wheel vibratory sensor arrangement associated with each of at least four of the wheels, said wheel vibratory sensor arrangement being configured to measure vibrational motion of the hub of each wheel along at least two perpendicular axes;

(b) a processing system in data communication with each of said wheel vibratory sensor arrangements, said processing system being configured to process the vibrational motion measurements to determine whether the current use of the vehicle exceeds safe performance criteria for the vehicle under current driving conditions; and

(c) a data storage system associated with said processing system and configured to store data indicative at least of instances in which safe performance criteria have been exceeded.

15. The system of claim 14, wherein said wheel vibratory motion sensor arrangement includes a tri-axial accelerometer associated with each of said at least four wheels and deployed for measuring vibrational motion of the hub of each wheel along three perpendicular axes.

16. The system of claim 14, wherein said data storage system is configured to store said data together with a time at which each instance of safe performance criteria exceedance occurs.

17. The system of claim 14, wherein said data storage system is configured to store said data together with data relating to a duration during which safe performance criteria were exceeded.

18. The system of claim 14, wherein said data storage system is configured to selectively retain data indicative of most recent instances in which safe performance criteria have been exceeded.

19. The system of claim 14, further comprising a vehicle disabling arrangement associated with said processing system and configured to disable operation of the vehicle if said data storage is prevented.