Control system and method for a vehicle

Abstract

A control system (208) configured to sense at least an area rearwards of the rear wheels Wr of a first vehicle 10 to obtain sensor 120a-g data, and determine in dependence on the sensor data, tracks (fig. 4) formed by the first vehicle in the sensed area of the terrain surface. The control system is then configured to analyse the determined tracks to estimate one or more surface characteristic of the terrain surface and output a surface signal indicative of the estimated one or more surface characteristic. In this way, the control system can utilise tracks formed by the front and / or rear wheels of a vehicle to ascertain the nature of the surface, and in particular its characteristics. Vehicle driving parameters may then be set or modified having regard to the surface characteristics, either or both for the vehicle itself, or of another vehicle, for example a following vehicle. This relies on the fact that the tread or ruts left in a surface by the front wheels of the vehicle as that vehicle traverses the surface are often a good indicator of the surface characteristics.

Description

TECHNICAL FIELD The present disclosure relates to a control system and method for a vehicle. Aspects of the invention relate to a control system, a control method, a vehicle and a computer program. BACKGROUND When an expert driver is driving a vehicle on a surface, they are often able to ascertain the surface quality in terms of how challenging it is to traverse, and are therefore able to change their driving behaviour or optimise the vehicle settings (for example by changing a driving mode) as required. However, less experienced drivers may not be capable of this, potentially leading to challenges in traversing the surface. If a vehicle sinks into a soft surface, such as sand or mud, this can make it very challenging to continue to make progress - especially when the vehicle sinks deep enough such that the suspension or body of the vehicle drags on the surface. While systems may be present on the vehicle to mitigate these scenarios, activation of these systems are dependent on the driver, relying on the driver’s experience being sufficient to deal with this scenario. It is an aim of the present invention to address one or more of the disadvantages associated with the prior art. SUMMARY OF THE INVENTION Aspects and embodiments of the invention provide a control system, a control method, a vehicle and a computer program as claimed in the appended claims. According to an aspect of the present invention there is provided a control system for a vehicle, the control system comprising one or more controller, the control system configured to: sense at least an area rearwards of the rear wheels of a first vehicle to obtain sensor data; determine in dependence on the sensor data, tracks formed by the first vehicle in the sensed area of the terrain surface; analyse the determined tracks to estimate one or more surface characteristic of the terrain surface; and output a surface signal indicative of the estimated one or more surface characteristic. In this way, the control system is able to utilise tracks (in the driving surface) formed by the front and / or rear wheels of a vehicle to ascertain the nature of the surface, and in particular its characteristics. Vehicle driving parameters may then be set or modified (optimised) having regard to the surface characteristics, either or both for the vehicle itself, or of another vehicle, for example a following vehicle. This relies on the fact that the tread or ruts left in a surface by the front wheels of the vehicle as that vehicle traverses the surface are often a good indicator of the surface characteristics. The one or more controllers may collectively comprise: at least one electronic processor having an electrical input for receiving the sensor data; and at least one memory device electrically coupled to the at least one electronic processor and having instructions stored therein; and wherein the at least one electronic processor is configured to access the at least one memory device and execute the instructions thereon so as to determine the tracks, analyse them to estimate the surface characteristics, and output the surface signal. The surface signal indicative of the estimated one or more surface characteristic may be output to a second vehicle. This enables that other vehicle to either notify its driver, or automatically (or semi-automatically) set its on driving parameters in preparation for traversing the surface. This is particularly beneficial where vehicles are travelling in convoy - enabling forward vehicles in the convoy to provide terrain information to rearward vehicles in the convoy. The surface signal may comprise data relating to setting or modifying one or more driving parameter of the second vehicle in dependence on the estimated one or more surface characteristic. The control system may be configured to set or modify one or more driving parameters of the first vehicle in dependence on the surface signal. The control system may be configured to analyse the determined tracks by determining changes to the tracks with respect to time. This may for example be carried out by determining a rate of change of depth of the tracks with respect to time, or a rate of collapse of the tracks with respect to time. The changes may be determined as the tracks are being formed by the rear wheels of the first vehicle or subsequently to the tracks being fully formed by the rear wheels of the first vehicle. It will be appreciated that both changes as the tracks are being formed, and changes following the formation of the tracks, may be implemented in a single system, and used together to determine the surface characteristics of the terrain. The analysis of the tracks may take into account a weight of the first vehicle, or a force exerted on the ground at each wheel of the first vehicle, and / or a diameter and / or width for the wheels of the first vehicle, and / or a speed of the first vehicle at the time the tracks were formed. In some implementations the control system may be configured to analyse the tracks by estimating a contact patch pressure for the wheels of the first vehicle, using the above parameters of the vehicle. The control system may be configured to analyse the tracks by identifying one or more of a depth of the tracks, an angle of the sides of the tracks and a presence of water within the tracks. The control system may be further configured to analyse the tracks by identifying a change with respect to time of one or more of the depth of the tracks, an angle of the sides of the tracks and a presence of water within the tracks. In this way, the present technique takes account of how the terrain changes dynamically due to a vehicle traversing it, allowing the vehicle to automatically adapt to the driving surface ahead, without requiring driver expertise to select appropriate driving parameters. The control system may be configured to estimate the one or more surface characteristics of the terrain surface by taking account of one or more other inputs, such as one or more of an ambient temperature, recent precipitation, and an expected terrain surface type. These additional inputs may allow the control system to determine which of multiple surface characteristics or surface types apply to the terrain when the track structure itself is not able to identify these alone - that is, where a given track structure could result from two or more different surface characteristics or terrain types. The driving parameters set in response to the determined surface characteristics may comprise one or more of a throttle map for the vehicle, a wheel slip parameter, and a ride height for a suspension system of the vehicle. The control system may be further configured to sense an area ahead of a vehicle and / or beneath the vehicle to obtain sensor data, to determine, from the sensor data, tracks ahead of and / or beneath the vehicle in a terrain surface, to analyse the determined tracks ahead and / or beneath the vehicle to estimate one or more surface characteristics of the terrain surface, and to set or modify one or more driving parameters of the vehicle in dependence on the estimated surface characteristics from the tracks ahead and / or beneath the vehicle, wherein the control system is configured to refine the estimated surface characteristics and / or the one or more driving parameters established based on the tracks determined ahead of and / or beneath the vehicle based on the analysis of the tracks behind the vehicle. According to another aspect, there is provided a vehicle comprising the control system according to the above, and one or more sensors configured to image the area behind the first vehicle to generate the sensor data. According to another aspect, there is provided a control method for a vehicle, the method comprising: sensing at least an area rearwards of the rear wheels of a first vehicle; determining tracks made by the vehicle in a terrain surface within the sensed area; analysing the determined tracks to estimate one or more surface characteristics of the terrain surface; and outputting a surface signal indicative of the estimated one or more surface characteristic. According to another aspect there is provided computer software that, when executed, is arranged to perform a method according to the above. The present disclosure describes three main embodiments, which may be used together or separately. The first implementation senses tracks ahead of the vehicle, made by another vehicle, or in some cases a pedestrian or even an animal. In this case vehicle parameters may be set or adjusted to prepare the vehicle to traverse the terrain ahead to take into account surface characteristics identified from the tracks. The second implementation senses tracks beneath, or to one or other side of the vehicle, made by the front wheels of the vehicle itself. The tracks in this case are sensed from close range, providing more detailed sensor data, and additionally the characteristics of the entity which made the tracks (the vehicle itself) are well known, with the result that the sensed data paired with knowledge of the vehicle itself may provide a better indication of the surface characteristics of the surface being traversed. In this case vehicle parameters of the vehicle itself are also set or adjusted, but in this case those vehicle parameters may particularly relate to settings controlling the rear wheels of the vehicle, which have yet to reach this portion of the terrain (although more general vehicle settings relevant to the front wheels may be adjusted too). Finally, the third implementation senses tracks on an area of the driving surface behind the rear wheels of the vehicle, made by the (front and / or rear) wheels of the vehicle. Similarly to the second embodiment, the tracks are in this case sensed at a high resolution, and with reliable and detailed knowledge of the nature of the entity which made the tracks (the vehicle itself). In the third embodiment, the information on surface characteristics is less valuable to the vehicle itself (since it has already traversed the surface), but may aid with estimating the surface characteristics in the first two embodiments (by determining errors in estimates made using the first two embodiments, and refining current and / or future estimates). In addition, the information on surface characteristics obtained in the third embodiment is particularly beneficial to other vehicles, which may subsequently traverse the same terrain (for example when driving in convoy). As such, in the third embodiment the estimated surface characteristics are communicated to a second vehicle, which may adapt one or more of its driving parameters to account for the terrain surface. Within the scope of this application it is expressly intended that the various aspects, embodiments, examples and alternatives set out in the preceding paragraphs, in the claims and / or in the following description and drawings, and in particular the individual features thereof, may be taken independently or in any combination. That is, all embodiments and / or features of any embodiment can be combined in any way and / or combination, unless such features are incompatible. The applicant reserves the right to change any originally filed claim or file any new claim accordingly, including the right to amend any originally filed claim to depend from and / or incorporate any feature of any other claim although not originally claimed in that manner. BRIEF DESCRIPTION OF THE DRAWINGS One or more embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which: Figure 1 shows a schematic illustration of a vehicle having a control system according to an embodiment of the invention; Figure 2 shows a schematic illustration of the control system within Figure 1; Figure 3 shows a schematic flow diagram of a high level method implemented by embodiments of the invention; Figure 4 shows a schematic flow diagram of a method according to a first embodiment of the invention relating to analysis of tracks ahead of the vehicle; Figure 5 shows a schematic illustration of tracks at a plurality of times following their formation; Figure 6 shows a schematic flow diagram of a variation on the first embodiment; Figure 7 shows a schematic flow diagram of a method according to a second embodiment of the invention relating to analysis of tracks beneath the vehicle; Figure 8 shows a schematic flow diagram of a method according to a third embodiment of the invention relating to analysis of tracks behind the vehicle; and Figure 9 shows a schematic flow diagram of a method of adjusting rear wheel steering to avoid tracks formed by the front wheels of the vehicle. DETAILED DESCRIPTION A vehicle and control system in accordance with an embodiment of the present invention is described herein with reference to the accompanying Figure 1. With reference to Figure 1, there is illustrated a vehicle 10 having a control system 208. The control system 208 comprises one or more controllers responsible for controlling various aspects of the vehicle, including the powertrain, suspension, infotainment and many others. The present technique is concerned with sensing a driving surface of terrain in the vicinity of (ahead of, beneath and behind) the vehicle, and as such the vehicle 10 is provided with a plurality of sensors 120a- 120g. The sensors 120a and 120b are disposed at or towards the front of the vehicle. The sensors 120a, 120b are responsible for capturing sensor data in relation to a portion of the driving surface ahead of the vehicle 10, or ahead of the front wheels of the vehicle 10 (that is, ahead of the line Wf which passes through the contact points of the front left and right wheels of the vehicle with the driving surface). The sensors 120c and 120d are disposed at the side of the vehicle 10, and in the present case on the wing mirrors of the vehicle 10. The sensor 120e is disposed beneath the vehicle 10. The sensors 120c, 120d, 120e are responsible for capturing sensor data in relation to a portion of the driving surface beneath or to the side of the vehicle 10 (for example between the front wheels and rear wheels of the vehicle 10). This portion of the driving surface is between the line Wf and a line Wr which passes through the contact points of the rear left and right wheels of the vehicle with the driving surface. The sensors 120f and 120g are disposed at the rear of the vehicle 10. The sensors 120f and 120g are responsible for capturing sensor data in relation to a portion of the driving surface behind the vehicle 10, or behind the rear wheels of the vehicle 10 (that is, behind the line Wr). The vehicle 10 also comprises a user interface 130, via which information can be presented to a driver (or occupant) of the vehicle 10, and via which the driver (or occupant) of the vehicle is able to enter information which controls, or influences the control of, the vehicle 10. The vehicle 10 also comprises a transceiver 140 which is capable of transmitting and receiving data to and from other vehicles, or to and from one or more servers. It will be appreciated that the vehicle 10 in practice comprises other vehicle systems, such as steering (front and optionally rear), apowertrain, heating, ventilation and air conditioning (HVAC), amongst others. Two example systems are shown in Figure 1, these being the powertrain 152 and the suspension system 154. The controller 208 is configured to control aspects of these (and other) systems, based in part of the sensor data acquired by the sensors 120a-120g. It will be appreciated that if only one of the three main embodiments described herein is being implemented in a vehicle, only a subset of the sensors 120a -120g may be required. It will also be appreciated that, in certain embodiments, a given sensor may capture sensor data usable for more than one of these embodiments, such as of (for example) the driving surface ahead of the vehicle (or at least ahead of the front wheels of the vehicle), the driving surface between the front and rear wheels of the vehicle, and the driving surface behind the rear wheels of the vehicle. In other words, the same sensor may contribute sensor data for use in one, two or even three embodiments (in principle, although in practice it is likely to be difficult to site a sensor on a vehicle in a position which permits sensing of all of the different areas of driving surface considered by the three embodiments). The embodiments differ, in part, in terms of how the sensor data is used. It will be appreciated that, in an alternative implementation, the sensor data may be captured by a sensor not mounted to the vehicle. For example, a user’s smartphone may be configured to capture image data of the driving surface ahead of, to one side, or behind the vehicle, and this image data may be used. Alternatively, a drone (controlled or autonomous) may hover above or proximate the vehicle, and obtain image data used in the present technique. For all embodiments, the control system 208 is configured to receive sensor data from the sensors 120a to 120g (or other sensors) and determine the presence of tracks on the sensed driving surface. The tracks are analysed by the control system 208, and surface characteristics of the driving surface estimated based on the analysed tracks. In other implementations the image data may be offboarded to a remote server or other device for analysis, and the results returned subsequently to the control system 208. The control system 208 may then output a control signal to either (or both) provide a driver of the vehicle 10 (via the user interface 130), or of another vehicle (communicating via the transceiver 140), with an indication of the nature, type or characteristics of the driving surface, or to control one of more aspects of the vehicle, such as torque delivery, ride height or wheel slip for example. The control system 208 is configured to implement any one or more of the methods described herein. Figure 2A illustrates how the control system 208 may be implemented. The control system 208 of Figure 2A illustrates a controller 200. In other examples, the control system 208 may comprise a plurality of controllers 200 onboard and / or off board the vehicle 10. In examples any suitable control system 208 can be used. The controller 200 of Figure 2A includes at least one processor 202; and at least one memory device 204 electrically coupled to the electronic processor 202 and having 4 instructions 206 (for example a computer program) stored therein, the at least one memory device 204 and the instructions 206 configured to, with the at least one processor 202, cause any one or more of the methods described herein to be performed. Also illustrated in the example of Figure 2A are one or more vehicle systems 226. In examples, the vehicle system(s) 226 can comprise any suitable vehicle system(s). For example, the vehicle system(s) 226 can comprise any suitable vehicle system(s) 226 from which the control system 208 can receive and / or to which the control system 208 can transmit, directly or indirectly, one or more signals 20, for example to control the suspension system or torque delivery system of the vehicle 10. The torque delivery system for example may be part of the powertrain 152 identified in Figure 1, while the suspension system may be the suspension system 154 identified in Figure 1. Figure 2B illustrates a non-transitory computer readable storage medium 218 comprising the instructions 206 (computer software). Accordingly, Figure 2B illustrates a non-transitory computer readable medium 218 comprising computer readable instructions 206 that, when executed by a processor 202, cause performance of at least one or more of the methods described herein. Figure 3 is a schematic flow diagram illustrating the present technique at a high level. The method shown in Figure 3 is a method of control of a vehicle 10, such as the vehicle 10 illustrated in Figure 1. In particular, the method is a method of estimating the type and / or surface characteristics of a driving surface to be, or being, traversed by the vehicle 10. The method may be performed by the control system 208 illustrated in Figures 1 and 2. In particular, the memory may comprise computer-readable instructions which, when executed by the processor, perform the method according to an embodiment of the invention. In particular, at a step A1 one or more vehicle-mounted sensors sense a portion of the terrain around or under the vehicle. The present disclosure considers separately the handling of captured sensor data dependent on whether it relates to an area of terrain in front of the (front wheels of the) vehicle (first embodiment), underneath or to the side of the vehicle (between the front and rear wheels thereof) (second embodiment) or behind the (rear wheels of the) vehicle (third embodiment). However, the steps of Figure 3, in a general sense, are common to all embodiments, which may be carried out in parallel, and may interact in various ways as will be explained below. At a step A2 the sensor data is analysed, to identify tracks, and to determine one or more surface characteristics of the terrain based on those tracks. As will be explained, in most cases the analysis utilises changes in the tracks over time. At a step A3 one or more actions is taken (for example one or more control signals is output) in dependence on the determined surface characteristics. These actions may be to set or adjust a vehicle parameter of the vehicle bearing the sensors, or to carry out an action at that vehicle, or to set or adjust a vehicle parameter of another vehicle (in the vicinity), or to carry out an action on that other vehicle, or to notify the driver of the vehicle or a driver of another vehicle of the nature of the terrain and I or recommended driving actions to take. Figures 4A to 4D show a cross sectional view of a track, as it changes over time, after it has been formed. In particular, Figure 4A shows a track immediately after it has been formed by the passage of a vehicle tyre. Figures 4B to 4D then show the same track at intervals over the following few seconds or minutes. As can be seen from Figures 4A to 4D, the track is a deformation of the terrain which comprises at least a depression into the terrain surface (that is, a portion which is below ground level, as indicated by the dashed line in Figures 4A to 4D), and which may comprise a raised portion which extends above ground level to one or both sides. In Figure 4A, the track has a depth D1 below ground level. The raised portion of the track, to either side of the depression, has a height H1 above ground level. As can be seen from Figures 4B to 4D, the height H1 decreases over time as the raised portions collapse. Although not represented in Figures 4A to 4D, the depth D1 may also increase, for example as the walls collapse and materials fills up the track from the bottom. In some cases the angle I pitch of the walls may decrease (become less steep) over time, as also represented in Figures 4A to 4D. Also shown in Figures 4A to 4D is a build-up of water in the tracks. Assuming that water is present in the terrain (but not covering the terrain, in which case the tracks would fill substantially immediately), then water in the material surrounding the track will fill the track over time. The rate at which this occurs may be indicative of the water content of the terrain surface. All of these characteristics (and others) can be measured by the sensors. In some cases the tyres may leave tread prints (not shown) in the base portion of the track. These tread prints may fade over time, again providing an indication of the characteristics of the driving surface in which the tracks have been formed. Summary of first embodiment (tracks ahead) The control system according to the first embodiment is configured to sense an area ahead of a vehicle to obtain sensor data. The control system is further configured to determine, in dependence on the sensor data, tracks ahead of the vehicle in a terrain surface. These tracks may have been formed by a further vehicle, for example a lead vehicle or a vehicle ahead when travelling in convoy. Alternatively, the present technique may also utilise tracks made by another entity such as a person or animal (for example a horse or camel, each of which are likely to traverse off road surfaces also usable by a suitably equipped vehicle). The sensor data may be captured by sensing devices including one or more of cameras, LIDAR, RADAR, and infrared (for example). Other sensing methodologies may also be used. The control system is configured to analyse the determined tracks to estimate one or more surface characteristics of the terrain surface. Various forms of analysis may be carried out, as will be discussed in detail below. Once the analysis has been performed, the control system is configured to set or modify one or more driving parameters of the vehicle in dependence on the estimated surface characteristics. Various vehicle systems may benefit from this terrain dependent control, such as a throttle map for the vehicle, a wheel slip parameter, and a ride height for a suspension system of the vehicle. In this way, the vehicle can be tuned to provide improved performance (and an improved driving experience) over the terrain surface ahead. Other parameters may also be set (in addition to or alternatively to those indicated above), such as steering adjustments to avoid or follow tracks, gear selection (to select a most appropriate transmission gear to apply the appropriate torque range for negotiating the surface, and to define when gear shifts should occur), braking system responsiveness (that is, how the braking system responds to a given brake pedal input from the driver), and torque weighting between the rear and front wheels of the vehicle. A vehicle set speed may also be applied, to cause the vehicle to (automatically) apply a suitable speed for traversal of the identified surface. Advantageously, the present technique does not consider tracks as a static feature of the terrain, but as a time-dependent feature having a structure which varies, either as the track is being formed, or over time after it has been formed. As such, the control system is configured to analyse the detected tracks by determining changes to the tracks with respect to time. Several features of the tracks may be analysed in this way. For example, the control system may be configured to determine changes to the tracks with respect to time by determining a rate of change of depth of the tracks with respect to time, a rate of collapse of the tracks with respect to time, an angle of the sides of the tracks and a presence of water within the tracks. The control system is also able to analyse the tracks by determining whether the tracks were left by a part of the further vehicle other than its wheels (such as the underside of the chassis). This would suggest that the further vehicle was sinking deeply into the terrain ahead, and may inform the determination of the surface characteristics (e.g. that the terrain ahead is soft). As will be discussed subsequently, a better estimate of the surface characteristics may be made by identifying one or more characteristics of the entity making the tracks. For example, where the entity making the tracks is a vehicle, the one or more characteristics used to improve the surface characteristic estimate may comprise a weight of the vehicle, a force exerted on the ground at each wheel of the vehicle, a diameter and / or width for the wheels of the vehicle forming the tracks, and a speed (and / or acceleration / deceleration / torque demand) of the vehicle at the time the tracks were formed. As an intermediate step, some of these characteristics may make it possible to estimate a contact patch pressure for each wheel of the vehicle forming the tracks, which when combined with the tracks actually formed, provide an indication of the nature of the terrain surface. In some embodiments the control system may be configured to estimate the one or more surface characteristics of the terrain surface more accurately by taking account of one or more other inputs, such as an ambient temperature, recent rainfall, and an expected terrain surface type. In addition to adjusting the vehicle driving parameters, the control system may also be configured to communicate the estimated surface characteristics to another vehicle, preferably in association with an indication of the geographical location corresponding to those surface characteristics. By sharing this information, other vehicles without direct visibility of the tracks (which may have been erased or eroded by the time the relevant portion of the terrain has been reached by that vehicle) may benefit from being advised of the surface characteristics ahead (or in the vicinity). The other vehicle may automatically set driving parameters accordingly, or may make the driver aware of the nature of the terrain ahead. More generally, the present technique may determine trafficability of the surface ahead of the vehicle based on the analysed tracks and / or the estimated surface characteristics, and further configured to notify the driver of the determined trafficability. Trafficability is the condition or a measure of the suitability of terrain for being travelled over, in the present case by a vehicle. The driver can then make an informed decision as to whether to attempt traversal of the terrain ahead. To aid in this, an audible or visual alert may be generated if the determined trafficability is indicative of the 6 vehicle being likely to have difficulty in safely traversing the terrain surface ahead. The determination of trafficability may be enhanced by identifying a type of vehicle which formed the tracks. For example, if the tracks ahead were formed by a vehicle with a greater capability on that surface type than the instant vehicle, then if the tracks indicate that the vehicle ahead only just successfully traversed the terrain, this would suggest it would be too challenging a surface for the instant vehicle. Accordingly, an indication to a driver that the terrain ahead is not suitable for their traversal (or may be challenging) may be made on the basis of (a) the capabilities and characteristics of the vehicle which made the tracks, (b) the estimated surface characteristics I analysed tracks, and (c) the capabilities and characteristics of the vehicle being driven by the driver. Referring to Figure 5, at a step B1 an area of the terrain (driving surface) ahead of the (front wheels of) the vehicle is imaged using the sensors 120a, 120b. For the purposes of explanation, the sensors 120a, 120b will be described as cameras, but itwill be understood that other sensing technologies may be used instead, as described elsewhere. The sensors 120a, 120b therefore capture an image of the terrain ahead of the vehicle 10. Ata step B2 the image is processed to extract the tracks. This could be achieved in many ways, for example by filtering for high frequency changes in the image, by image recognition (tracks having a known appearance), by stereoscopic imaging (for example combining together the images from the two cameras 120a, 120b to form a point cloud, and then comparing the point cloud to an assumed surface plane to identify track features such as depth, height, wall angle for example) or any other method. Where other types of sensor are used, such as LIDAR or RADAR, determining the 3D structure of the terrain surface and tracks may be achieved more directly. Such techniques for isolating a 3D structure (track) from sensor data are well known to the person skilled in the art. At a step B3 the track data extracted from the sensor data is analysed. Various features may be extracted from the track data, such as (but not limited to) those discussed above in relation to Figures 4A to 4D. This is repeated multiple times (for example, in relation to successive image frames), to acquire track data over time. At a step B4, the track data at the multiple times are compared to identify changes in the tracks over time. More specifically, changes to each of the measured characteristics of the track with respect to time are identified. This may result in the determination of (for example) a rate of change of depth (depth change per second or minute for example), a rate of change of height (height change per second or minute for example), a rate of change of wall angle (number of degrees per second or minute for example), and a rate of fill of water (for example depth change per second or minute for example). At a step B5, vehicle information (of the vehicle which formed the tracks) and other inputs are obtained. As described previously, the vehicle information may include wheel size, vehicle mass, mass distribution (between front and rear wheels), tyre pressure, vehicle speed and vehicle acceleration or deceleration (at the time the tracks were formed by the vehicle). The other inputs may include environmental information such as ambient temperature, and recent rainfall. At a step B6, the track information (static and / or time variant), vehicle information and other information are analysed to identify one or more surface characteristics of the driving surface, and to classify the nature of the driving surface in which the track has been formed. For example, the driving surface (or a portion thereof) may be classified based on these parameters into one of a plurality of predetermined classes, each of which is considered to relate to a particular set and / or range of surface characteristics, and each of which is to be treated differently by the driver and / or vehicle, and each of which corresponds to a particular combination of surface characteristics. For example, considering four terrain types which may be available for classification (optionally along with others), these may be Type A, Type B, Type C and Type D. A first set and / or range of surface characteristics correspond to the Type A surface. Generally, these may be characteristics corresponding to dry sand. A second set and / or range of surface characteristics correspond to the Type B surface. Generally, these may be characteristics corresponding to wet sand. A third set and / or range of surface characteristics correspond to the Type C surface. These may be characteristics corresponding to dry grass. A fourth set and / or range of surface characteristics correspond to the Type D surface. These may be characteristics corresponding to wet grass. At a step B7, one or more driving parameters of the vehicle are set or adjusted (directly or indirectly) in dependence on the determined surface characteristics and / or in dependence on the classified surface type. In some implementations the driving parameters may be set directly based on the surface characteristics, but preferably the surface characteristics are used to classify the terrain type, and the classified terrain type is used to set or adjust the driving parameters. In some implementations the driving parameters may be set automatically, whereas in other implementations a driving parameter change may be suggested to the driver, who is able to accept the recommendation (in which case the driving parameters are changed), or to set driving parameters themselves (for example by selecting a driving mode, or terrain mode). In some cases, the analysis of the tracks ahead, optionally paired with other inputs relating to the vehicle which made the tracks, the vehicle being driven, and other environmental factors, may suggest that the terrain ahead is not safe for traversal. This may cause the vehicle to automatically stop, or to recommend to the driver that progress be halted. In parallel with the step B7, at a step B8, the surface characteristics and / or the terrain type are communicated externally of the vehicle, for example to another vehicle. The other vehicle is then able to notify its driver and / or set or adjust its own driving parameters when approaching or traversing the driving surface. It will therefore be appreciated that the first embodiment seeks to monitor (a change in) tracks ahead of the vehicle as (or after) a preceding vehicle (or other entity) passes over them to infer how best the vehicle should be controlled over the approaching terrain. Sensors and existing cameras on the vehicle may be used to monitor the change in tracks, such as rate of change of depth of tracks or rate of collapse of the tracks, to determine the response of the terrain to the vehicle. As an additional feature, vehicle identification for the vehicle ahead could be factored in, to better influence the control (e.g. determination of the mass of the vehicle in front if vehicle type can be identified). Summary of the second embodiment (tracks beneath) The control system of the second embodiment is configured to sense an area beneath or to a side of the vehicle to obtain sensor data. In the case where both the first and second embodiment are implemented (in practice by the same control system), this means that the area being imaged by the technique of the second embodiment may have been previously imaged (and analysed) by the first embodiment. The analysis carried out in both cases may be similar. In any case, the control system is configured to determine, in dependence on the sensor data, tracks formed by the front wheels of the vehicle in the sensed area of the terrain surface. In some cases, the determined tracks may be matched against the tracks determined in the first embodiment, and compared. The control system is configured to analyse the determined tracks to estimate one or more surface characteristics of the terrain surface. Then, one or more driving parameters of the vehicle may be set or adjusted in dependence on the estimated one or more surface characteristics. While the driving parameters adjusted by the second embodiment may be the same as those of the first embodiment, with the second embodiment there is an opportunity to target the adjustments to the rear wheels of the vehicle, which have yet to reach the portion of the terrain surface being sensed by the second embodiment. Preferably then, the one or more driving parameters being set or modified comprise one or more parameters influencing the rear wheels of the vehicle. For example, these may be parameters used to control one or more electric machines providing torque to the rear wheels of the vehicle separately to the front wheels of the vehicle. Alternatively (or additionally), the parameters may relate to a rear wheel steering system of the vehicle, in which case the control system may be configured to control a rear wheel steering system of the vehicle to adjust the direction of the rear wheels of the vehicle to avoid the rear wheels of the vehicle travelling over the tracks left by the front wheels (or in other cases to follow the tracks left by the front wheels). In either case, the control system may be configured to determine whether to control the rear wheel steering system to avoid the rear wheels travelling over the tracks left by the front wheels, the determination being made in dependence on one or both of the analysis of the tracks and the estimated surface characteristics. While the second embodiment is principally concerned with detecting the tracks left by the front wheels of the vehicle, other tracks beneath the vehicle, and in particular tracks already determined and analysed by the first embodiment, may also be considered. In this case, a refinement of the analysis already carried out based on the tracks ahead of the vehicle may be carried out using sensor data captured of the tracks beneath the vehicle. The sensors viewing the underside of the vehicle can be expected to be closer to the tracks than those viewing ahead of the vehicle, leading to higher resolution sensor data and enhanced analysis. This may identify errors in the analysis carried out by the first embodiment, which can be used to 8 generate feedback for the first embodiment to use on new tracks in the same driving surface. Further, since more information may be available from monitoring changes to the tracks over an extended period of time, by monitoring changes to tracks both while they are in front of the vehicle and beneath it (and, using the third embodiment described below, behind it), the changes can be monitored over a longer period of time, leading to more robust analysis. As with the first embodiment, the second embodiment does not consider tracks as a static feature of the terrain, but as a time-dependent feature having a structure which varies, either as the track is being formed (for example by the front wheels of the vehicle), or over time after it has been formed. As such, the control system is configured to analyse the detected tracks by determining changes to the tracks with respect to time. Several features of the tracks may be analysed in this way. For example, the control system may be configured to determine changes to the tracks with respect to time by determining a rate of change of depth of the tracks with respect to time, a rate of collapse of the tracks with respect to time, an angle of the sides of the tracks and a presence of water within the tracks. Again similarly to the first embodiment, the analysis of the tracks may take into account a weight of the vehicle, a force exerted on the ground at each wheel of the vehicle, a diameter and / or width for the wheels of the vehicle, a speed of the vehicle at the time the tracks were formed. In the second embodiment though, these parameters can generally be known more accurately than with the first embodiment, since the tracks are being formed by the vehicle itself (in relation to which its weight, tyre size and other characteristics may be well known). As with the first embodiment, these parameters may be used in an intermediate step of estimating a contact patch pressure for the front wheels of the vehicle. As with the first embodiment, the control system may be configured to estimate the one or more surface characteristics of the terrain surface more accurately by taking account of one or more other inputs, such as an ambient temperature, recent rainfall, and an expected terrain surface type. In addition to adjusting the vehicle driving parameters, the control system may also be configured to communicate the estimated surface characteristics to another vehicle, preferably in association with an indication of the geographical location corresponding to those surface characteristics, as per the first embodiment. Referring to Figure 6, at a step C1 an area of the terrain (driving surface) beneath and / or to the side of the vehicle (at or rearward of the front wheels of the vehicle, and forward of the rear wheels of the vehicle) is imaged using the sensors 120c, 120d, 120e. For the purposes of explanation the sensors 120c to 120e will be describes as cameras, but it will be understood that other sensing technologies may be used instead, as described elsewhere. The sensors 120c to 120e therefore capture an image of the terrain beneath the vehicle 10. At a step C2 the image is processed to extract the tracks. This could be achieved in many ways, as explained above in relation to Figure 5. Additionally, to the extent that Figure 6 is mainly concerned with tracks formed by the front wheels of the vehicle 10 itself, the location of the tracks within the field of view of the sensors 120c to 120e is known in advance, making their extraction from the sensor data more straightforward. At a step C3 the track data extracted from the sensor data is analysed. Various features may be extracted from the track data, such as (but not limited to) those discussed in relation to Figures 4A to 4D. This is repeated multiple times, to acquire track data over time. At a step C4, the track data at the multiple times are compared to identify changes in the tracks over time. More specifically, changes to each of the measured characteristics of the track with respect to time are identified. This may result in the determination of (for example) a rate of change of depth (depth change per second or minute for example), a rate of change of height (height change per second or minute for example), a rate of change of wall angle (number of degrees per second or minute for example), and a rate of fill of water (for example depth change per second or minute for example). In the example of Figure 6, it will be appreciated that a particular portion of track will only remain within the field of view of the sensors 120c to 120e for a relatively short time. However, in an off-road environment where the vehicle 10 is typically travelling relatively slowly, meaningful changes to the tracks may still be detectable over this time period. At a step C5, vehicle information and other inputs are obtained. As described previously, the vehicle information may include wheel size, vehicle mass, mass distribution (between front and rear wheels), tyre pressure, vehicle speed and vehicle acceleration or deceleration (at the time the tracks were formed by the vehicle). The other inputs may include environmental information such as ambient temperature, and recent rainfall. At a step C6, the track information (static and / or time variant), vehicle information and other information are analysed to identify one or more surface characteristics of the driving surface, and to classify the nature of the driving surface in which the track has been formed. For example, the driving surface (or a portion thereof) may be classified based on these parameters into one of a plurality of predetermined classes, each of which is considered to relate to a particular set and / or range of surface characteristics, and each of which is to be treated differently by the driver and / or vehicle, and each of which corresponds to a particular combination of surface characteristics. For example, considering four terrain types which may be available for classification (optionally along with others), these may be Type A, Type B, Type C and Type D. A first set and / or range of surface characteristics correspond to the Type A surface. Generally, these may be characteristics corresponding to dry sand. A second set and / or range of surface characteristics correspond to the Type B surface. Generally, these may be characteristics corresponding to wet sand. A third set and / or range of surface characteristics correspond to the Type C surface. These may be characteristics corresponding to dry grass. A fourth set and / or range of surface characteristics correspond to the Type D surface. These may be characteristics corresponding to wet grass. At a step C7, one or more driving parameters of the vehicle are set or adjusted (directly or indirectly) in dependence on the determined surface characteristics and / or in dependence on the classified surface type. In some implementations the driving parameters may be set directly based on the surface characteristics, but preferably the surface characteristics are used to classify the terrain type, and the classified terrain type is used to set or adjust the driving parameters. In parallel with the step C7, at a step C8, the surface characteristics and / or the terrain type are communicated externally of the vehicle, for example to another vehicle. The other vehicle is then able to notify its driver and / or set or adjust its own driving parameters when approaching or traversing the driving surface. It will therefore be appreciated that the second embodiment seeks to monitor the (change in) tracks under the vehicle as the front wheels of the vehicle itself pass over the terrain. This may in some implementations require the use of additional cameras (or other sensors) on the vehicle to those already present, in addition to the downward facing cameras on the wing mirrors, for example. Summary of the third embodiment (tracks behind) The control system of the third embodiment is configured to sense at least an area rearwards of the rear wheels of the first vehicle to obtain sensor data. In the case where the third embodiment is implemented (in practice by the same control system) along with one or both of the first and second embodiments, this means that the area being imaged by the technique of the third embodiment may have been previously imaged (and analysed) by the first embodiment and / or the second embodiment. The analysis carried out in each case may be similar. In any case, the control system is configured to determine, in dependence on the sensor data, tracks formed by the first vehicle in the sensed area of the terrain surface. This may be the tracks made by the front wheels of the first vehicle, tracks made by the rear wheels of the vehicle, or tracks made by both (that is, by the rear wheels passing along the track made by the front wheels). In some cases the determined tracks may be matched against the tracks determined in the first embodiment and / or the second embodiment, and compared. The control system is configured to analyse the determined tracks to estimate one or more surface characteristics of the terrain surface. Then, a surface signal is output, the surface signal being indicative of the estimated one or more surface characteristics. This may be used in a variety of ways. For example, the surface signal may be communicated to a second vehicle, for example a vehicle following the first vehicle in a convoy, to (for example) permit the second vehicle to adjust its own driving parameters to best deal with the terrain, and / or to notify the driver of the second vehicle of the nature of the terrain surface ahead. In the case of the surface signal being communicated to the second vehicle, the surface signal may comprise data relating to setting or modifying one or more driving parameters of the second vehicle in dependence on the estimated one or more surface characteristics. The surface signal may also be used internally by the first vehicle. In particular, one or more driving parameters of the first vehicle may be set or adjusted in dependence on the surface signal (and thus on the estimated one or more surface characteristics). Generally, the driving parameters adjusted by the third embodiment may be the same as those of the first embodiment and / or the second embodiment. It will be appreciated that, while with the third embodiment the surface characteristics being measured are in some senses obtained too late to inform the first vehicle of the nature of the specific portion of the terrain (since it has already been traversed), it is likely that the terrain ahead (in terms of surface composition for example) is similar, and so may have similar surface characteristics. Since the third embodiment has access to the most comprehensive set of high resolution sensor data of tracks left by the first vehicle itself (from both the front and rear wheels), the accuracy of the estimate still results in very useful information. Additionally, the surface characteristics obtained using the third embodiment represents the state of the surface behind the vehicle following its traversal by at least the first vehicle. The surface characteristics may in some cases be indicative that the driving surface behind the vehicle has been too damaged by the passage of the first vehicle. In this case, the non-trafficability of that portion of driving surface may be communicated to one or more other vehicles (so that they can avoid driving on it), and the driver of the first vehicle may be notified that the area behind is no longer-trafficable, to discourage them from reversing over it for example. In some cases the first vehicle may be inhibited from traversing this area. As with the first two embodiments, the third embodiment does not consider tracks as a static feature of the terrain, but as a time-dependent feature having a structure which varies, either as the track is being formed (for example by the rear wheels of the vehicle), or over time after it has been formed. As such, the control system is configured to analyse the detected tracks by determining changes to the tracks with respect to time. Several features of the tracks may be analysed in this way. For example, the control system may be configured to determine changes to the tracks with respect to time by determining a rate of change of depth of the tracks with respect to time, a rate of collapse of the tracks with respect to time, an angle of the sides of the tracks and a presence of water within the tracks. Again similarly to the first embodiment, the analysis of the tracks may take into account a weight of the vehicle, a force exerted on the ground at each wheel of the vehicle, a diameter and / or width for the wheels of the vehicle, a speed of the vehicle at the time the tracks were formed. As with the second embodiment, these parameters can generally be known more accurately than with the first embodiment, since the tracks are being formed by the vehicle itself (in relation to which its weight, tyre size and other characteristics may be well known). As with the first and second embodiments, these parameters may be used in an intermediate step of estimating a contact patch pressure for the front wheels of the vehicle. As with the first and second embodiments, the control system may be configured to estimate the one or more surface characteristics of the terrain surface more accurately by taking account of one or more other inputs, such as an ambient temperature, recent rainfall, and an expected terrain surface type. While the third embodiment is principally concerned with detecting the tracks left by the wheels of the first vehicle, other tracks behind the (rear wheels of the) first vehicle, and in particular tracks already determined and analysed by the first and / or second embodiments, may also be considered. In this case, a refinement of the analysis already carried out based on the tracks ahead of the vehicle and / or those beneath or to one side of the vehicle, may be carried out using sensor data captured of the tracks behind the vehicle. In this way, the same tracks can be analysed two or three times, by different sensor systems, at slightly different times, providing more accurate estimates of the surface characteristics as a result of the increased volume of data. To the extent that time-varying features of the tracks are being detected and analysed, it will be appreciated that the time-variation could extend across two or more of the three embodiments. That is, if a track is first detected and analysed by the first embodiment over a first time period, it may then be detected and analysed by the second embodiment over a second time period following the first time period, and it may then be detected and analysed by the third embodiment over a third time period following the second time period. The analysis of the time varying features of the tracks may therefore be assessed over a longer duration, and a more accurate estimate of the surface characteristics thereby received. Referring to Figure 7, at a step D1 an area of the terrain (driving surface) behind the vehicle (at or rearward of the rear wheels of the vehicle) is imaged using the sensors 120f, 120g. For the purposes of explanation the sensors 120f and 120g will be described as cameras, but it will be understood that other sensing technologies may be used instead, as described elsewhere. The sensors 120f an 120g therefore capture an image of the terrain immediately behind the vehicle 10, or immediately behind the rear wheels but under the vehicle 10. At a step D2 the image is processed to extract the tracks. This could be achieved in many ways, as explained above in relation to Figure 5. Additionally, to the extent that Figure 7 is mainly concerned with tracks formed by the rear wheels of the vehicle 10 itself, the location of the tracks within the field of view of the sensors 120f and 120g is known. At a step D3 the track data extracted from the sensor data is analysed. Various features may be extracted from the track data, such as (but not limited to) those discussed in relation to Figures 4A to 4D. This is repeated multiple times, to acquire track data over time. At a step D4, the track data at the multiple times are compared to identify changes in the tracks over time. More specifically, changes to each of the measured characteristics of the track with respect to time are identified. This may result in the determination of (for example) a rate of change of depth (depth change per second or minute for example), a rate of change of height (height change per second or minute for example), a rate of change of wall angle (number of degrees per second or minute for example), and a rate of fill of water (for example depth change per second or minute for example). At a step D5, vehicle information and other inputs are obtained. As described previously, the vehicle information may include wheel size, vehicle mass, mass distribution (between front and rear wheels), tyre pressure, vehicle speed and vehicle acceleration or deceleration (at the time the tracks were formed by the vehicle). The other inputs may include environmental information such as ambient temperature, and recent rainfall. At a step D6, the track information (static and / or time variant), vehicle information and other information are analysed to identify one or more surface characteristics of the driving surface, and to classify the nature of the driving surface in which the track has been formed. For example, the driving surface (or a portion thereof) may be classified based on these parameters into one of a plurality of predetermined classes, each of which is considered to relate to a particular set and / or range of surface characteristics, and each of which is to be treated differently by the driver and / or vehicle, and each of which corresponds to a particular combination of surface characteristics. For example, considering four terrain types which may be available for classification (optionally along with others), these may be Type A, Type B, Type C and Type D. A first set and / or range of surface characteristics correspond to the Type A surface. Generally, these may be characteristics corresponding to dry sand. A second set and / or range of surface characteristics correspond to the Type B surface. Generally, these may be characteristics corresponding to wet sand. A third set and / or range of surface characteristics correspond to the Type C surface. These may be characteristics corresponding to dry grass. A fourth set and / or range of surface characteristics correspond to the Type D surface. These may be characteristics corresponding to wet grass. At a step D7, one or more driving parameters of the vehicle are set or adjusted (directly or indirectly) in dependence on the determined surface characteristics and / or in dependence on the classified surface type. In some implementations the driving parameters may be set directly based on the surface characteristics, but preferably the surface characteristics are used to classify the terrain type, and the classified terrain type is used to set or adjust the driving parameters. In parallel with the step D7, at a step D8, the surface characteristics and / or the terrain type are communicated externally of the vehicle, for example to another vehicle. The other vehicle is then able to notify its driver and / or set or adjust its own driving parameters when approaching or traversing the driving surface. It will therefore be appreciated that the third embodiment seeks to monitor the (change in) tracks behind the vehicle using rearward cameras. Existing cameras, such as reversing cameras, could be used for this purpose, if necessary supplemented with additional cameras or other sensors. By viewing tracks left by the vehicle, it is possible to identify how well the driving surface recovers from the passage of the vehicle. For example, a rate or extent to which the material displaced by the passage of the vehicle wheels falls back into the tracks may be monitored. This may be used in a determination of trafficability of the driving surface, and whether the vehicle is able to reverse over the surface, or would be at risk of being stuck. The trafficability of the surface may also be provided to a vehicle behind (for example in a convoy of vehicles), to aid it in traversing that terrain. In some surfaces, the following vehicle may be guided to follow the ruts left by the vehicle in front, whereas for other surfaces (and / or track depths) the following vehicle may be guided to avoid those ruts and travel primarily on undisturbed ground. Considering the example of sand traversal, it is desirable to understand the trafficability characteristics of the sand driving surface. In particular, it is desirable to identify how much rolling resistance is expected from driving the vehicle on that surface (that is, the magnitude of the force resisting motion of the vehicle due to the characteristics of the surface), since vehicle characteristics such as a torque map or torque parameter may beneficially be set in dependence on the expected rolling resistance to aid with traversal of the driving surface. It is also desirable to identify an expected amount of degradation of that surface due to passage of the vehicle, since as the surface degrades it may cause traversal issues for the vehicle forming the tracks or a subsequent vehicle traversing the same driving surface. Again, this may lead to vehicle parameters of the vehicle (or another vehicle) being set or modified to aid in traversal of the terrain, or it may be used to warn the driver of possible traversal issues. Considering the example of clay surface traversal, it is desirable to be able to recognise a clay surface, since this is a challenge surface to traverse. When traversing a clay surface it is desirable to reduce slip by setting a low torque limit (to limit maximum torque). A clay surface may be identified from a track which is substantially or completely smooth. When a smooth track is detected, one or more driving parameters is therefore set or modified to aid with traversal, such as a torque map, or a torque limit. Similar principles may apply to a grass surface. In this case, a trail of mud (or churned grass and mud) running through a grass surface represents a heavily damages portion of the grass driving surface. In this case it is desirable to avoid the churned areas. In some implementations the driving surface may be mapped, with the mapping indicating trafficability of different portions of driving surface, and a vehicle controller being configured to present a driver with the trafficability mapping, or presenting the driver with recommended routes across the surface in dependence on trafficability. Similar principles to the above may be applied when the tracks are formed. In particular, as the tracks are being formed it is possible to acquire sensor data of this over time. In this case it is possible to determine how quickly the tyre of the vehicle sinks into the driving surface, a rate at which material is forced out of the tracks, and how far material or water is ejected from the tracks. This is particularly relevant to the second and third embodiments described above, in which case the imaging (or other sensing) is being carried out at short range, at a known position, and may specifically image the region around and behind the front and rear tyres respectively. This makes it possible to inform the system of the navigability and / or surface characteristics of the surface / terrain by (for example) monitoring the area immediately in front of and / or around the contact patch to predict failure modes as the vehicle progresses, and to pre-empt other characteristics of the surface when further pressure or torque from the wheel(s) is applied. The present technique makes it possible to improve control of the vehicle when travelling over deformable terrain types, such as sand or mud. By predicting the conditions ahead for the vehicle, and understanding the nature of the terrain to be traversed, systems on the vehicle (for example ride height and torque demand) can be adjusted so that the vehicle responds better to the terrain ahead. More specifically, the manner in which the surface behaves as the vehicle (or another vehicle at a similar time) is travelling over the surface is indicative of the nature of the surface and, hence, how the vehicle should best travel over it. This can be used to feedback automatically into vehicle control aspects. Generally, the present technique relies on determining the structure of tracks, or in the alternative on determining changes in the structure of tracks overtime. Examples of measurable parameters of the tracks include: (1) Depth of the tracks (below the plane of the driving surface) (2) Angle of the sides of the tracks (below the driving surface) (3) Angle of the sides of the tracks (above the driving surface) (4) Height of the portion of the tracks extending above the driving surface (5) Amount of material displaced from the indentations left by the wheels (6) Visibility of tread marks within the indentations (7) Amount of water within the indentations (8) rate of change of each of (1) to (7) above A surface type for the driving surface bearing the tracks may be identified by comparing one or more of the above measurable parameters with one or more threshold values. Each given surface type may correspond to a particular combination of parameter ranges for at least some of the parameters given above. At least some of such surface types may have one or more associated vehicle parameters which are or can be set (automatically, or subject to driver approval) to configure the vehicle to traverse that surface type. At least some of those surface types may also result in the driver (or a driver of another vehicle) being notified of the surface type and / or being provided with driving recommendations, such as to follow ruts, avoid ruts, to maintain at least a minimum speed or to remain below a maximum speed, or to not reverse or not attempt a particular traversal. The depth of the indentations formed in a driving surface by a vehicle (for example) is a function of not only the surface characteristics of the driving surface, but also the mass of the vehicle, the speed of the vehicle, the acceleration of the vehicle, and the weight distribution between the various wheels of the material. It follows that if the vehicle mass, speed, acceleration and mass distribution are known, and the depth of the indentations can be measured, the surface characteristics of the driving surface can be estimated. If only some of the vehicle mass, speed, acceleration and weight distribution are known, an estimate can still be made, but may be less accurate (but more accurate than if no vehicle information is available at all). The same principle can apply to each of the other measurable parameters indicated above, each of which are a function of not only the surface type, but of characteristics of the vehicle itself and how it is traversing the surface. Even without any knowledge of the vehicle, changes over time of the tracks following their formation may also give valuable insights into the surface characteristics of the driving surface. For example, the tracks may be monitored with respect to time, and a rate at which the tracks fill up with water determined. This is indicative of how waterlogged the ground is or the moisture content of the ground is. Using the cameras and / or other sensor sets on the vehicle this data can be processed to indicate surface deformation and clearances under / around the vehicle. This idea would allow the vehicle to understand these conditions and inform systems. The control system is configured to monitor progress relative to known surface geometry (rocks etc) and monitor deformable geometry (for example sand / mud) it can use the deformation data to predict characteristics such as surface sheer strength or rolling resistance allowing the vehicle to respond to the changing torque delivery requirements to maintain driver demand. The surface characteristics may be estimated in dependence on the detection of marks I tracks formed by parts of a vehicle (lead vehicle of the instant vehicle) other than the wheels. For example, if the bottom of the vehicle scrapes along the driving surface, this is indicative that the driving surface is not able to support the weight of the vehicle (at least not at the vehicle speed / acceleration applied by the vehicle when the tracks were formed). Further information on the characteristics of the surface / terrain may be inferred from the change in the tracks after they are made, for example sand flowing into ruts made by a vehicle, or mud falling back into freshly made mud ruts. Certain changes may happen relatively quickly, to the extent that they are observable over a time period in which the full length of the vehicle passes over the same point of the driving surface. Examples of such changes which occur after the track has been formed, may include water flowing into 14 the tracks (in the case of a waterlogged surface), and displaced sand flowing back into the tracks (in the case of a sandy surface). Examples of such changes which occur prior to the track being formed may include visible temporary deformation of the surface around (in front of, behind or to a side of) the wheel, such as waterlogged grass being dragged towards the wheel, or water being expelled under compression from the driving surface in the vicinity of the wheel. Figure 8 describes one implementation of the present technique, directed to the first embodiment. At a step E1, the driving surface identification system is activated. This may be when the vehicle is switched on, when the vehicle enters a specified driving mode, or upon manual driver selection. At a step E2 the system receives various sensor data, including from cameras, infra-red sensors, LIDAR, RADAR or V2X (vehicle to everything). At a step E3 the system looks for a vehicle ahead. If no vehicle ahead is found, then at a step E4 the process of Figure 8 pauses until one is found. If the vehicle ahead cannot be identified, then the process may return to the step E3. If the vehicle ahead is identified at the step E4, then at a step E5 it is determined whether the vehicle ahead is on a track being driven on by the instant vehicle. If not, then the vehicle ahead may not be useful in estimating the surface characteristics of the driving surface ahead of the instant vehicle, and so the process returns to the step E3. If it has been determined at the step E5 that the vehicle is on a track being driven on by the instant vehicle, then at a step E6 it is determined whether that vehicle is approaching or driving away from the instant vehicle. If approaching, then the vehicle ahead may not be useful in estimating the surface characteristics of the driving surface ahead of the instant vehicle, and so the process returns to the step E3. If the vehicle ahead is determined at the step E6 to be driving away from the instant vehicle, then at a step E7 the vehicle being followed is identified, and characteristics from the identified vehicle obtained from a vehicle characteristic database. The vehicle may be identified in various different ways, as will be known to the person skilled in the art. For example, the vehicle ahead may be imaged by a forward facing camera on the instant vehicle, and the image processed to determine the make and model of the vehicle ahead. Alternatively, the vehicle ahead may self-report its location, make and model (for example), this data being made available to the instant vehicle, either through direct vehicle-to-vehicle communication, or via a centralised (server based) function. In some cases it may not be possible to identify the vehicle ahead. In this case, this is determined at a step E8. If the identity of the vehicle can be determined, then surface characteristics of the driving surface (being) traversed by the vehicle ahead are estimated at a step E9 based on the tracks formed by or left by the vehicle ahead, and based on that vehicle’s contact patch pressure. The contact patch is the area of the vehicle tyres which is in contact with the driving surface, and the contact patch pressure is the pressure exerted on the driving surface by those contact patches. This can be calculated from the mass of the vehicle, the tyre size and pressure, the mass distribution of the vehicle (front and rear tyres may be subject to different pressure for example), and the speed and / or acceleration of the vehicle. If the identity of the vehicle cannot be determined, then the surface characteristics are still estimated, but at a step E10 and without the benefit of knowledge of vehicle characteristics (and thus the contact patch pressure). For example, a default average contact patch pressure may be assumed. The steps E9 or E10 may also take into account other factors, such as known information on the terrain being driven on (for example mud, sand, snow), weather information (temperature or rainfall for example), in each case being associated with location information, which the instant vehicle may use as a look up based on the current location of the instant vehicle. In either case, at a step E11 the estimated surface characteristics of the driving surface being traversed by the vehicle ahead are either or both of communicated to the driver of the instant vehicle and / or of other vehicles (by way of transmission from the instant vehicle, either directly to another vehicle, or to a centralised server where it is available to other vehicles), and / or used to (preferably automatically) set or adjust one or more vehicle driving parameters of the instant vehicle. Other characteristics of tracks which may be sensed and used to infer the surface characteristics of the driving surface may include water depth (in the depression of the tracks), mud depth, and water splashes (on the surface to either or both sides of the depression of the tracks), based on speed and relative to a contact patch location. It will be appreciated that, in some implementations the surface characteristics may be provided to a driver, or may be used to define a further metric or recommendation to be provided to the driver. The driver is then able to take this as advisory information on the nature of the driving surface, and may take this into account when driving on the surface (or choosing not to). For example, a recommended maximum (or minimum) speed may be suggested. That is, where a surface type Alternatively (or additionally) the surface characteristics may be used to automatically set or modify a vehicle parameter in a manner which optimises the operation of the vehicle for traversal of the surface. Crabbing In some cases, repeated traversal of the same portion of driving surface may be undesirable, for example if the surface becomes damaged each time a wheel passes over it. In some embodiments this type of surface may be identified from its surface characteristics, and the rear wheels steered to avoid passing over the same portion of ground as the front wheels. This embodiment requires the vehicle to have a rear wheel steering system. In other cases, having the rear wheels follow the tracks made by the front wheels may be desirable, such as resulting in a smoother ride, for example in cases where there is an underlying driving surface (e.g. a road) which would prevent the wheels from sinking too deep. I n this case the rear wheels may be steered to retain them in the tracks left by the front wheels. An example of this process is shown in Figure 9 At a step F1, the sensors scan the driving surface at or behind the front wheels of the vehicle. At a step F2 the tracks left by the front wheels are extracted (for example isolated) from the sensor data captured at the step F1. At a step F3 the track data is analysed, for example to determine the surface characteristics, and in this embodiment to identify the amount of damage caused to the driving surface by the front wheels. At a step F4 it is determined whether the rear wheels should be permitted to intersect or follow the tracks left by the front wheels, in dependence on one or both of the surface characteristics and the damage identified at the step F3. If it is determined that the rear wheels should not intersect with or follow the tracks, the rear wheel steering is controlled at a step F5 such that the rear wheels avoid the tracks left by the front wheels (or to reduce the extent to which the rear wheels move along or across the tracks). Otherwise, if it is determined that the rear wheels are permitted to follow or intersect with the tracks, no action is taken to use the rear wheel steering system to avoid this. Alternatively, in some circumstances it may be desirable for the rear wheels to follow (move in and along) the tracks left by the front wheels, for example to result in a smoother ride (repeatedly passing over the edges of tracks may give rise to a bumpy ride). This may be contingent on the driving surface having certain surface characteristics (for example that the vehicle will not sink too deep into the driving surface if the rear wheels further deepen the tracks left by the front wheels), and / or the damage to the driving surface caused by the front wheels being less than a threshold amount. I n this case, the step F4 instead determines whether the rear wheels should be guided to follow the track left by the front wheels, and the step F5 controls the rear wheel steering system to achieve this. Damage in this embodiment may be equated to track depth in some examples, although other metrics of damage might also be envisaged by the skilled person. Note that the use of rear wheel steering in these ways may feel unnatural to the driver, and so the driver may need to be notified, via a user interface / display or other notification (for example audio) that this is happening. The system may have access to a surface characteristic library. This may map particular combinations of measured track features (and changes over time), in combination with vehicle characteristics and other information, to one or more surface characteristics. In other words, the sensed track data and other parameters may be configured as inputs to a lookup table which outputs surface characteristics or types. The outputted surface characteristics or types are then used to alert a driver of a vehicle and / or to set or adjust vehicle parameters. The system may have access to a human / animal characteristic library. This may provide information required to determine a specific surface characteristic from given track information when the entity leaving the tracks is a human or an animal. It will be appreciated that a human will typically leave different tracks (for example shallower) than an animal such as a horse or camel. The system may have access to a vehicle characteristics library. This may indicate, for each of a plurality of vehicle types, characteristics such as the weight of the vehicle (or a weight identifier - for example heavy, medium and light for example), tyre width, usable in combination with the surface characteristic library. To use these libraries, the system first seeks to identify the nature of the entity (vehicle, animal or human) which has formed the tracks. This may be based on analysis of the sensor data. The human / animal characteristic library and the vehicle characteristic library may contain the information required to make this determination. As an example, when traversing a sandy driving surface, particularly off road, camel tracks might be more common than those of another vehicle. The system carries out initial identification of the entity forming the tracks based on the various libraries. Tracks are then identified, and the system monitors for changes in those tracks. The initial identification is carried out on the basis of a number of inputs including IR sensor, cameras, LIDAR, RADAR, e-Horizon, weather, V2X, temperature and automatic terrain response related data. 16 The system uses the vehicle / animal / human database to extract information (for example contact patch pressure - hoofs / shoes / tyres). This forms an input to the determination of surface characteristics. The surface characteristics database may store, as a function of the one or more surface characteristics, appropriate vehicle parameter settings. The track structure extracted from the sensor data may be processed in a number of ways. For example, for each measurement, such as track depth, track height, wall angle, water depth, splash location etc., an average may be taken over a distance of track (for example one metre). The average for each measurement may be used in estimating the surface characteristics, although it will be appreciated that a maximum or minimum value could be used instead, or additionally. Various surface characteristics maybe determined in this way, including a surface sheer strength and a rolling resistance. The wall angle of the tracks may be indicative of the water content of the terrain, particularly for sand. In some cases, the structure of the tracks, and even the changes to that structure over time, may not be adequate to accurately estimate the surface characteristics of the driving surface. In this case other parameters, such as temperature, rainfall etc. may be used to distinguish between multiple candidate surface types. For example, two different surface types may give the same type of track, but at different temperatures. In this case, knowledge of the ambient temperature may make it possible to distinguish, and thus more accurately determine the surface type. The additional data may also influence the changes to vehicle parameters. For example, navigational data may indicate whether the vehicle is currently operating on road or off road, as well as a width of the road upon which the vehicle is travelling. Weather data may indicate whether it is raining or snowing. At low temperatures, at which water is likely to be in a frozen state, the vehicle is unlikely to sink into the driving surface. Where the driving surface is currently covered in snow, water or sand, navigation data may also be able to inform the controller of what is beneath the covering. The controller may differently adapt the vehicle parameters dependent on whether a road surface is beneath the covering, or a mud or grass surface is beneath the covering. Another surface characteristic may be trafficability - a measure of how readily the surface can support the traversal of a vehicle. In some implementations a universal metric of trafficability may be used. This may be presented to the driver via a user interface to inform them about the driving surface ahead, enabling them to make a decision as to whether / how to proceed. In some cases the trafficability metric may be used to automatically configure one or more vehicle parameters. Various vehicle parameters may be adjusted in dependence on the estimated surface characteristics. For example, a throttle map for mapping accelerator pedal position to engine torque may be adjusted in dependence on the estimated surface characteristics. In particular, as a tyre rolls over a driving surface, how it generates grip as it rotates may be dependent on the surface characteristics of the driving surface. For example, if the surface characteristics are indicative of dry sand, the throttle map may be adjusted to rotate the wheels faster for a given pedal position. This is because on this type of surface a high degree of grip can be achieved with a high wheel speed. Alternatively, if the surface characteristics are indicative of wet grass, the throttle map may be adjusted to rotate the wheels more slowly for a given pedal position. This is because on this type of surface higher wheels speeds are likely to cause substantial slippage on the grass surface, resulting in damage to the grass and the surface becoming damaged. The surface characteristics may be defined as a predetermined set of surface classes, with the track analysis being used to categorise the terrain into one of those classes. The vehicle parameter adjustment is then made in dependence on the class into which the terrain has been categorised. Another vehicle parameter which may be adjusted is permitted wheel slip. In particular, the traction control system of the vehicle may be controlled by adjusting a permitted amount of wheel slip in dependence on the surface type. In a vehicle having an adjustable ride height (achieved via the vehicle suspension system), the ride height may be adjusted automatically in dependence on the estimated surface characteristics. This may be to raise the ride height to reduce the likelihood of the body of the vehicle dragging on the driving surface in the event that the surface characteristics suggest that the vehicle would be likely to sink relatively deeply into the road surface. The present technique may also be used in navigation, for example to avoid taking a route which traverses a road surface having estimated surface characteristics which are likely to be problematic for the vehicle, or which would result in another route (avoiding that driving surface) being preferable. In this case, the vehicle parameter may relate to controlling navigation. If the present technique identifies, for example, that a vehicle ahead has only just managed to successfully navigate the driving surface (based on the analysis of the tracks and knowledge of that vehicle), then if the instant vehicle is less capable on a driving surface of that type, the user interface may suggest to the driver that traversal of that surface not be attempted. By analysing the tracks behind the (rear wheels of the) vehicle, and comparing these to the tracks previously analysed between the front and rear wheels of the vehicle, it is possible to identify how effective the changes to the vehicle parameters have been in facilitating effective traversal of the driving surface. Similarly, tracks behind the vehicle and / or tracks between the front and rear wheels of the vehicle may be compared with those in front of the vehicle to identify how effective the changes made based on the tracks in front of the vehicle were. This information can be used to refine the algorithms used to control the adjustment of vehicle parameters and / or to refine the algorithms used to control the determination of surface characteristics. The vehicle parameter may be the selection of a driving mode (for example a normal road-driving mode, a sports road optimised for track use, or an off road driving mode) or a terrain mode (for example a sand-mode, a rock-crawl mode, a wading mode etc.) for the vehicle. Automatic selection of a terrain mode is known, but the present technique makes it possible to identify the surface characteristics of the terrain more accurately, and thus increases the reliability of terrain mode selection in choosing the most appropriate driving / terrain mode for the vehicle. When a vehicle is being driven on a water soaked grass surface, the grass surface ahead of a tyre can be observed being pulled towards the tyre. The grass mat will deform before breaking. Movement without breaking gives an indication of the strength of the grass mat. Accordingly, by imaging an area in front of one or more wheels of the vehicle as the surface is being traversed, an amount of elastic deformation, or a breakage, of the grass mat may be observed in the images. This can be used to limit or otherwise control torque delivery to avoid, or at least reduce, damage to the grass mat by the vehicle. The area in front of the front wheels, or the rear wheels, may be used for this purpose. The present technique may be used to monitor the accumulation of damage over time to the driving surface. Different analysis may for example be carried out in dependence on whether the tracks represent a first pass of a vehicle over that portion of the driving surface, or an accumulation (superposition) of multiple tracks over the same portion of the driving surface. For a first pass, consideration may be made to how much the surface deformed as the wheel passed over it, how quickly the track fills with water, mud or sand, how much material has been ejected laterally as the wheel formed the tracks, and how these factors relate to characteristics of the vehicle which formed those tracks (vehicle mass, contact patch pressure etc.), and the driving characteristics of that vehicle at the time the tracks were formed (speed, torque delivery etc.). For subsequent vehicle passes, consideration may be made to how the subsequently formed tracks interact with (for example deepen and / or widen) the previously formed tracks. A number of examples of observed characteristics (in the imaged tracks), corresponding surface classifications, and corresponding parameter adjustments will now be provided. In the case of smooth tracks I grooves being observed, the surface may be identified as wet clay, and a torque limit may be imposed as a parameter adjustment. In the case of a track depth being greater than a threshold depth and / or the presence of chassis marks on the terrain surface, the surface may be identified as soft terrain, with a high degree of sinking. In this case, a ride height of the vehicle suspension system may be adjusted to increase clearance and reduce drag. In the case of the sides of the tracks (either of the indentation formed in the surface, or of the ridges of expelled material to either side of the indentation), if an angle of repose is greater than a threshold value, in a sandy surface, a shear strength of the surface may be considered relatively high, whereas if the angle of repose is not greater than the threshold value, the shear strength of the surface may be considered relatively low. For a relatively high shear strength surface, a first sand throttle map may be used to control torque to maintain momentum through the sand. For a relatively high shear strength surface, a second (different) throttle map may be used instead. In the case of a grass surface replaced with mud in a track region, that region may be identified as damaged grass. Vehicle parameters such as applying a torque limit may be implemented (to prevent further damage) and / or guidance may be provided to the driver (or an autonomous driving controller) to avoid damaged (muddy) regions of the grass surface. It will be appreciated that various changes and modifications can be made to the present invention without departing from the scope of the present application.

Claims

1. A control system for a vehicle, the control system comprising one or more controller, the control system configured to:sense at least an area rearwards of the rear wheels of a first vehicle to obtain sensor data;determine in dependence on the sensor data, tracks formed by the first vehicle in the sensed area of the terrain surface; analyse the determined tracks to estimate one or more surface characteristic of the terrain surface; andoutput a surface signal indicative of the estimated one or more surface characteristic.

2. The control system of claim 1, wherein the surface signal indicative of the estimated one or more surface characteristic is output to asecond vehicle.

3. The control system of claim 2, wherein the surface signal comprises data relating to setting or modifying one or more driving parameterof the second vehicle in dependence on the estimated one or more surface characteristic.

4. The control system according to any preceding claim, comprising setting or modifying one or more driving parameters of the first vehicle in dependence on the surface signal.

5. The control system according to any preceding claim, configured to analyse the determined tracks by determining changes to the trackswith respect to time.

6. The control system according to claim 5, configured to determine changes to the tracks with respect to time by determining a rate of change of depth of the tracks with respect to time, or a rate of collapse of the tracks with respect to time.

7. The control system according to claim 6, wherein the changes are determined as the tracks are being formed by the rear wheels of thefirst vehicle or subsequently to the tracks being fully formed by the rear wheels of the first vehicle.

8. The control system according to any preceding claim, wherein the analysis of the tracks takes into account a weight of the first vehicle,or a force exerted on the ground at each wheel of the first vehicle, or a diameter and / or width for the wheels of the first vehicle or a speed of the first vehicle at the time the tracks were formed.

9. The control system according to claims 8, configured to analyse the tracks by estimating a contact patch pressure for the wheels of thefirst vehicle.

10. The control system according to any preceding claim, configured to analyse the tracks by identifying one or more of a depth of the tracks, an angle of the sides of the tracks and a presence of water within the tracks.

11. The control system according to claim 10, configured to analyse the tracks by identifying a change with respect to time of one or more of the depth of the tracks, an angle of the sides of the tracks and a presence of water within the tracks.

12. The control system according to claim 3 or claim 4, wherein the driving parameters set in response to the determined surface characteristics comprise one or more of a throttle map for the vehicle, a wheel slip parameter, and a ride height for a suspension system of the vehicle.

13. The control system according to any preceding claim, further configured to sense an area ahead of a vehicle and / or beneath the vehicle to obtain sensor data, to determine, from the sensor data, tracks ahead of and / or beneath the vehicle in a terrain surface, to analyse the determined tracks ahead and / or beneath the vehicle to estimate one or more surface characteristics of the terrain surface, and to set or modify one or more driving parameters of the vehicle in dependence on the estimated surface characteristics from the tracks ahead and / or beneath the vehicle, wherein20the control system is configured to refine the estimated surface characteristics and / or the one or more driving parameters established based on the tracks determined ahead of and / or beneath the vehicle based on the analysis of the tracks behind the vehicle.

14. A vehicle comprising the control system according to any preceding claim, and one or more sensors configured to image the area behind 5 the first vehicle to generate the sensor data.

15. A control method for a vehicle, the method comprising:sensing at least an area rearwards of the rear wheels of a first vehicle;determining tracks made by the vehicle in a terrain surface within the sensed area;10 analysing the determined tracks to estimate one or more surface characteristics of the terrain surface; andoutputting a surface signal indicative of the estimated one or more surface characteristic.

16. Computer software that, when executed, is arranged to perform a method according to claim 15.15

Images