WO2010132014A1

WO2010132014A1 - Method for calibrating a hinge angle sensor on a vehicle and a vehicle

Info

Publication number: WO2010132014A1
Application number: PCT/SE2010/050517
Authority: WO
Inventors: Johan Larsson; Michael Krasser
Original assignee: Atlas Copco Rock Drills Ab
Priority date: 2009-05-14
Filing date: 2010-05-11
Publication date: 2010-11-18
Also published as: SE0950339A1; SE533777C2

Abstract

The present invention refers to an arrangement and a method for calibrating a hinge angle sensor (110) on a vehicle (100) comprising a first part (100b) and a second part (100a) connected by a hinge (107) provided with a hinge angle sensor (110) for measuring a hinge angle (α) between the first part (100b) and the second part (100a). In accordance with the invention, the following steps are involved: placing the vehicle (100) so that the hinge angle (α) is roughly 0° and a hypothetical sight line (120) points towards an object (124), in which connection the hinge angle sensor (110) indicates a measured hinge angle; beaming of a medium from a beam-generating device (121) so that the medium generates an image; projection of the image on the object (124); determination of the first position of the image(125) on the object (124); movement of the vehicle (100) towards or away from the object (124) so that the hinge angle sensor (110) continues to display the same measured hinge angle; determination of the second position of the image(126) on the object (124); and adjustment of the zero position of the hinge angle sensor (110).

Description

CALIBRATION METHOD AND VEHICLE

TECHNICAL AREA

The present invention refers to a method for calibrating a hinge angle sensor, a method for calibrating a distance scanner and a vehicle arranged for these methods.

THE PRIOR ART

In many areas, constant work takes place to improve efficiency, productivity and safety. One of these is underground mining. One of the areas in connection with such mining in which changes/improvements are implemented is to automate the objects used in the mine. For example, it is desirable for the above to consist of certain functions being carried out autonomously, i.e. vehicles should not only be drivable without a driver, they should also be able to carry out functions entirely autonomously.

One example of such a vehicle in which automated operation is desirable is an LHD (load/haul/dump) vehicle. These vehicles are often used to remove mined rock and haul it to a specific location where the removed rock is dumped, after which the vehicle returns to the same location to be reloaded. These vehicles thus often carry out the same manoeuvre again and again, which means that this type of vehicle/manoeuvre is particularly well suited for automation.

These vehicles have previously been driven manually by a driver on board the vehicle or using radio control, for example. On account of factors such as driver safety, risk of accident and labour costs, it is desirable, however, to be able to drive such load vehicles fully automatically.

One type of existing system (see, for example, WO2007/012198) for achieving such fully automatic operation is based on a three-step principle in which the vehicle, in a route teaching step, is first driven manually along the route it will subsequently drive autonomously, while signals from various sensors arranged on the vehicle are stored. Step two involves route profiling, in which, based on at least part of the above stored sensor signals, a coordinate system is created, covering the area in which the vehicle is to be driven. The route driven by the vehicle when the route is taught is described in this coordinate system with information on, for example, the suitable speed for various parts of the route.

The third step consists of playback, in which the coordinate information on how the vehicle was driven manually and a representation of the environment are used to drive the vehicle autonomously along the route along which the vehicle was driven manually in step one.

During autonomous playback (tramming) of a route, the position of the vehicle is determined, for example via estimation, in the coordinate system in which representation of the environment and the desired route are defined.

The representation of the environment used in connection with determining the position may consist of a map representation of the walls in the tunnels in which the vehicle is to be driven autonomously and may either be generated in advance or generated using the sensor information acquired, as described above. When the route profiling has been completed, a representation may thus have been generated.

Regardless of how the representation of the environment has been generated, in a system of the above type it is very important for it to be a correct description of the environment as, if for any reason it is a poor representation of what the environment actually looks like in corresponding parts of the route, there is a risk of vehicles getting lost or driving the wrong way and causing damage to vehicles or the environment.

The processes used to generate such representations of the environment can be complicated and much can go wrong. As mentioned above, the map generation may be based on sensor information from many different sensors. Examples of such sensors include waist angle sensors or other hinge angle sensors, sensors for measuring distance travelled and one or more laser distance scanners or other distance scanners. One of the problems with this system is that it is sensitive to how well the hinge angle sensor is calibrated. If the vehicle is to follow the taught route with a high level of precision, it is important for the hinge angle sensor to be correctly calibrated, i.e. for the hinge angle to be known at which longitudinal displacement of the vehicle does not result in lateral displacement or a change of the angle between the lateral centre line of the vehicle and an external reference. The hinge angle sensor is best calibrated so that its zero position coincides with the stated hinge angle. The precision requirement to allow vehicles to follow a taught route at speeds in the order of 4-5 m/s or more with constant route tracking precision is for the hinge angle sensor to be calibrated with a tolerance of in the order of a few tenths of a degree. An example of a specific tracking algorithm is that a calibration error of 1° is equivalent to 25 cm lateral offset on tracking at 5 m/s. The corresponding figure at 8.3 m/s is 0.7 m as the offset increases exponentially. This precision requirement is much stricter than the precision requirement for a standard LHD vehicle that is driven manually.

In the same way, it is sensitive to how well the distance scanners are calibrated. The precision requirement here is for a tolerance of in the order of a few tenths of a degree.

Therefore, there is a need for a method for calibration that meets these high precision requirements.

DESCRIPTION OF THE INVENTION

One aim of the present invention is to provide a method for calibrating a hinge angle sensor with a high level of precision. This aim is achieved with a method characterised as described in claim 1.

The present invention refers to an arrangement and a method for calibrating a hinge angle sensor on a vehicle, which vehicle includes a first part with a first centre line and a second part with a second centre line. The first part and the second part are connected by a hinge, which hinge is provided with a hinge angle sensor for measuring a hinge angle between the first part and the second part. In accordance with the invention, the method involves the following steps. Placing the vehicle so that the hinge angle is roughly 0° and a hypothetical sight line points towards a first object. The hypothetical sightline passes between a beam-generating device arranged in the first part at a first distance from the first centre line, and an obstacle arranged on the second part at the first distance from the second centre line. The hinge angle sensor then indicates a measured hinge angle. Beaming of a medium from the beam-generating deive, so that the medium partially passes the obstacle and the medium is partially stopped by the obstacle, so that the medium generates an image. Projection of the image on a first object. Determination of the first position of the image on the first object. Movement of the vehicle a length towards or away from the first object so that the hinge angle sensor continues to display essentially the same measured hinge angle. Determination of the second position of the image on the first object. Adjustment of the zero position of the hinge angle sensor, when a second distance between the first position of the image and the second position of the image is greater than the second distance equivalent to the minimum acceptable error in the hinge angle measurement for the movement performed.

The advantages of this method are that precise calibration is achieved in a fast, reliable manner and this can be done by a single operator.

According to one embodiment of the invention, it also includes an arrangement and a method for calibrating a distance scanner on a vehicle, which vehicle includes a first part and a second part connected by a hinge, which hinge is provided with a hinge angle sensor for measuring a hinge angle between the first part and the second part. In accordance with the invention, the method involves the following steps: Calibration of the hinge angle sensor. Placing the vehicle so that the hinge angle sensor indicates the zero position. Generation of an image in a sight line. Projection of the image on a second object. Adjustment of the zero position of the distance scanner, so that the second object is at the zero position of the distance scanner.

The advantages of this embodiment are that precise calibration is achieved in a fast, reliable manner and this can be done by a single operator. DESCRIPTION OF DRAWINGS

The invention will now be explained in further detail using preferred embodiments and with reference to the attached drawings, in which:

Figs. IA-C show a vehicle from the side and from above, in connection with which the present invention can advantageously be used.

Fig. 2 shows an example of a mine in which the present invention can advantageously be applied. Figs. 3A-B show an arrangement for calibrating a hinge angle sensor. Fig. 4 shows a flow chart for calibrating a hinge angle sensor. Figs. 5A-C show an arrangement for calibrating a distance scanner. Fig. 6 shows a flow chart for calibrating a distance scanner.

PREFERRED EMBODIMENT

Figs. IA-C show a vehicle 100 from the side and from above. The vehicle 100 is a load vehicle to which the present invention can advantageously be applied. The vehicle 100 includes a scoop 101, wheels 102-105 connected to axles 113, 114, and a control unit 106, which controls various of the functions of the vehicle 100. As shown in Fig. IB, the vehicle consists of an articulated vehicle, whereby a front part 100a is joined to a rear part 100b via a hinge 107. The hinge 107 can be set to the desired hinge angle α. A hinge angle sensor 110 is arranged at the hinge 107 and measures the current hinge angle α and transfers these signals to the control unit 106. The vehicle may also include, for example, a wheel rotation sensor 108 such as an odometer, which may be arranged in connection with the shaft proceeding from the transmission and emit signals representing the rotation of the drive wheels and/or the distance travelled. Moreover, the vehicle 100 includes an engine or other device (not shown) to propel the vehicle 100.

The vehicle 100 also includes a front 111 and a rear 112 laser distance scanner, each in a laser housing, which laser distance scanners 111, 112 are also connected to the control unit 106 and emit sensor signals representing distance measured, i.e. the distance to the nearest obstacle that stops the path of the laser beam. The laser distance scanners 111, 112 may, for example, be arranged to measure the distance in certain directions at an angular interval. In the present example, laser distance scanners are used to measure the distance to the nearest object in the longitudinal direction of the front part 100a forwards (or in the longitudinal direction of the rear part 100b backwards) and the distance to the nearest object (such as rock) for each full degree ± 90° from each longitudinal direction. Each laser distance scanner thus measures distance at 181 measuring points. Laser distance scanners that measure distance in considerably more directions may, of course, be used, as well as those that measure distance in considerably fewer directions. Likewise, just one omnidirectional laser may be used instead. In an alternative example embodiment, only the scanner that is currently pointing in the direction of travel is used (i.e. the front scanner 111 if the vehicle is driving forwards and vice versa). Nor is it essential for the invention for the directions to be measured via laser distance scanners. Any distance scanners may be used, provided that they can supply distance measurements with an acceptable level of precision. Examples of other types of possible distance scanner include those based on radar or sonar technology.

Moreover, in the example embodiment shown here, a distance scanner is used that measures distance in just one plane (the horizontal plane of the vehicle). However, distance scanning may take place in more than one plane, for example also in a vertical plane to measure tunnel/gallery height or in other planes between the horizontal and vertical planes, thus refining the opportunities for correct position estimation. In another alternative example embodiment, it is possible instead or in addition to use one or more scanners that are directed towards the sides.

Moreover, the above sensors emit sensor signals to the control unit 106 at relevant times, for example continuously or every 40 ms or more frequently or more rarely. The control unit 106 then uses signals received as will be described below.

Fig. 2 shows an example of a mine in which the present invention can advantageously be applied. In the example shown, the vehicle 100 uses the scoop 101 to load rock mass at location A and then hauls the mass loaded for dumping at location B. When the vehicle 100 needs to be set up for autotramming, the three-step principle described above can, for example, be applied, i.e. route teaching takes place first, with the activation of storage of signals from the above sensors.

The load-haul-dump-return method can be arranged to be performed as a single route. However, travel from A to B is usually performed as a first separate route and travel from B to A as a second separate route. Sensor signal storage is thus activated to teach a route from A to B. An operator with the vehicle located at point A reverses towards point C to turn the vehicle; then the vehicle is hauled along the dotted line to point B where the route teaching is stopped.

Based on the sensor signals stored, the route is then created, i.e. how the vehicle is to be driven and the speed at which the vehicle should be driven in different parts of the route. As mentioned above, the sensor signals can be read off every 40 ms, for example. If each sensor signal reading were a route point, the number of route points would be very large. For this reason, the route points can instead consist of signals determined at a position every 50 cm, for example, the vehicle has travelled. The data stored for the route are preferably the position, the direction of the vehicle and the desired speed. This produces a route that, in principle, consists of a number of points. Each point thus indicates where the vehicle should be, the direction it should have and the speed at which it should be driven in the subsequent autotramming.

However, when the vehicle subsequently travels along the route autonomously, this information is not normally sufficient to allow the desired travel to take place, as mentioned above, for example because uncertainties in the sensor signals mean that the end position will, with a high degree of probability, deviate from the calculated position, with the result that the start position of the next route will also deviate from the original position. For this reason, a representation of the environment such as route maps is also used to allow a comparison of signals measured during autotramming with the map to allow more reliable determination of the position of the vehicle and restore uncertainties in the estimated position at regular intervals. The representation of the environment (route maps) may, for example, consist of a coordinate system which may advantageously be local for the specific route and may also be created on the basis of the sensor signals stored.

Therefore, the coordinate system only needs to cover the area in which the vehicle is to be driven and may have its origin where the point on the vehicle that constitutes the reference in connection with positioning, such as the centre of the front axle of the vehicle, is located when storage begins.

The route driven by the vehicle when the route is taught can then be described in this coordinate system with information on, for example, the suitable speed for various parts of the route.

When the route profiling has been completed, a representation corresponding to that shown in Fig. 2 may thus have been generated. This can then be used in connection with subsequent route playback. Therefore, it is possible to say that, after the map has been generated, the state of each square on the map is a function of the measured distance of all laser beams that have struck/passed through the square during the measurements.

The map can be represented by a relatively fine-meshed grid, for example with a resolution of 1 cm or 10 cm/square.

As mentioned above, one problem with such maps, whatever their type, is that the environment in which the route is taught may be such that the quality of the map is poor, for example on account of objects or surfaces that absorb or reflect light away from the laser distance scanners. Moreover, map generation is based on sensor information from many different sensors such as hinge angle sensors, sensors for measuring the distance travelled and laser distance scanners, and in some cases also information from a gyro that measures the direction of travel of vehicles. If one or more of these sensors produces incorrect information, the map is probably also incorrect. Detailed geometric information on the vehicle design and sensor locations on vehicles is also required for the maps to be a relevant representation of the environment. During the development of the system, a manual method was used to calibrate the hinge angle sensor 110. This method involves a person standing on the 'bumper' behind the radiator of the vehicle 100 and creating a sight line over the centre of the laser housings 111 , 112. By observing how this sight line moves relative to any distant, stationary object when the vehicle 100 is driven a distance of 20-30 m, a measure is obtained of whether the vehicle is in line with the hinge, i.e. if the hinge angle is in its correct zero position, the previously determined sight line is fixed on the same point on the distant object.

If, on the other hand, the hinge 7 is slightly angled, the vehicle 100 will turn and the sight line will no longer point to the same point when the vehicle 100 is driven forwards or backwards. By using this sight line method to evaluate how the vehicle 100 turns in different hinge angle positions, the position corresponding to the vehicle 100 travelling straight forwards can be determined, i.e. the angle to be regarded as the zero position of the hinge angle.

This method has several shortcomings, the most important of which are as follows: 1) In some countries, there are rules that do not permit people to stand on or hang onto the vehicle while it is being driven. 2) The method requires at least two people, a driver and a person who stands and assesses the movement of the sight line. 3) Precision depends on how well the person reading the sight line can assess how vehicles turn. 4) The method is very time-consuming as the vehicle operator and the sighter must communicate between each test on how the hinge angle should be changed.

Figures 3 A and B show an arrangement for an automated variant of this method. We imagine, cf. Fig. 1C, that there is a lateral centre line 115 for the first part 100b and a lateral centre line 116 for the second part 100a, where a centre line is a line that passes midway between the wheels 103, 105, 102, 104 on the respective parts 100b, 100a and is perpendicular to the axles of the respective parts 113, 114. If the hinge angle α is 0°, these centre lines 115 and 116 will coincide or be parallel. This knowledge is utilised in the embodiment of the invention below. In the example in Figures 3 A and B, laser housings 111 and 112 are arranged on the centre lines 1 15, 1 16 (cf. Fig. 1C), for which reason the arrangement of the laser housings 111, 112 is beneficial. A split laser 121 is mounted on the first laser housing 112 on the first part 100b, and an obstacle 122 in the form of a panel or similar with a hole 123 is mounted on the second laser housing 111 on the second part 100a. Preferably, the hole 123 is a vertical slit that is, for example, approximately 10 mm wide and approximately 5-10 cm high. A hypothetical sight line 120 passes the split laser 121 and the obstacle 122.

The split laser 121 is designed so that it generates a laser line in a horizontal plane and has attachments so that it can be attached to the first laser housing 112 so that the centre of the split laser 121 is located in the centre line 1 15 of the first part 100b. The invention naturally also works if the split laser 121 and the obstacle 122 are arranged on a part 100b, 100a each, in places along the centre lines 115, 116 other than on the laser housings 111, 112, or are arranged in parallel at equal distances from the centre lines 115, 116.

It is also possible to use types of light other than split lasers. In the same way, it is conceivable to use a medium other than light, for example other electromagnetic waves, preferably radar. Moreover, it is conceivable to use sound, preferably based on sonar technology, or some type of particle beam such as water or dye.

When the split laser 121 beams forwards, the obstacle 122 will partially obscure the light, except in the slit 123. This produces a spot laser, the spot of which can be projected as an image on a first object 124 for calibration of the hinge angle sensor 110.

Naturally, a projected image other than a spot can be used, for example a vertical line or a more complicated image. It would also be possible to create a projection not through a hole but, for example, along an edge or on both sides of a pin. The word 'image' should be interpreted broadly and shall also include a pattern generated, for example, with sound or in some other way. The first object 124 can be anything on which the image is visible. It is easiest if the first object 124 is stationary.

It would also be possible to calibrate the hinge angle sensor 110 by selecting a property of the image dependent on the hinge angle other than its position on the first object 124.

The method for calibration of the hinge angle sensor is then, cf. Fig 4, that the vehicle 100 is placed so that the hinge angle α is roughly 0° and the sight line 120 points towards the first object 124, step 210. In this situation, the hinge angle sensor 110 indicates a measured hinge angle that should be roughly 0°. The spot laser 121, 122 generates an image in the form, for example, of a spot, step 220, that is projected on the first object 124 in a first position 125, step 230. The first position 125 of the image is determined, step 240. This can be done manually, for example by remembering where the first position 125 of the image is or by marking the position with a pen, for example. This can also take place automatically by the first object 124 being sensitive to light or another medium that reaches the first object 124 so that, for example, the first position 125 of the image is stored electronically.

The vehicle 100 is then moved towards or away from the first object 124 perhaps 20-30 metres without actively changing the hinge angle on the vehicle so that the hinge angle sensor 1 10 at all times essentially continues to demonstrate the same measured hinge angle, step 250. The image is then, after the vehicle has been moved, in a second position 126, which second position 126 is determined, step 260. If the hinge angle α is now in its perfect zero position, the second position 126 is the same as the first position 125.

However, if the hinge 107 is slightly angled, the vehicle 100 will turn slightly and the sight line 120 will no longer point to the same place on the first object 124, i.e. the image has moved to a second position 126. The distance between the first position of the image 125 and the second position of the image 126 corresponds to the error in the previous hinge angle calibration. The direction in which the image has moved indicates the direction in which calibration must take place. The zero position of the hinge angle sensor 110 can then be adjusted, step 270, for example by changing the offset in the measurement. Normally, if the image on the first object 124 moved to the right, viewed from the vehicle, when the vehicle was driven forwards, then the vehicle has turned clockwise and the actual zero position of the hinge angle is then located at an angle anticlockwise from the hinge angle measured during movement.

If the first object 124 can store the first position of the image 125 and the second position of the image 126, calibration can take place automatically, for example using the control unit 106. Otherwise, an operator can calculate manually how much and in which direction the zero position of the hinge angle sensor 110 should be adjusted.

After calibration of the hinge angle sensor 110, the test should be carried out again to check that it is now correctly calibrated. Otherwise, the method should be repeated until adequate precision has been achieved. The vehicle can either be driven forwards and backwards in the same area or it can continue in the same direction with adjustment of the hinge angle between each length driven. The hinge angle sensor 110 should be calibrated so that the error is less than 0.5° and preferably less than 0.1°.

The advantages of calibrating the hinge angle sensor 110 in this way are that only one operator is required to carry out the calibration. The calibration is also faster as the operator does not need to communicate with others. The calibration is also more precise as a quantitative measure of the calibration precision is obtained.

A well-calibrated hinge angle sensor 110 can also be used to calibrate the laser distance scanners of the vehicle 111, 112 or other distance scanners, for example radar or sonar scanners. To make a correct map of the environment, it is necessary also for the distance scanners 11 1 , 112 to be mounted in the correct location and at the correct angle in relation to a reference point and the direction of the vehicle 100. It is relatively easy to measure with satisfactory precision that the distance scanners 111, 112 are mounted in the correct location, while it is considerably more difficult to measure whether the distance scanners 111, 112 are mounted in the correct direction. The precision requirement for the direction of the distance scanners 1 1 1 , 1 12 in relation to the reference direction of the vehicle 100 is in the order of a few tenths of a degree. The reference direction should be chosen to be parallel to or coincident with the centre line

115, 116.

During the development of the system, a manual method was used to verify that the distance scanners were mounted in the correct direction. This method involves two people holding wooden panels in the measurement plane of the respective distance scanners 111, 112 while data from the distance scanners 111, 112 is stored in a file. The wooden panels are held so that they form corners corresponding to the corners of the vehicle 100 but displaced in parallel up to the measurement plane of the distance scanners 111, 112. By analysing the data stored in the file mentioned above to find out whether the corners registered in the data appear symmetrical, it is possible to obtain a measure of whether the distance scanners 111, 112 are mounted in the correct direction.

One of the problems with this method is that several people need to help. The precision also depends on how well those who hold the wooden panels are able to hold them in a position that produces a correct image of the corners of the vehicle 100 in the laser plane. There is also great uncertainty in the measurement as it is based on the corners of the vehicle 100 being symmetrically located in relation to the lateral centre line of the vehicle 100.

Figs. 5A-C show an arrangement for calibrating the front distance scanner 111. A sight line needs to be known to make it possible to calibrate the distance scanner. In this case, the sight line 120 should be a hypothetical line that passes through the distance scanners 111, 112 that are to be calibrated and that coincides with or is parallel to the centre line 115, 116 on the part 100a, 100b on which the distance scanner 111, 112 is arranged. One easy way to find the sight line is to use the method described above for the hinge angle sensor 110, although other methods are, of course, also available. Therefore, the hinge angle sensor 110 should preferably be calibrated first, step 310 in Fig. 6, before the distance scanner 111 is calibrated.

The same equipment can be used to calibrate the distance scanner 111 as is used to calibrate the hinge angle sensor 110, i.e. a split laser or similar 121 and an obstacle in the form, for example, of a panel or similar 122 with a slit or similar 123, which together function as a spot laser or similar.

A second object 140, preferably in the form of a vertical bar, is also used. It is also conceivable for the second object 140 to be a large sheet, for example, with a vertical hole in it. The important element in this context is to have a second object 140 that makes it possible, in a simple way, to detect a difference in distance between the second object 140 and the environment in such a way that the sight line 120 can be detected. For example a vertical edge can thus also function as a second object 140.

The method for calibrating the distance scanner 111 is then to place the vehicle 100 so that the calibrated hinge angle sensor 110 indicates the zero position, step 320. A spot or another image 143 is generated using the split laser 121 and the obstacle 122 in the sight line 120 in the same way as described above, step 330. The image 143 is then an indication of where the sight line 120 lies and the line along which the distance scanner 1 11 should be calibrated. The second object 140 is placed in the sight line 120 by ensuring that the image 143 is projected on the second object 140, step 340. The second object 140 should be placed approximately 10-20 m in front of the vehicle.

The distance scanner 111 then measures the distance to the object within an area 141. If the distance from the distance scanner to other objects 142 in the proximity of the sight line 120 is greater than to the second object 140, a clear indication of the position of the second object 140 is produced in the data from the distance scanner 1 1 1 and the direction of the distance scanner 1 1 1 in relation to the sight line 120 can thus be determined. This may, for example, be done using an offline analysis by measured laser data being displayed on the display of the vehicle or via an automatic analysis of the data initiated by the operator. The distance scanner 111 can then be adjusted manually or automatically, for example by changing its offset so that its zero position coincides with the sight line 120, step 350. The distance scanner 111 should be calibrated so that the error is less than 0.5° and preferably less than 0.3°.

The rear distance scanner 112 can be calibrated in a corresponding manner by changing the location of the split laser 121 and the obstacle 122. The advantages are that one person is sufficient to carry out the measurement. Calibration is also faster as it can be carried out directly by the operator without offline analysis of laser data. In addition, this method produces higher precision.

Although the description above has only concerned the calibration of hinge angle sensors and distance scanners on self-propelling mine vehicles, the invention can, of course, also be used in other vehicles.

The invention is, of course, not limited to the example described above. It can be modified within the framework of the attached claims.

Claims

I) A method for calibrating a hinge angle sensor (110) on a vehicle (100), which vehicle (100) includes a first part (100b) with a first centre line (115) and a second part (100a) with a second centre line (116), which first part (100b) and second part (100a) are connected by a hinge (107), which hinge (107) is provided with a hinge angle sensor (110) for measuring a hinge angle (α) between the first part (100b) and the second part (100a), characterised by the following steps:

- (210) placing the vehicle (100) so that the hinge angle (α) is roughly 0° and a hypothetical sight line (120) points towards a first object (124), which hypothetical sight line passes between a beam-generating device (121), arranged on the first part (100b) at a first distance from the first centre line (115), and an obstacle (122), arranged on the second part (100a) at the first distance from the second centre line (116), whereby the hinge angle sensor (110) indicates a measured hinge angle; - (220) beaming of a medium from the beam-generating device (121), so that the medium partially passes the obstacle (122) and the medium is partially stopped by the obstacle (122), so that the medium generates an image;

- (230) projection of the image on the first object (124);

- (240) determination of the first position of the image (125) on the first object (124); - (250) movement of the vehicle (100) a length towards or away from the first object

(124), so that the hinge angle sensor (110) continues to display essentially the same measured hinge angle;

- (260) determination of the second position of the image (126) on the first object (124); and - (270) adjustment of the zero position of the hinge angle sensor (110) when a second distance between the first position of the image (125) and the second position of the image (126) is greater than the second distance equivalent to the minimum acceptable error in the hinge angle measurement for the movement performed.

2) A method for calibrating a hinge angle sensor in accordance with claim 1, characterised in that the first distance is zero. 3) A method for calibrating a hinge angle sensor in accordance with claim 1 or 2, characterised in that the medium is one of the following: electromagnetic waves (preferably laser or radar); sound (preferably sonar); particles (preferably water or dye).

4) A method for calibrating a hinge angle sensor in accordance with any of the claims 1- 3, characterised by calibration until the error in the zero position is less than 0.5°, and preferably less than 0.1°.

5) A method for calibrating a distance scanner (121, 122) on a vehicle (100), which vehicle (100) includes a first part (100b) and a second part (100a) connected by a hinge

(107), which hinge (107) is provided with a hinge angle sensor (110) for measuring a hinge angle (α) between the first part (100b) and the second part (100a), characterised by the following steps:

- (310) calibration of the hinge angle sensor (110) in accordance with any of the claims 1-4;

- (320) placing the vehicle (100) so that the hinge angle sensor (110) indicates the zero position;

- (330) generation of an image in a sight line (120);

- (340) projection of the image on a second object (140); and - (350) adjustment of the zero position of the distance scanner (121, 122) so that the second object (140) is at the zero position of the distance scanner (121, 122).

6) A method for calibrating a distance scanner in accordance with claim 5, characterised by calibration until the error in the zero position is less than 0.5°, and preferably less than 0.3°.

7) A vehicle comprising a first part (100b) with a first centre line (115) and a second part (100a) with a second centre line (116), which first part (100b) and second part (100a) are connected by a hinge (107), which hinge (107) is provided with a hinge angle sensor (110) for measuring a hinge angle (α) between the first part (100b) and the second part (100a), characterised in that the vehicle (100) includes: a beam-generating device (121) arranged on the first part (100b) at a first distance from the first centre line (115), an obstacle (122) arranged on the second part (100a) at the first distance from the second centre line (116), which beam-generating device (121) is arranged to beam a medium so that the medium partially passes the obstacle (122) and is partially stopped by the obstacle (122) so that an image is generated; an arrangement for placing the vehicle (100), so that the hinge angle (α) is roughly 0° and so that a hypothetical sight line (120) between the beam-generating device (121) and the obstacle (122) points towards a first object (124), whereby the hinge angle sensor (1 10) is arranged to indicate a measured hinge angle; an arrangement (121, 122) for projecting the image on the first object (124) in a first position (125); an arrangement for moving the vehicle (100) some way towards or away from the first object (124), so that the hinge angle sensor (110) continues to display the same measured hinge angle; an arrangement (121, 122) for projecting the image on the first object (124) in a second position (126); and an arrangement for adjusting the zero position of the hinge angle sensor (110) if a second distance between the first position of the image (125) and the second position of the vehicle (126) is greater than the second distance equivalent to the minimum acceptable error in the hinge angle measurement for the movement performed.

8) A vehicle in accordance with claim 7, characterised in that the first distance is zero.

9) A vehicle in accordance with claim 7 or 8, characterised in that the medium is one of the following: electromagnetic waves (preferably radar or laser); sound (preferably sonar); particles (preferably water or dye).

10) A vehicle in accordance with any of the claims 7-9, characterised in that the obstacle (122) is a panel with a hole (123).

H) A vehicle in accordance with any of the claims 7-10, characterised in that the first object (124) is a screen sensitive to the medium.

12) A vehicle in accordance with any of the claims 7-11, characterised in that the vehicle is arranged to calibrate until the error in the zero position is less than 0.5°, and preferably less than 0.1°. 13) A vehicle (100) in accordance with any of the claims 7-11, characterised in that the vehicle also includes an arrangement for placing the vehicle (100) so that the hinge angle sensor (110) indicates the zero position; an arrangement (121, 122) for generating an image in a sight line (120), which image can be projected on a second object (140); and an arrangement for adjusting the zero position of the distance scanner (111, 112) so that the second object (140) is at the zero position of the distance scanner (111, 112).

14) A vehicle in accordance with claim 13, characterised in that the vehicle is arranged to calibrate the distance scanner (111, 112) until the error in the zero position is less than 0.5°, and preferably less than 0.3°.

15) A vehicle in accordance with claim 13 or 14, characterised in that the distance scanner (121) is from the group of laser, radar and sonar.

16) A vehicle in accordance with any of the claims 7-15, characterised in that the vehicle is a mining and/or construction vehicle.

17) A vehicle in accordance with any of the claims 7-16, characterised in that the vehicle is a self-propelling vehicle.