Method and system for identifying dangerous objects around a vehicle
By detecting objects around the vehicle and determining their orientation parameters, judging hazardous conditions based on motion parameters, identifying and activating corresponding safety functions, the reliability and efficiency of identifying hazardous objects in the prior art are solved, and the safety and handling of vehicles are improved efficiency.
Patent Information
- Application Number
- CN202010597152.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-06-28
- Filing Date
- 2020-06-28
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2040-06-28
AI Technical Summary
The prior art has reliability and efficiency problems in identifying dangerous objects around the vehicle, especially when processing large amounts of sensor data, it is difficult to effectively avoid the risk of incorrect activation of safety functions, resulting in potential accidents.
The computer hardware components detect objects around the vehicle, determine the orientation parameters of the object, determine whether the hazardous conditions are met based on the orientation parameters and other motion parameters, thereby identifying the hazardous object and activating the corresponding safety function according to the type of the hazardous object.
It improves the identification accuracy of dangerous objects and the activation efficiency of safety functions, reduces the risk of incorrect activation, and enhances the safety and handling complexity of the vehicle.
Smart Images

Figure CN112141093B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to a method and system for identifying dangerous objects around a vehicle. Various functions of the vehicle can be activated based on one or more identified dangerous objects, such as a warning function for the driver (e.g., forward collision warning, FCW) or an autonomous driving function (e.g., automatic emergency braking, AEB). Background Art
[0002] Sensor technologies are used in automotive applications to implement various comfort or safety functions, such as collision warnings and active collision avoidance measures (e.g. AEB). Different sensor technologies can be used alone or in combination, wherein possible sensor types include (as a non-exhaustive list of examples) radar sensors, light detection and ranging (laser) sensors, imaging devices (e.g. cameras) and / or ultrasonic sensors. Although the different sensor types are able to capture the vehicle environment under various conditions very well, it remains a challenge to identify potential dangerous objects from the sensor data in an effective and reliable manner. In principle, with an increase in the amount of processed sensor data, a higher reliability can be achieved in identifying dangerous objects, which are considered to be relevant for the desired safety function. On the other hand, the processing of the data needs to be carried out in a very short time, in particular in real time, so that safety or other control functions (such as AEB) can still effectively avoid collisions. Therefore, it is generally not feasible to process a large amount of sensor data about different dangerous situations using the data processing resources of a standard vehicle. A further challenge is to minimize the risk of erroneous activation of safety functions. False activation means that activation is actually unnecessary and can therefore lead to surprises or even accidents. For example, erroneous activation of the AEB function in a vehicle traveling at high speed on a crowded highway should be avoided in view of potential accidents. Correct identification of dangerous objects is therefore an important aspect of all functions based on this information. This also includes partially or fully autonomous driving functions.
[0003] Therefore, there is a need to provide an improved method for identifying hazardous objects around a vehicle. Summary of the invention
[0004] The present disclosure provides a computer-implemented method, a computer system, and a non-transitory computer-readable medium. Implementations are presented in the specification and drawings.
[0005] In one aspect, the present disclosure relates to a computer-implemented method for identifying dangerous objects around a vehicle, the method comprising the following steps performed (in other words: carried out) by a computing hardware component: detecting an object in the surrounding environment of the vehicle; determining an orientation parameter of the object, the orientation parameter representing the difference between the orientation of the object and the orientation of the vehicle; determining whether the object satisfies at least one dangerous condition based on the orientation parameter; and identifying the object as a dangerous object if at least one dangerous condition is satisfied.
[0006] The term "orientation" refers to a direction or heading. For example, when considering the orientation of a vehicle, the orientation can be defined relative to a predetermined axis of the vehicle or another predetermined geometric element (which can be a coordinate system with a fixed spatial relationship relative to the vehicle body). In a specific example, the orientation of the vehicle can be defined by the longitudinal axis of the vehicle, which can be an axis parallel to the longer part of the vehicle body. The longitudinal axis can also be parallel to the direction of travel along a straight line when the steering element is in a neutral steering position. The longitudinal axis can have a direction that is substantially pointing away from the outside of the windshield (i.e., the front panel) of the vehicle. Alternatively, the orientation can be defined as the current direction of movement of the vehicle. The same type of definition can also be applied to the orientation of an object.
[0007] The orientation parameter may be, for example, an angle and may indicate an angular displacement between the orientation of the vehicle and the orientation of the object. In one example, the orientation parameter represents the orientation of the object relative to a vehicle coordinate system, wherein the vehicle coordinate system has a fixed spatial relationship relative to the vehicle. In this way, the orientation parameter directly represents the difference in orientation between the vehicle and the object.
[0008] Typically, the orientation parameter may be a numerical value representing the orientation difference between the vehicle and the object. As further noted above, the orientation parameter may be represented as an angle, but other representations may also be used. For example, the orientation parameter may be represented as a categorical variable having a finite number of categories, each of which is associated with one of a plurality of orientation differences. The categories may be represented as numbers, letters, or other symbols.
[0009] The orientation parameter is preferably a numerical value, wherein the magnitude indicates the degree of the difference in orientation. In particular, when the orientation parameter is zero, the orientations of the vehicle and the object are considered to be substantially the same, i.e., the vehicle and the object are aligned parallel. When the vehicle and the object are oriented in opposite directions, the orientation parameter has a maximum value, which corresponds to an angle of 180 degrees. However, other ways of expressing the difference in orientation are also possible and are not limited to angles.
[0010] It has been found that orientation parameters are well used for identifying dangerous objects in an efficient and reliable manner. On the one hand, the orientation difference between the vehicle and the object represents a large amount of useful information in a concise way (i.e. in one variable). That is, the orientation parameters not only encode the spatial relationship between the vehicle and the object, but also carry information about their current motion paths. For example, if the orientation difference is 90 degrees, the motion paths of the object and the vehicle cross at essentially the same angle. Of course, this does not necessarily mean that the vehicle and the object have collided, but the expected directions of motion are assumed to be very different.
[0011] The identification of hazardous objects based on orientation parameters allows an intuitive and concise definition of one or more hazardous conditions. If more than one hazardous condition is used, different hazardous scenarios can be considered. This allows different hazard qualities or categories to be distinguished so that a safety function can be selected depending on the hazard quality or category. For example, if the hazard quality is low, a warning can only be presented to the driver visually or acoustically, while when the hazard quality is high, an emergency braking function can be directly activated. In other words, different types of hazardous objects can be considered, wherein different types of actions can be performed depending on the type of hazardous object.
[0012] It should be understood that the at least one hazard condition need not be defined solely based on the orientation parameter, although this can be done if desired. In one example, the at least one hazard condition may be defined solely based on the orientation parameter, wherein only when the at least one hazard condition is met, further hazard conditions are considered. In this way, a stepwise approach may be employed to increase the overall efficiency of the process. However, as another example, the at least one hazard condition may also be defined by other parameters in addition to the orientation parameter.
[0013] It can also be understood from the above that the identification of a dangerous object does not necessarily imply that the safety functions of the vehicle will be automatically activated, although this can be done if necessary. In one example, the method may include activating at least one safety function for controlling the vehicle based on one or more identified dangerous objects. Therefore, typically, the identification of a dangerous object regards the object as a potential danger to the vehicle, which distinguishes it from other objects around the vehicle. In this way, the method is configured to classify the detected objects around the vehicle into at least two groups, namely dangerous objects and other objects. This includes that if at least one danger condition is not met, the detected object is not identified as a dangerous object.
[0014] The at least one dangerous condition may be defined by one or more logical or mathematical expressions. For example, the dangerous condition may include one or more thresholds for distinguishing dangerous objects from other objects based on orientation parameters. More specific examples of dangerous conditions will become apparent from the following.
[0015] Objects around the vehicle and therefore potentially dangerous objects can also be the vehicle itself. In this case, the object is a vehicle tool other than the self or host vehicle referred to in the method (usually, the term "vehicle" does not refer to this object). However, the object can also be another member of the traffic, for example, a pedestrian or a cyclist. The vehicle can be, for example, a car, a motorcycle, a truck, etc.
[0016] The detection of one or more objects can be performed using a detection algorithm and based on data representing the vehicle environment. The detection algorithm can usually be configured with respect to the type of data, which can be sensor data captured by one or more sensors installed on the vehicle. The detection algorithm can depend on the underlying sensor technology of the sensor. Preferably, a radar sensor system can be installed in the vehicle, and the detection of the object is performed based on the radar sensor data provided by the radar sensor system. Then, angle information useful for determining the orientation parameter is provided as part of the radar detection point (the radar detection point is part of the typical radar sensor data). In addition, at least for the radar sensor data, reliable object detection algorithms are known in the art. Therefore, when using a radar sensor system, the orientation parameter can be determined efficiently and reliably. However, other sensor technologies can also be used as a substitute or supplement to the radar system. One or more sensors do not necessarily have to be installed in the vehicle, because in principle the surrounding environment of the vehicle can also be captured from outside the vehicle.
[0017] It will be appreciated that at least some or all of the steps of the method may generally be associated with a common time instance. In particular, one or more objects and their orientation relative to the vehicle may be determined in a first time instance, wherein one or more dangerous conditions are checked in the first time instance. The process may be repeated in the same manner in a second time instance and a third time instance, and so on. In other words, the method may be performed periodically, thereby monitoring the surroundings of the vehicle over time, wherein one or more functions of the vehicle may be dynamically activated based on the detected dangerous objects. Continuous scans for detecting objects around the vehicle may be processed together in order to further enhance the reliability of the method, which also improves the safety of the vehicle.
[0018] According to one embodiment, at least one hazardous condition specifies an orientation range for an orientation parameter. For example, if the orientation parameter is within the orientation range of at least one hazardous condition, the corresponding hazardous condition is considered to be satisfied (i.e., met). The range can be limited by two thresholds, which represent a lower limit and an upper limit, respectively, and the range extends between or beyond these limits. A single threshold may also be sufficient to define a range, for example, from a predefined value x to infinity. Although the orientation range can cover all values of the range, it can alternatively be limited to a subset of values or other symbols. It has been found that the orientation range is useful for defining one or more hazardous conditions in an effective manner. The orientation range can be interpreted as a set of different orientations that have essentially the same relevance from a hazardous perspective.
[0019] According to another embodiment, the method further comprises determining at least one vehicle motion parameter of the vehicle, wherein at least one hazardous condition specifies a vehicle motion range of the at least one vehicle motion parameter. The vehicle motion parameter represents the motion of the vehicle. In particular, the at least one vehicle motion parameter may comprise or represent the speed of the vehicle and / or a steering parameter of the vehicle, wherein the steering parameter indicates the direction of the steering, e.g. left or right. As is generally understood according to the term "parameter", the parameter may be represented, for example, as a numerical value or as another symbol, e.g., a letter.
[0020] The speed of the vehicle may be provided by a device for determining speed, which is preferably installed in the vehicle. For example, the speed may be provided by a speedometer of the vehicle. Alternatively, the speed may be provided from a global positioning system (GPS) or other source via a network.
[0021] The vehicle motion range may be defined according to one or more thresholds, as discussed generally in connection with the orientation range. By defining the hazard condition according to the speed of the vehicle, the method may be made more reliable. It should be understood that the vehicle motion parameter is not an orientation parameter.
[0022] According to another embodiment, the method comprises determining an object motion parameter of the object in addition to the orientation parameter, wherein at least one danger condition specifies an object motion range of the object motion parameter. The object motion parameter represents the motion of the object. In particular, the object motion parameter comprises or represents the speed of the object. The speed may be expressed relative to the speed of the vehicle or relative to the ground.
[0023] The speed of the object can be determined from data captured by one or more sensors mounted on the vehicle (e.g., the radar system). The corresponding sensor data can be the same as the data used to determine the orientation parameter. By considering the object speed in addition to the orientation, the reliability can be further improved. For example, if the orientation difference between the vehicle and the corresponding object is large, the additional consideration of the object speed can help reduce the false identification of dangerous objects.
[0024] In another embodiment, at least one danger condition specifies a predetermined relationship between a vehicle motion parameter and an object motion parameter. An example of a predetermined relationship is that the vehicle motion parameter is lower than the object motion parameter. In both cases, the parameter can be given by a speed value. Incorporating said relationship into the danger condition allows to further improve the reliability of the method.
[0025] According to one embodiment, the method uses a plurality of different risk conditions. This means that in the step of determining whether an object meets at least one risk condition, a plurality of risk conditions are taken into account, preferably three to seven risk conditions are taken into account, and particularly preferably five risk conditions are taken into account. This allows to classify objects using different compliances between objects with different risk relevances.
[0026] The plurality of hazard conditions may comprise at least one pair of hazard conditions that are axially symmetric with respect to a longitudinal axis of the vehicle. In particular, the individual members of the pair may have orientation ranges, wherein the two orientation ranges are axially symmetric. The longitudinal axis may be defined in the same manner as the orientations. The longitudinal axis may be defined as an imaginary straight line that is parallel to the longer stretch of the vehicle body. The symmetric definition of the hazard conditions makes the entire set of hazard conditions more concise, thereby improving the reliability and efficiency of the method.
[0027] According to one embodiment, the plurality of hazard conditions include: (i) a first hazard condition specifying a first orientation range of an orientation parameter; and (ii) at least one second hazard condition specifying a second orientation range of an orientation parameter; wherein the first orientation range is at least partially lower than and / or smaller than the second orientation range. Preferably, the first orientation range covers lower orientation parameter values than the second range. The first orientation range may be centered around a zero value of the orientation parameter, which preferably indicates that the orientation difference is zero. Also preferably, the first orientation range and the second orientation range do not overlap.
[0028] The first hazard condition may further specify a first vehicle motion range of a first vehicle motion parameter of the vehicle, wherein the second hazard condition further specifies a second vehicle motion range of the first vehicle motion parameter; wherein the first vehicle motion range is at least partially higher and / or larger than the second vehicle motion range. For example, the first vehicle motion range covers higher parameter values than the second vehicle motion range, for example up to 160 kilometers per hour (km / h). In contrast, the second vehicle motion range may only cover values of, for example, up to 60 km / h. However, the first and second vehicle motion ranges may overlap, wherein the first vehicle motion range may be larger than the second vehicle motion range. The first vehicle motion parameter may be defined as further explained more generally above in conjunction with the vehicle motion parameter. It may be, for example, a speed or a steering parameter of the vehicle.
[0029] At least one second hazard condition may further specify a first object motion range of a first object motion parameter of the object, the first object motion range being defined relative to the first vehicle motion parameter. In particular, the first object motion range may depend on the first vehicle motion parameter. For example, the second hazard condition may specify that the object motion parameter needs to be higher than the first vehicle motion parameter in order to satisfy the second hazard condition, wherein one or more additional offsets may be incorporated for adjustment. The object motion parameter may be defined as further explained more generally above in conjunction with the vehicle motion parameter. Preferably, both the first vehicle motion parameter and the object motion parameter represent a speed.
[0030] According to one embodiment, the plurality of hazard conditions include at least one third hazard condition, the at least one third hazard condition specifying (i) a third orientation range of the orientation parameter and (ii) a third vehicle motion range of the first vehicle motion parameter; wherein the third orientation range is at least partially higher than the second orientation range, and wherein the third vehicle motion range is at least partially lower than and / or smaller than the second vehicle motion range. For example, an upper limit of the third vehicle motion range may be lower than an upper limit of the second vehicle motion range.
[0031] The third orientation range may overlap with the second orientation range, but preferably does not overlap with the first orientation range. Conversely, the third vehicle motion range may overlap with the first vehicle motion range and the second vehicle motion range, i.e., the third range may be a subset of the first range and / or the second range. As described above, the upper limit of the third vehicle motion range is preferably lower than the upper limit of the second vehicle motion range, i.e., the vehicle motion needs to be relatively small in order to allow the third hazard condition to be met relative to the vehicle motion. This further reduces the risk of "misidentifying" an object as a hazardous object.
[0032] According to another embodiment, the third hazardous condition may further specify a second vehicle motion range for a second vehicle motion parameter of the vehicle. This allows the method to be further improved in terms of reliability. The second vehicle motion parameter is preferably a steering parameter as further described above. In particular, the second vehicle motion parameter may include two categories, one category indicating a steering action of the driver in one direction (e.g., to the left), and the other category indicating a steering action of the driver in another direction (e.g., to the right).
[0033] According to one embodiment, if the object is identified as a dangerous object in response to at least one dangerous condition being met, the method comprises the steps of: determining whether at least one additional dangerous condition associated with the at least one dangerous condition is met; and activating at least one safety function for controlling the vehicle if the at least one additional dangerous condition is met. This is a multi-step method for activating one or more safety functions (e.g., AEB), wherein the reliability (reduced false activation) and efficiency (low processing complexity) of the method are further enhanced.
[0034] A plurality of hazard conditions may cover all possible values of the orientation parameter by means of one or more orientation ranges. In one example, there may be a first hazard condition, two second hazard conditions and two third hazard conditions, wherein the orientation ranges cover the entire circle, i.e. 360 degrees. In addition, the orientation ranges of the second hazard condition and the third hazard condition may be defined in pairs, respectively, wherein the respective pairs cover sectors of the orientation parameter having axial symmetry about an axis parallel to the longitudinal axis of the vehicle. It should be understood that the axial symmetry may also be formally expressed with respect to another axis, which need not necessarily be the longitudinal axis.
[0035] In another aspect, the present disclosure relates to a computer system configured to perform several or all of the steps of the computer-implemented method described herein.
[0036] In yet another aspect, a computer system capable of being connected or coupled to a sensor configured to capture the surroundings of a vehicle is provided, and the computer system includes a plurality of computer hardware components configured to perform a computer-implemented method of at least one of the embodiments described herein based on data captured by the sensor. The computer system and / or the sensor may be mounted on a vehicle. It should be understood that more than one sensor or sensor system may be used to capture the environment, thereby increasing the robustness of the method.
[0037] The computer system may include a plurality of computer hardware components (e.g., a processing unit, at least one storage unit, and at least one non-transitory data storage). It will be appreciated that additional computer hardware components may be provided and used to perform the steps of the computer-implemented method in the computer system. The non-transitory data storage and / or storage unit may include a computer program for instructing a computer, for example, using a processing unit and at least one storage unit to perform several or all steps or aspects of the computer-implemented method described herein.
[0038] In another aspect, the present disclosure relates to a non-transitory computer-readable medium including instructions for performing several or all of the steps or aspects of the computer-implemented methods described herein. The computer-readable medium may be configured as: an optical medium such as a compact disk (CD) or a digital versatile disk (DVD); a magnetic medium such as a hard disk drive (HDD); a solid-state drive (SSD); a read-only memory (ROM) such as a flash memory; etc. In addition, the computer-readable medium may be configured as a data storage accessible via a data connection such as an Internet connection. The computer-readable medium may be, for example, an online data repository or cloud storage.
[0039] The present disclosure also relates to a computer program for instructing a computer to execute several or all steps or aspects of the computer-implemented method described herein. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Example implementations and functionality of the present disclosure are described herein in conjunction with the following drawings, which are schematically shown:
[0041] Figure 1 A method for identifying hazardous objects around a vehicle is shown;
[0042] Figure 2 showing a first orientation range of orientation parameters of objects surrounding the vehicle;
[0043] Figure 3 A second orientation range of the orientation parameter is shown;
[0044] Figure 4 A third orientation range of the orientation parameter is shown;
[0045] Figure 5 a fourth orientation range of the orientation parameter is shown; and
[0046] Figure 6 A fifth orientation range of the orientation parameter is shown.
[0047] Reference Number List
[0048] 10 Steps of Object Detection
[0049] 12 Steps to determine orientation parameters
[0050] 14 Steps to determine if hazardous conditions are met
[0051] 16 Steps to identify dangerous objects
[0052] 18 Vehicles
[0053] 20 Objects
[0054] 20' Object
[0055] 22 Arrow
[0056] 24 Angle
[0057] A. The first dangerous condition
[0058] B. Second Dangerous Condition
[0059] C. Third Danger Condition
[0060] D. Fourth Dangerous Condition
[0061] E Fifth Dangerous Condition
[0062] OR Orientation Range
[0063] OMR Object Motion Range
[0064] L Arrow Line
[0065] VR Vehicle Motion Range DETAILED DESCRIPTION
[0066] Figure 1 A computer-implemented method for identifying hazardous objects around a vehicle is depicted, the method comprising: detecting an object in the surroundings of the vehicle (see step 10, "object detection"); determining an orientation parameter of the object (see step 12), the orientation parameter representing a difference between the orientation of the object and the orientation of the vehicle; determining whether the object satisfies at least one hazardous condition based on the orientation parameter (see step 14); and if at least one hazardous condition is satisfied, identifying the object as a hazardous object (see step 16). Each of the above steps 10, 12, 14, 16, and other steps can be performed by computer hardware components.
[0067] The orientation parameter is preferably formed by an angle, wherein the angle represents the difference in orientation between the vehicle and the object. By definition, the corresponding angle is within a potential range between -180 degrees and +180 degrees, i.e. [-180, 180], which covers a full range of 360 degrees. The angle extends within a plane that is considered to be parallel to the ground on which the vehicle and the potential object are positioned. It will be appreciated that the orientation parameter may also be defined in other ways, for example as an angle within the range of [0, 360], or expressed in radians or other units.
[0068] Combine vehicle 18 and object 20 in Figure 2 and Figure 3 20. The arrow dashed line L indicates the orientation of the vehicle 18 and the arrow 22 indicates the orientation of the object 20. The line L is the longitudinal axis of the vehicle 18 and additionally defines the axial symmetry of the body of the vehicle 18. Under normal driving conditions, the driver of the vehicle 18 usually looks through the windshield (not shown) of the vehicle 18 in the direction of the arrow line L. Similarly, the object 20 can be a vehicle or a similar type of object, wherein the arrow 22 can be defined in the same way relative to the object 20. This is in Figure 2 exemplified in.
[0069] from Figure 2 It can also be seen that the orientation of the vehicle 18 and the orientation of the object 20 are the same, since the direction of the arrow line L and the arrow 22 are parallel and point in the same direction. Therefore, the difference in orientation, and therefore the orientation parameter, is zero. It should be understood that in Figure 2 , the position of the object 20 relative to the vehicle 18 is only an example. In particular, the vehicle 18 and the object 20 do not have to be positioned on the line L, but can be arbitrarily displaced, while the orientation difference remains the same. In addition, the object 20 does not have to be in front of the vehicle 18, but can also be behind the vehicle 18, as shown by the object 20' in dashed lines.
[0070] exist Figure 3 An example of a non-zero orientation parameter is shown in . As can be seen, the angle 24 between the line L and the arrow 22 is non-zero. This also applies to the object 20 ′, which is shown to better illustrate the nature of the orientation parameter.
[0071] In step 14, a plurality of different dangerous conditions are considered. In one example, five different dangerous conditions A, B, C, D, and E are considered, respectively. Figures 2 to 6 This hazardous condition is partially illustrated in FIG.
[0072] like Figure 2As shown, the danger condition A specifies a first orientation range OR(A) of the orientation parameter, wherein the first orientation range OR(A) is between a lower threshold and an upper threshold. The first range OR(A) is between -10 degrees and +10 degrees and is therefore centered around zero, i.e., OR(A) = [-10, +10] degrees, see Figure 2 As in Figure 2 As can also be seen in FIG. 1 , the first orientation range OR(A) "extends" not only in front of the vehicle 18, but also in the rear, as indicated by the dashed line. It is emphasized that the orientation parameters do not carry information about the placement of the object relative to the vehicle. Such positioning information can be taken into account in further hazard conditions, which are only checked if one or more of the hazard conditions A to E are met.
[0073] Preferably, the hazard conditions A to E are defined such that a given detected object cannot satisfy more than one hazard condition of the plurality of hazard conditions A to E at one instance in time.
[0074] Regardless of other positioning differences between the vehicle 18 and the object 20 , if the angular displacement (ie, the angle 24 of the arrow 22 of the object 20 relative to the line L) is within OR(A), then the first hazard condition A is satisfied relative to the first orientation range OR(A).
[0075] The danger condition A further specifies a first vehicle movement range VR(A) of the speed of the vehicle 18. VR(A) covers a speed range between zero and a first upper limit value, in particular [0, 160] km / h.
[0076] Given the hazardous condition A, if (i) the orientation parameter is within OR(A) and (ii) the speed of the vehicle 18 is within VR(A), then the object 20, 20' is identified as a hazardous object. Figure 2 In FIG. 2 , both objects 20 and 20 ′ satisfy the danger condition A with respect to their orientation parameters.
[0077] As for the second orientation range OR(B), Figure 3 The second dangerous condition B is illustrated in FIG. OR(B) is between the lower threshold and the upper threshold, such as Figure 3 The OR(B) is shown to be [10, 105] degrees. The range OR(B) in the case where the object 20' is on the opposite side of the vehicle 18 is also shown in dashed lines.
[0078] The second danger condition B further specifies a second vehicle movement range VR(B) of the speed of the vehicle 18. VR(B) covers a speed range between zero and a second upper limit value, in particular, [0, 60] km / h. Therefore, VR(A) is partially higher than and greater than VR(B).
[0079] The second danger condition B further specifies a first object motion range OMR(B) of the speed of the object 20. In the example, it is required that the speed of the object 20 is greater than the speed of the vehicle 18 minus the offset k. That is, the second danger condition specifies a predetermined relationship between the speed of the vehicle 18 and the object 20, which can be expressed as V(vehicle) - k < V(object) or V(vehicle) < V(object) + k, where V() represents speed. The offset k can have a value in the range of 10 km / h to 30 km / h, especially 20 km / h. It should be understood that all speeds are preferably expressed in the same unit.
[0080] Regarding the third orientation range OR(C), the third danger condition C is illustrated in Figure 4 . OR(C) is between a lower threshold and an upper threshold. As Figure 4 shows, this OR(C) is [-105, -10] degrees. The range OR(C) in the case where the object 20' is on the opposite side of the vehicle 18 is also shown by a dashed line.
[0081] The third danger condition C further specifies a third vehicle motion range VR(C) of the speed of the vehicle 18, and this third vehicle motion range VR(C) is the same as the danger condition B, that is, VR(C) = VR(B), and VR(C) is [0, 60] km / h.
[0082] The third danger condition C also specifies a second object motion range OMR(C) of the speed of the object 20, and this second object motion range OMR(C) is the same as that in condition B, that is, OMR(C) = OMR(B).
[0083] It should be understood that the second danger condition B and the third danger condition C form a pair of danger conditions, and they are axially symmetric about the line L with respect to the orientation range OR(B) and the range OR(C).
[0084] In a similar manner, the fourth danger condition D and the fifth danger condition E form a pair of axially symmetric danger conditions, which are illustrated in Figure 5 and Figure 6 respectively. The fourth orientation range OR(D) is shown in Figure 5 , and OR(D) is [90, -175] degrees. Similarly, the fifth orientation range OR(E) is shown in Figure 6 , and OR(E) is [-90, 175] degrees.
[0085] In addition to the orientation range, the fourth and fifth hazard conditions both specify the vehicle motion range, i.e., VR(D) and VR(E), VR(D) and VR(E) are the same and equal to [0, 30] km / h. Therefore, VR(A) is partially higher than and greater than VR(B), VR(C), and partially higher than and greater than VR(D), VR(E). In addition, VR(B) and VR(C) are partially higher than and greater than VR(D), VR(E).
[0086] It should be understood that the hazard conditions A, B, C, D, E are only examples and that other definitions are possible to achieve at least similar results. However, it has been found that the presented structure and values of hazards provide significant advantages in terms of reliability and efficiency in identifying hazardous objects.
Claims
1. A computer-implemented method for identifying dangerous objects around a vehicle, The computer-implemented method comprises the following steps performed by computer hardware components: - detecting objects (20, 20') in the surroundings of the vehicle (18); - determining an orientation parameter (24) of the object (20), the orientation parameter (24) representing the difference between the orientation of the object (20) and the orientation of the vehicle (18), the orientation parameter being an angle; - determining whether the object (20) satisfies at least one danger condition (A, B, C, D, E) based on the orientation parameter (24), wherein For each of the different areas of the environment around the vehicle, the at least one hazard condition (A, B, C, D, E) specifies an orientation range (OR) of the orientation parameter (24); and - if the at least one dangerous condition is met, identifying the object (20) as a dangerous object, Therein, in the step of determining whether the object (20) satisfies at least one hazard condition, a plurality of hazard conditions are taken into account in order to distinguish different hazard qualities or categories so as to select different safety functions according to the hazard qualities or categories.
2. The computer-implemented method of claim 1, The computer-implemented method further comprises the following steps performed by the computer hardware components: determining at least one vehicle motion parameter of the vehicle (18), in, The at least one hazard condition specifies a vehicle motion range (VR) of the at least one vehicle motion parameter.
3. The computer-implemented method of claim 2, The computer-implemented method further comprises the following steps performed by the computer hardware components: determining object motion parameters of the object (20) other than the orientation parameters (24), in, The at least one hazard condition (B, C, D, E) specifies an object motion range (OMR) of the object motion parameter.
4. The computer-implemented method of claim 3, in, The at least one hazard condition (B, C) specifies a predetermined relationship between the vehicle motion parameter and the object motion parameter.
5. The computer-implemented method of claim 1, in, The plurality of hazardous conditions include at least one pair of hazardous conditions (B, C) that are axially symmetric relative to a longitudinal axis (L) of the vehicle (18).
6. The computer-implemented method of claim 1, in, The multiple hazardous conditions include: a first hazard condition (A), the first hazard condition (A) specifying a first orientation range (OR) of the orientation parameter (24), and at least one second risk condition (B, C), the at least one second risk condition (B, C) specifying a second orientation range (OR) of the orientation parameter (24), Wherein, the first orientation range is at least partially lower than and / or smaller than the second orientation range.
7. The computer-implemented method of claim 6, in, The first hazard condition (A) further specifies a first vehicle motion range (VR) of a first vehicle motion parameter of the vehicle (18), wherein the at least one second dangerous condition (B) further specifies a second vehicle motion range (VR) of the first vehicle motion parameter, Therein, the first vehicle motion range is at least partially higher than and / or larger than the second vehicle motion range.
8. The computer-implemented method of claim 7, in, The at least one second hazard condition (B, C) further specifies a first object motion range (OMR) of a first object motion parameter of the object (20), the first object motion range (OMR) being defined relative to the first vehicle motion parameter.
9. A computer-implemented method according to claim 7 or 8, in, The multiple hazardous conditions include: At least one third danger condition (D, E), the at least one third danger condition (D, E) specifying a third orientation range (OR) of the orientation parameter (24), and a third vehicle motion range (VR) of the first vehicle motion parameter; Wherein the third orientation range (OR) is at least partially higher than the second orientation range (OR), and wherein the third vehicle motion range (VR) is at least partially lower and / or smaller than the second vehicle motion range (VR).
10. The computer-implemented method of claim 9, in, The at least one third hazard condition (D, E) further specifies a second vehicle motion range (VR) of a second vehicle motion parameter of the vehicle (18).
11. The computer-implemented method of claim 1 , in, If the object (20) is identified as a hazardous object in response to the at least one hazardous condition being satisfied, the computer-implemented method further comprises the following steps performed by the computer hardware components: - determining whether at least one additional risk condition associated with the at least one risk condition is satisfied; as well as - activating at least one safety function for controlling the vehicle (18) if the at least one additional risk condition is met.
12. The computer-implemented method of claim 2, in, The at least one vehicle motion parameter comprises a speed (V) of the vehicle (18) and / or a steering parameter of the vehicle (18), the steering parameter indicating a direction of steering.
13. The computer-implemented method of claim 3, in, The object motion parameters of the object (20) include a velocity of the object (20).
14. A computer system connectable to a sensor configured to capture the surroundings of a vehicle, and The computer system comprises a plurality of computer hardware components configured to perform the steps of the computer-implemented method according to any one of claims 1 to 13 based on data captured by the sensor.
15. A non-transitory computer-readable medium comprising instructions for executing the computer-implemented method according to any one of claims 1 to 13.
