METHOD FOR ISSUING COLLISION WARNINGS FOR AN AGRICULTURAL VEHICLE TRACTION

Description

[0001] The invention relates to a method for issuing collision warnings for an agricultural vehicle combination, comprising an agricultural tractor and an attachment or accessory device mounted on the agricultural tractor, in particular in a rear area.

[0002] German patent DE 10 2013 011 089 A1 discloses an anti-collision system for an agricultural vehicle. The anti-collision system comprises a first sensor for detecting the dimensions of a clearance profile available in front of the vehicle in the direction of travel, a second sensor for detecting the dimensions of a load being moved by the vehicle in the form of a silage trailer, and a processing unit connected to the first and second sensors. This processing unit compares the dimensions of the clearance profile with the dimensions of the trailer based on the signals from the first and second sensors and issues a warning signal in the event of an imminent collision between the trailer and a limit of the clearance profile.

[0003] The function of the known anti-collision system is limited to a comparison of the dimensions between the transport trailer and the clearance profile available in the direction of the preceding roadway, but neglects, among other things, the influence of likely swerving movements of the transport trailer, for example when turning at a fork in the road or intersection.

[0004] US 2021 / 122368 A1 provides a method and system for monitoring a vehicle's surroundings to assist drivers in traffic situations and facilitate vehicle handling, particularly in areas with poor visibility or complex traffic situations. The system is especially useful for agricultural and construction vehicles, particularly those with trailers, in road traffic and in areas with poor visibility such as intersections, entrances and exits, and yards.

[0005] It is therefore an object of the present invention to further develop a method of the type mentioned at the outset with regard to an extended warning of the driver against possible collisions with obstacles adjacent to a driving path.

[0006] This problem is solved by a method having the features of claim 1.

[0007] The procedure for issuing collision warnings for an agricultural vehicle combination comprising an agricultural tractor and an attachment or accessory mounted on the agricultural tractor, particularly in a rear area, provides that a control unit (a) a risk zone enveloping the vehicle combination is determined, the risk zone representing a surface area on the earth's surface swept by the outer contours of the vehicle combination as a result of its expected movement along a route to be traveled, (b) objects located in the risk zone and / or those expected to enter it as a result of the direction of travel and / or the trajectory of the vehicle combination are identified from a cartographic database and / or by sensory environmental detection, (c) a driver warning is issued when an object located in the risk zone and / or that is expected to enter it is identified and thus classified as an obstacle, (d) the course of the risk zone is adjusted according to driving-related lateral and / or pivoting movements of the attachment or accessory.

[0008] Based on a predictive approach that assesses the expected movement of the vehicle combination, future events that lead to a reassessment of the collision situation can be included for the purpose of providing the driver with enhanced warnings. These include, among other things, swerving movements of the implement or attachment on the agricultural tractor along the route, for example, when turning at a fork in the road or an intersection.

[0009] The surface area affected by the outer contour of the vehicle combination, i.e., the area along the path ahead defined by its structural limitations, is significantly influenced by the model and type of the agricultural tractor and the attached implement or attachment. In the case of a towed implement or attachment, the angle of attack or articulation relative to the agricultural tractor must also be considered, as this angle results from the expected movement of the vehicle combination. The corresponding model- or type-specific contour information can be stored in a memory unit communicating with the control unit and can be selected by the operator via a user interface communicating with the control unit.It is also conceivable to obtain this information for a specific implement or attachment through optical measurement using an imaging device assigned to the agricultural tractor. This device could be a camera system or a lidar. The measurement results can simultaneously be used to update or supplement the contour information stored in the memory unit.

[0010] Predicting the expected movement of the vehicle combination, including its direction and trajectory, is possible using parameters characteristic of the route to be traveled. These parameters include the steering angle observed at the steerable wheels or steering handle of the agricultural tractor, as well as its speed. Further parameters, such as the tractor's sideslip or yaw angle, can be taken into account for refinement. It is also conceivable to incorporate cartographic information regarding the road or path along the route, which can be provided by a GPS-based navigation system.

[0011] The predictive range of the expected movement is typically limited to a few tens of meters in advance and assumes that, under normal circumstances, the steering angle and speed of the agricultural tractor change steadily and not abruptly. However, given the comparatively low speeds typical in agriculture, this predictive horizon is sufficient to detect potential collision risks with obstacles in the path in time and to prompt the driver to take appropriate countermeasures.

[0012] To determine whether an object is located within the identified risk zone or is likely to enter it, the control unit ascertains its current position. This can be done in different ways, depending on the situation. One possibility is for the control unit to map the course of the identified risk zone by linking it with GPS-acquired position information. This is then compared with the position of the object in question, assuming that this position is stored in the cartographic database. A second possibility is for the vehicle combination or agricultural tractor to serve as the relevant reference frame for the course of the risk zone and the position of the object. This can be particularly useful for objects detected during sensor-based environmental monitoring whose absolute (cartographic) position is unknown. A reference frame based on the vehicle combination or...Position determination for agricultural tractors is also important in cases where GPS-based acquisition of position information is not possible due to insufficient satellite reception. Such situations arise, for example, in covered areas such as warehouses or barns. Advantageous embodiments of the method according to the invention are described in the dependent claims.

[0013] To improve the accuracy of position determination, sensory environmental sensing can be performed using at least one imaging device assigned to the agricultural tractor and / or the implement or attachment, whereby a large number of image data points acquired from different perspectives along the vehicle combination's path are superimposed. This allows for the targeted reduction of errors in individual image data points caused by noise or other influences.

[0014] According to the invention, the course of the risk zone is adjusted by the control unit in accordance with driving-related lateral and / or pivoting movements of the attachment or accessory. Such driving-related lateral and / or pivoting movements can be caused by lateral overshooting of the attachment or accessory. Such overshooting occurs, for example, with attachments or accessories mounted on a three-point linkage in a raised position during transport journeys due to existing lateral play of the three-point linkage around a rest position of the attachment. The extent of the driving-related lateral and pivoting movements can be predicted based on the expected movement pattern of the vehicle combination along the route to be traveled, taking into account the lateral dynamic behavior of the attachment or accessory on the three-point linkage.The lateral dynamic behavior of the attachment or accessory is assumed to be known and can be determined in advance through testing.

[0015] Furthermore, it is possible that static objects detected during sensor-based environmental monitoring—that is, fixed and therefore usually permanently present objects—are mapped by the control unit and stored in the database for the purpose of completion or updating. These objects can include guardrails, road signs, fences, large boulders, roadside buildings, and the like.

[0016] The sensory environment detection can be repeated upon a renewed approach to a specific object, allowing the system to determine whether the object is still present. If it is not, it is deleted from the cartographic database. To reduce data volume, it is also conceivable to retain an entry in the cartographic database after a predetermined storage period only if the associated object was previously classified as an obstacle and therefore poses a potential risk when the same route is traveled again.

[0017] To minimize the risk of collision, the control unit may be designed to automatically initiate emergency braking if the operator fails to respond to a warning issued by the driver (typically by braking or executing an evasive maneuver). This emergency braking can be triggered after a predetermined warning period by automatically activating the relevant wheel brakes of the agricultural tractor, independent of the driver.

[0018] In this context, it is conceivable to alert the driver to an increasing risk of collision by having the control unit gradually escalate the driver warning before triggering the automatic emergency braking process. A first visual warning stage is conceivable, followed by a second visual and audible warning stage, after which, if there is no reaction from the operator, the automatic emergency braking process is finally initiated. Additionally, at least one of the warning stages could include haptic feedback through a vibration or similar sensation transmitted to the steering wheel or driver's seat. The first and second warning stages would each have fixed warning durations, which could be on the order of a few seconds.

[0019] Instead of the aforementioned time-based escalation of the driver warning, the control unit can also assign a specific collision probability to the detected objects, with the driver warning being adjusted accordingly. Different signal colors can be assigned to the first visual and second visual-audible warning levels, for example, "yellow" for the first warning level indicating a moderate risk of collision and "red" for the second warning level indicating an acute risk. If a non-critical object is identified that poses no risk of collision, a further warning level of "green" can be provided. This serves more as information for the driver than as a warning in the strictest sense.The use of such a "signal light" to visualize the probability of a collision is familiar to the driver and therefore quickly grasped intuitively. If there are no objects in the vicinity of the identified risk zone, the "signal light" can also be deactivated.

[0020] Characteristics for assigning a specific collision probability include, in addition to the estimated collision time based on the current driving speed (i.e., the time elapsed between the vehicle combination and the collision point), the shape and dimensions of the identified object. These can be determined without significant effort during sensor-based environmental monitoring, while simultaneously assessing whether a collision with the identified object can be ruled out from the outset, given the ground clearance of the agricultural tractor or any attached implement.

[0021] On the other hand, an assessment can also be made as to whether the identified object, based on its nature and location within the risk zone, could actually cause damage to the vehicle combination. For example, damage is unlikely for branches or similar objects that merely graze the edge of the risk zone. Accordingly, a driver warning can be omitted. The necessary information regarding the type of object in question is also stored in the cartographic database.

[0022] The method according to the invention can be extended to include the detection of height restrictions, such as those caused by bridges, traffic lights, gateways, or the like. The corresponding clearance heights, along with their current location, can be stored in the cartographic database or predictively determined by means of sensor-based environmental sensing along the route to be traveled. If passage is not possible due to the dimensions of the agricultural tractor or the attached implement or accessory resulting from the relevant model- or type-specific contour information, a further driver warning can be issued, and an emergency braking procedure can be initiated.

[0023] The inventive method for issuing collision warnings for an agricultural vehicle combination is described in more detail below with reference to the accompanying drawings. These show: Fig. 1 shows a vehicle combination from above with an agricultural tractor and a plow attached to it as it approaches a bend, Fig. 2 shows a schematic arrangement for carrying out the method according to the invention, and Fig. 3 shows an illustration of the method according to the invention in the form of a flowchart.

[0024] Fig. 1Figure 14 shows a vehicle combination 14 consisting of an agricultural tractor 10 and an implement 12. The implement 12 is, for example, a plow 16, which is attached to a conventional three-point linkage 18 at the rear 20 of the agricultural tractor 10. A typical driving situation is depicted during a transport operation shortly before turning onto a side road 22.

[0025] It should be noted that instead of a plow 16, the implement 12 can also be an attachment or accessory of any other design, including a towed implement such as a transport or loading trailer, which is attached to a coupling jaw of the agricultural tractor 10 via a drawbar. Furthermore, a front linkage can be provided instead of the rear three-point linkage 18 shown.

[0026] Furthermore, it shows Fig. 2A schematically illustrated arrangement 24 for carrying out the method according to the invention. The arrangement 24 comprises a microprocessor-controlled control unit 26, which is connected to a cartographic database 28, a storage unit 30, a first imaging device 32 assigned to the agricultural tractor 10, a second imaging device 34 assigned to the attachment or accessory 12, an operator interface 36, a GPS-based navigation system 38, a sensor system 40 for recording vehicle dynamic parameters, and a control unit 42 for actuating wheel brake devices 44 of the agricultural tractor 10.

[0027] The cartographic database 28 is assigned, for example, to the agricultural tractor 10; alternatively, it can also be formed by a data cloud to which the control unit 26 has access via a wireless interface. In this case, the entries contained in the data cloud can be made available to other agricultural vehicles via a central farm management system, such as a harvester following the agricultural tractor 10 in the processing process, or the like.

[0028] Further details of arrangement 24 will be explained in the context of its operation using the flowchart in Fig. 3 The flowchart illustrates an exemplary embodiment of the method according to the invention.

[0029] The method according to the invention is called up in a start-up step 100 when the agricultural tractor 10 is started up or at the instigation of a driver via the operator interface 36 which communicates with the control unit 26. For this purpose, the operator interface 36 comprises an input unit 46 in the form of a touch-sensitive display (see Fig. 2 ), via which the relevant function can be selected using a menu.

[0030] In a subsequent first step 102, the control unit 26 determines a risk zone 48 surrounding the vehicle combination 14 (see Fig. 1 The risk zone 48 represents a surface area 50 which is swept over or stressed by the outer contours of the vehicle combination 14 due to the expected movement pattern along a travel path 52.

[0031] As in Fig. 1As can be seen, the course of risk zone 48 is determined by the structural limitations 54 of the vehicle combination 14. The latter determine the surface area 50 affected along the preceding travel path 52 and are significantly influenced by the model or type of the agricultural tractor 10 and the attached implement 12. In the case of a towed implement 12, the angle of attack or articulation relative to the agricultural tractor 10 must also be taken into account, which in turn results from the expected movement of the vehicle combination 14.

[0032] The corresponding model- or type-specific contour information is stored in the memory unit 30 and can be selected by the operator via the operator interface 36, which communicates with the control unit 26. Optionally, this information is obtained by optical measurement of the implement 12 using the first imaging device 32 assigned to the agricultural tractor 10. The first imaging device 32 is a camera system or lidar mounted in the roof area of ​​the driver's cab 56 of the agricultural tractor 10 for 360° surround detection. The measurement results are used to update or supplement the contour information contained in the memory unit 30.

[0033] In a second step 104, the control unit 26 identifies objects located in the risk zone 48 and / or objects likely to enter it due to a direction of travel and / or a direction of travel of the vehicle combination 14 from the cartographic database 28 and / or by sensory environment detection using the first and second imaging devices 32, 34.

[0034] In this case, the second imaging device 34 is also designed as a camera system or lidar, which extends the sensory environmental detection carried out by the first imaging device 32 in the area surrounding the agricultural tractor 10 with regard to the attachment or accessory device 12.

[0035] To improve the accuracy of the position determination, the control unit 26 overlays a large number of image data acquired from various perspectives along the route 52 of the vehicle combination 14. This allows for the targeted reduction of errors in individual image data caused by noise or other influences. If lower data quality requirements are necessary, optionally only the first imaging unit 32 can be used.

[0036] In addition, static objects detected during sensory environmental perception—that is, fixed and therefore generally permanently present—are mapped by control unit 26 and stored in database 28 for the purpose of its completion or updating. As shown in Fig. 1 These objects include, for example, a perimeter development in the form of two buildings 58, 60 and a motor vehicle 62 parked between them.

[0037] In the second step 104, the prediction of the expected movement of the vehicle combination 14, as well as the direction and course of travel, is carried out using parameters characteristic of the route 52 to be traveled. These parameters include, in addition to the steering angle observed at the steerable wheels or steering handle of the agricultural tractor 10, its speed, whereby further parameters in the form of a sideslip or yaw angle occurring on the agricultural tractor 10 can be taken into account for refinement. The relevant vehicle dynamic parameters are recorded by the sensor system 40 arranged in the agricultural tractor 10 and transmitted to the control unit 26. The control unit 26 also incorporates cartographic information regarding the road or path course in the direction of travel, which is provided by the GPS-based navigation system 38.

[0038] To determine whether an object is located in the identified risk zone 48 or is likely to enter it, control unit 26 ascertains its current location. This is done in different ways depending on the situation.

[0039] A first procedure involves the control unit 26 mapping the course of the identified risk zone 48 using the navigation system 38 by linking it with GPS-based position information. This is then compared with the location of the object in question, assuming that this location is stored in the cartographic database 28. For example, such a determination is carried out for the in Fig. 1The first and second buildings 58 and 60 are located at the roadside, with the first building 58 being situated within the identified risk zone 48 and therefore likely obstructing the turning maneuver of the vehicle combination 14 with regard to the attached attachment or accessory 12. The second building 60, however, lies outside the identified risk zone 48 and is therefore considered non-critical.

[0040] A second approach proposes that the vehicle combination 14 or the agricultural tractor 10 forms the relevant reference system for the course of risk zone 48 and the location of the object. This can be particularly useful for objects identified during sensor-based environmental monitoring whose absolute (cartographic) location is unknown. In this case, this applies to the motor vehicle 62 parked between the two buildings 58 and 60; however, it could just as easily be a static obstacle not recorded in the cartographic database 28.

[0041] The course of risk zone 48 is adjusted by the control unit 26 in a subsequent third step 106 according to driving-related lateral and / or pivoting movements of the attachment or auxiliary device 12. Such driving-related lateral and / or pivoting movements can occur when turning due to a lateral overshoot 64 of the attachment or auxiliary device 12 in a raised state as a result of existing lateral play of the three-point linkage 18 around an equipment rest position 66 (see Fig. 1The extent of the driving-related lateral and pivoting movements is predicted by the control unit 26 based on the expected movement pattern of the vehicle combination 14 along the travel path 52, taking into account the lateral dynamic behavior of the attachment or accessory 12 on the three-point linkage 18. The lateral dynamic behavior of the attachment or accessory 12 is assumed to be known and can be determined in advance through testing. The data thus determined are model- or type-specific for the respective attachment or accessory 12 and are stored, along with the associated contour information, in the memory unit 30.

[0042] In a fourth step 108, the control unit 26 issues a driver warning by activating the operator interface 36 if, based on the information obtained in the third step 106, an object located in and / or likely to enter risk zone 48 is identified and thus classified as an obstacle. In this case, this applies to the first building 58 and the motor vehicle 62 parked next to it. Both objects are considered obstacles due to their location within risk zone 48.

[0043] Alternatively, 108 different approaches are pursued for issuing the driver warning in the fourth step: Time-based driver warning

[0044] To minimize the risk of collision, the control unit 26 is designed to automatically initiate an emergency braking process if the operator fails to respond as expected following the issued driver warning. This emergency braking process is triggered after a predetermined warning period by the control unit 42, which communicates with the control unit 26, by intervening in the wheel brakes 44 of the agricultural tractor 10 independently of the driver.

[0045] To alert the driver to an increasing risk of collision, the driver warning is escalated in stages by control unit 26 before the automatic emergency braking process is triggered. This includes an initial visual warning, followed by a second visual and audible warning. If the driver fails to react, the automatic emergency braking process is then initiated. The first and second warning stages each have fixed warning durations, typically a few seconds. Risk-based driver warning

[0046] Instead of the aforementioned time-based escalation of the driver warning, the control unit 26 assigns a specific collision probability to the detected objects, and the driver warning is adjusted accordingly. Different signal colors are assigned to the first visual and second visual-acoustic warning levels. For example, the first warning level is yellow, indicating a moderate risk of collision, and the second warning level is red, indicating an acute risk of collision. If a non-critical object is identified that poses no risk of collision, a further warning level of green is provided. This serves more as information for the driver than as a warning in the strictest sense. The use of such a "traffic light" system to visualize the collision probability is familiar to the driver and therefore quickly grasped intuitively.If there are no objects in the vicinity of the identified risk zone, the "signal light" is deactivated by control unit 26.

[0047] The characteristics used to assign a specific collision probability include, in addition to a collision time estimated from the current driving speed (i.e., the time elapsed between the vehicle combination 14 and the time of collision), the shape and dimensions of the identified object. These are determined during sensor-based environmental perception, whereby the control unit 26 simultaneously assesses whether a collision with the identified object can be ruled out from the outset, given the ground clearance of the agricultural tractor 10 or the attached implement 12. The control unit 26 also considers the type of identified object. If damage to the vehicle combination 14 is deemed unlikely based on its shape, dimensions, type, and position, no driver warning is issued.The necessary information on the type of object in question is also stored in the cartographic database 28.

[0048] A typical sequence for issuing the driver warning is that the warning level is displayed in the color "green" at a time interval of 5 seconds or more, and the warning level in the color "yellow" at a time interval of between 1 and 5 seconds.

[0049] The process is then terminated in a final step 110.

Claims

1. Method for issuing collision warnings for an agricultural vehicle combination comprising an agricultural tractor (10) and an accessory or ancillary implement (12) attached to the agricultural tractor (10), in particular in a rear region (20), characterized in that a control unit (26) (a) determines a risk zone (48) surrounding the vehicle combination (14), wherein the risk zone (48) represents, on the ground surface, a surface region (50) over which the outer contours of the vehicle combination pass due to a movement course of the vehicle combination (14) to be expected along a route (52) to be covered, (b) determines objects located in the risk zone (48) and / or anticipated to enter the risk zone due to a direction of travel and / or a direction of travel course of the vehicle combination (14) from a cartographic database (28) and / or by sensor-based surroundings detection, (c) issues a driver warning when an object located in the risk zone (48) and / or anticipated to enter the risk zone is identified and thus classified as an obstacle, (d) adjusts the course of the risk zone (48) in accordance with the lateral and / or swivel movements of the accessory or ancillary implement (12) as a result of travel.

2. Method according to Claim 1, characterized in that the sensor-based surroundings detection is carried out by means of at least one imaging device (32, 34) associated with the agricultural tractor (10) and / or the accessory or ancillary implement (12), wherein a large amount of image data recorded along the route (52) of the vehicle combination (14) from different perspectives is superimposed.

3. Method according to Claim 1 or 2, characterized in that the travel-related lateral and / or swivel movements of the accessory or ancillary implement (12) are detected by means of at least one imaging device (32) associated with the agricultural tractor (10).

4. Method according to at least one of the preceding claims, characterized in that static objects determined during the course of the sensor-based surroundings detection are mapped by the control unit (26) and stored in the database (28).

5. Method according to at least one of the preceding claims, characterized in that an automatic emergency braking operation is initiated by the control unit (26) in the absence of an operator response to be expected due to the issued driver warning.

6. Method according to Claim 5, characterized in that the driver warning is gradually escalated by the control unit (26) before the automatic emergency braking operation is initiated.

7. Method according to at least one of the preceding claims, characterized in that a specific collision probability is assigned to the determined objects by the control unit (26), wherein the driver warning is adjusted in accordance with the assigned specific collision probability.

8. Method according to Claim 7, characterized in that the shape and dimensions of the identified object are used by the control unit (26) as characteristics for the assignment of a specific collision probability.