Method and device for dynamic determination of safety distance and vehicle

Abstract

The embodiment of the application provides a kind of dynamic determination method, device and vehicle of safety distance, it is related to intelligent driving technical field.The method comprises: based on the multi-dimension data of vehicle is carried out fuzzy processing, obtains multi-dimension fuzzy result, and multi-dimension fuzzy result includes environment bad fuzzy result, perception quality fuzzy result, communication quality fuzzy result and motion stability fuzzy result;According to environment bad fuzzy result, perception quality fuzzy result, communication quality fuzzy result and motion stability fuzzy result carries out fuzzy reasoning, obtains credibility discount factor;Based on the target running data corresponding to the formation operation state of vehicle, determine initial safety distance;According to credibility discount factor, initial safety distance is adjusted, and target safety distance is obtained.The method is used to reach safety distance with working condition dynamic adjustment, improve the adaptive capacity of safety distance control to dynamic driving working condition.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, and in particular to a method, device and vehicle for dynamically determining a safe distance. Background Technology

[0002] In intelligent driving, it is necessary to determine the safe distance between vehicles in order to control the vehicle's braking, acceleration, or constant speed following. Therefore, the accuracy of the safe distance has a significant impact on the vehicle safety of intelligent driving.

[0003] In related technologies, safe distances are usually determined based on fixed parameter models of vehicle dynamics. However, actual driving scenarios are complex and varied. The safe distances determined by fixed parameter models may be insufficient or redundant, making it difficult to meet the needs of dynamically changing and complex operating conditions. Summary of the Invention

[0004] This application provides a method, apparatus, and vehicle for dynamically determining safe distance, so as to achieve dynamic adjustment of safe distance according to operating conditions and improve the adaptability of safe distance control to dynamic driving conditions.

[0005] In a first aspect, embodiments of this application provide a method for dynamically determining a safety distance, comprising:

[0006] The multidimensional data of the vehicle is fuzzed to obtain multidimensional fuzzification results, including environmental degradation fuzzification results, perception quality fuzzification results, communication quality fuzzification results, and motion stability fuzzification results.

[0007] Fuzzy inference is performed based on the fuzzification results of environmental severity, perception quality, communication quality, and motion stability to obtain the credibility discount factor;

[0008] The initial safe distance is determined based on the target operation data corresponding to the vehicle formation operation status;

[0009] The initial safety distance is adjusted based on the credibility discount factor to obtain the target safety distance.

[0010] In one possible embodiment, adjusting the initial safety distance based on a confidence discount factor to obtain the target safety distance includes:

[0011] The initial safe distance is adjusted based on the confidence discount factor to obtain the candidate safe distance;

[0012] When the vehicles are in a convoy and the following vehicle is in a tail-to-tail state, the candidate safe distance is taken as the target safe distance.

[0013] When the vehicles are in platooning with the lead vehicle in the lead vehicle position, the larger of the reaction distance and the candidate safe distance is taken as the target safe distance; the reaction distance is determined based on the vehicle's speed and safe travel distance.

[0014] In one possible embodiment, adjusting the initial safety distance based on a confidence discount factor to obtain the target safety distance includes:

[0015] Determine the target confidence interval to which the confidence discount factor belongs;

[0016] The first compensation coefficient is determined based on the target confidence interval. If the target confidence interval is a low confidence interval, the first compensation coefficient is greater than 1. If the target confidence interval is a low confidence interval, the first compensation coefficient is equal to 1. If the target confidence interval is a high confidence interval, the first compensation coefficient is less than 1.

[0017] The product of the first compensation coefficient and the initial safety distance is taken as the target safety distance.

[0018] In one possible embodiment, multi-dimensional data of the vehicle is subjected to fuzzification processing to obtain a multi-dimensional fuzzification result, including:

[0019] Determine the environmental severity score based on the vehicle's environmental data;

[0020] The perceived quality score is determined based on the vehicle's perception data.

[0021] The communication quality score is determined based on the vehicle's communication data;

[0022] The motion stability score is determined based on the vehicle's motion stability data.

[0023] The environmental severity score, perception quality score, communication quality score, and motion stability score are fuzzified to obtain the membership degrees under multiple environmental severity fuzzy levels, multiple perception quality fuzzy levels, multiple communication quality fuzzy levels, and multiple motion stability fuzzy levels.

[0024] In one possible embodiment, the environmental degradation fuzzification result includes membership degrees corresponding to multiple environmental degradation fuzzification levels, the perception quality fuzzification result includes membership degrees corresponding to multiple perception quality fuzzification levels, the communication quality fuzzification result includes membership degrees corresponding to multiple communication quality fuzzification levels, and the motion stability fuzzification result includes membership degrees corresponding to multiple motion stability fuzzification levels.

[0025] Based on the fuzzy results of environmental degradation, perception quality, communication quality, and motion stability, fuzzy inference is performed to obtain a credibility discount factor, including:

[0026] The activation intensity of multiple preset fuzzy rules is determined based on the membership degrees corresponding to multiple environmental severity fuzzy levels, multiple perception quality fuzzy levels, multiple communication quality fuzzy levels, and multiple motion stability fuzzy levels.

[0027] Defuzzing is performed based on the activation intensity of multiple preset fuzzy rules and the preset center value corresponding to the credibility fuzziness level of multiple preset fuzzy rules to obtain the credibility discount factor.

[0028] In one possible embodiment, the initial safe distance is determined based on target operation data corresponding to the vehicle platooning status, including:

[0029] When the vehicles are in a convoy and the following vehicle is in a tail-to-tail state, the initial safe distance is determined based on the speed of the preceding vehicle, the speed of the following vehicle, the load factor, and the road gradient.

[0030] When the vehicles are in platooning with the lead vehicle in the lead vehicle position, the initial safe distance is determined based on the vehicle speed, load factor, and road gradient.

[0031] In one possible embodiment, the initial safe distance is determined based on the speed of the vehicle in front, the speed of the vehicle behind, the load factor, and the road gradient, including:

[0032] The deceleration threshold is adjusted based on the load factor and road slope to obtain the equivalent deceleration;

[0033] The first braking distance is determined based on the speed of the vehicle in front, the speed of the vehicle itself, and the equivalent deceleration.

[0034] The reaction distance is determined based on the vehicle's speed and safe travel distance.

[0035] The initial safety distance is determined based on the first braking distance, the reaction distance, and the preset safety margin.

[0036] In one possible embodiment, when the vehicle platoon is in a lead vehicle mode, an initial safe distance is determined based on the vehicle speed, load factor, and road gradient, including:

[0037] When the lead vehicle is obstructed during convoy operation, the deceleration threshold is adjusted based on the load factor and road gradient to obtain the equivalent deceleration.

[0038] The second braking distance is determined based on the vehicle speed and equivalent deceleration;

[0039] The reaction distance is determined based on the vehicle's speed and safe travel distance.

[0040] The initial safety distance is determined based on the second braking distance, the reaction distance, and the preset safety margin.

[0041] In one possible embodiment, when the vehicle platoon is in a lead vehicle mode, an initial safe distance is determined based on the vehicle speed, load factor, and road gradient, including:

[0042] When the vehicle platoon is in the lead vehicle passing state, the acceleration threshold is adjusted according to the load factor and road gradient to obtain the equivalent acceleration.

[0043] The acceleration distance is determined based on the equivalent acceleration, vehicle speed, and acceleration time.

[0044] The reaction distance is determined based on the vehicle's speed and safe travel distance.

[0045] The initial safety distance is determined based on the acceleration distance, reaction distance, and preset safety margin.

[0046] Secondly, embodiments of this application provide a device for dynamically determining a safe distance, the device comprising:

[0047] The fuzzing processing module is used to perform fuzzing processing on multi-dimensional vehicle data to obtain multi-dimensional fuzzing results, including environmental degradation fuzzing results, perception quality fuzzing results, communication quality fuzzing results, and motion stability fuzzing results.

[0048] The fuzzy inference module is used to perform fuzzy inference based on the fuzzification results of environmental degradation, perception quality, communication quality, and motion stability to obtain the credibility discount factor.

[0049] The distance determination module is used to determine the initial safe distance based on the target running data corresponding to the vehicle formation running status;

[0050] The distance adjustment module is used to adjust the initial safety distance based on the confidence discount factor to obtain the target safety distance.

[0051] Thirdly, embodiments of this application provide a vehicle, including: a processor, and a memory communicatively connected to the processor;

[0052] The memory stores instructions that the computer executes;

[0053] The processor executes computer execution instructions stored in memory to implement the methods provided above.

[0054] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods provided above.

[0055] Fifthly, embodiments of this application provide a computer program product, including computer execution instructions, which, when executed by a processor, implement the method provided above.

[0056] The method, apparatus, and vehicle for dynamically determining safe distance provided in this application embodiment fuzzify multi-dimensional vehicle data to obtain fuzzification results for environmental adverseness, perception quality, communication quality, and motion stability. These results are then converted into a credibility discount factor to quantify system reliability through fuzzy inference. The credibility discount factor characterizes the comprehensive confidence level of the vehicle across multiple dimensions, including environment, perception, communication, and motion stability. The distance is determined in real-time based on target operation data corresponding to the vehicle's platooning status. An initial safety distance is defined, but this initial safety distance is not a fixed threshold. This improves the adaptability of the initial safety distance to vehicle dynamics and actual driving scenarios. A credibility discount factor is used to adjust the initial safety distance, incorporating adverse environmental conditions, perception quality, communication quality, and motion stability into the safety distance calculation. This means that the safety distance is no longer determined solely by fixed dynamic parameters, but is adjusted in real time based on the system credibility under the current driving conditions. This improves the problem of insufficient adaptability of safety distances determined by fixed parameter models to complex driving conditions, enabling dynamic adjustment of the safety distance according to driving conditions and enhancing the dynamic adaptability of safety distance control to complex driving conditions. Attached Figure Description

[0057] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0058] Figure 1 A flowchart illustrating the dynamic determination method for the safety distance provided in this application;

[0059] Figure 2 A schematic diagram of the membership function provided in this application;

[0060] Figure 3 A schematic diagram of the fuzzy inference system provided in this application;

[0061] Figure 4 A schematic diagram of the unmanned mining truck safety distance system provided in this application;

[0062] Figure 5 A schematic diagram of the structure of the dynamic determination device for safe distance provided in this application;

[0063] Figure 6 This is a structural diagram of the vehicle provided in this application.

[0064] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0065] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0066] In the safety control of intelligent driving (including driverless and autonomous driving), it is necessary to determine the safe distance between vehicles in order to control the vehicle's braking, acceleration, or constant speed following.

[0067] In related technologies, safe distances are usually determined based on fixed parameter models of vehicle dynamics. This approach is applicable to environments with stable road conditions, reliable perception, normal communication, and minimal changes in vehicle status. However, in complex operating conditions where vehicles are often in dynamic changes, such as when the driving environment is affected by various factors including weather, road conditions, traffic participants, and communication quality, using fixed parameter models to determine safe distances can result in insufficient or redundant safe distances, making it difficult to meet the requirements of dynamic and complex operating conditions.

[0068] For example, in production and transportation scenarios, vehicles operate in platooning or single-vehicle cruising modes. The safety distance serves as a constraint on the following vehicle's following and the lead vehicle's acceleration. If the safety distance is insufficient, it will increase the probability of collisions, affecting vehicle safety and overall production and transportation safety. If the safety distance is redundant, it will lead to large platoon spacing, reducing platooning transportation efficiency and production efficiency. Therefore, in production and transportation scenarios, it is difficult to dynamically adapt to complex and ever-changing production and transportation scenarios by using a fixed parameter model to determine the safety distance.

[0069] To address the aforementioned issues, a dynamic method for determining safe distance is proposed. This method involves fuzzifying multi-dimensional vehicle data to obtain fuzzification results for environmental severity, perception quality, communication quality, and motion stability. Fuzzy inference is then performed based on these fuzzification results to derive a confidence discount factor. An initial safe distance is determined based on target operational data corresponding to the vehicle platoon's operating state. Finally, the initial safe distance is adjusted according to the confidence discount factor to obtain the target safe distance. This technical approach...

[0070] By employing fuzzy reasoning on environmental conditions, perception quality, communication quality, and motion stability, a credibility discount factor is obtained under the current driving conditions. This credibility discount factor is then used to adjust the initial safety distance, incorporating adverse environmental conditions, perception quality, communication quality, and motion stability into the safety distance calculation. This allows the safety distance to be dynamically adjusted based on the current driving conditions, rather than being solely determined by fixed dynamic parameters. For example, when environmental conditions, perception quality, communication quality, and motion stability are relatively good, the target safety distance is appropriately reduced to improve vehicle operating efficiency. Conversely, when environmental conditions, perception quality, communication quality, and motion stability are relatively deteriorated, the target safety distance is correspondingly increased to enhance safety redundancy. This addresses the issue of insufficient adaptability of safety distances determined by fixed parameter models to complex driving conditions, enabling dynamic adjustment of the safety distance according to driving conditions and improving the adaptability of safety distance control to dynamic driving conditions.

[0071] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0072] Figure 1 A flowchart illustrating the dynamic determination method for safe distance provided in this application. This method can be applied to vehicles, such as... Figure 1 As shown, the dynamic determination method for safe distance includes:

[0073] S10. Perform fuzzification processing on the multi-dimensional data of the vehicle to obtain multi-dimensional fuzzification results, including environmental degradation fuzzification results, perception quality fuzzification results, communication quality fuzzification results, and motion stability fuzzification results.

[0074] Among them, the vehicle can be an autonomous driving vehicle or a driverless vehicle; in practical applications, driverless vehicles can be driverless transport vehicles, that is, driverless vehicles that perform transportation tasks on roads or in production scenarios, such as driverless mining trucks used for transportation in mining areas, or driverless freight trucks used for transportation on roads or in factory areas.

[0075] Multidimensional data refers to a collection of heterogeneous data from multiple sources used to characterize the vehicle's operating environment, perception system status, communication link status, and the vehicle's own dynamics.

[0076] Fuzzification refers to the process of mapping continuously changing physical quantities with different dimensions and uncertain boundaries into fuzzy linguistic variables through a preset membership function.

[0077] The environmental degradation fuzzification result is used to characterize the impact of the current operating environment on vehicle safety; the perception quality fuzzification result is used to characterize the reliability of the perception results of the perception system; the communication quality fuzzification result is used to characterize the stability of the vehicle-to-ground communication link and the vehicle-to-vehicle cooperative link; and the motion stability fuzzification result is used to characterize the stability of the vehicle's current driving.

[0078] In practice, the vehicle controller receives raw data from environmental sensors, vehicle radar, vehicle camera, inertial measurement unit, wheel speed sensor, brake actuator feedback unit, and vehicle communication module. The raw data includes environmental data, perception data, communication data, and motion stability data.

[0079] The environmental data may include, but is not limited to: dust concentration, light intensity, weather information, and proximity to slopes; the perception data may include target recognition accuracy, continuous frame tracking stability, point cloud integrity, image clarity, and image occlusion rate; the communication data may include communication latency, signal strength, packet loss rate, and number of retransmissions; and the motion stability data may include longitudinal acceleration, yaw rate, vehicle pitch angle, and tire slip rate.

[0080] Specifically, environmental data, perception data, communication data, and motion stability data are scored separately to obtain environmental severity score, perception quality score, communication quality score, and motion stability score. Then, through a preset membership function, the scores are mapped to obtain fuzzy results for environmental severity, perception quality, communication quality, and motion stability. The membership function can be a triangular membership function, a trapezoidal membership function, or a Gaussian membership function.

[0081] The environmental severity fuzzification results include the membership degree of the environmental severity score under multiple environmental severity fuzziness levels. Optionally, the environmental severity fuzziness levels are divided into three levels: low, medium, and high.

[0082] The result of perceptual quality fuzzification includes the membership degree of the perceptual quality score under multiple perceptual quality fuzziness levels. Optionally, the perceptual quality fuzziness levels are divided into three levels: low, medium, and high.

[0083] The notification quality fuzzification results include the membership degree of the communication quality score under multiple communication quality fuzzy levels. Optionally, the communication quality fuzzy levels are divided into three levels: unstable, general, and stable.

[0084] The motion stability fuzzification result includes the membership degree of the motion stability score under multiple motion stability fuzziness levels. Optionally, the motion stability fuzziness levels are divided into three levels: unstable, general, and stable.

[0085] For example, if at least one of the following conditions is met: long communication delay, low signal strength, high packet loss rate, and many retransmissions, the communication quality score is low. Correspondingly, the communication quality fuzzy level is unstable with a high degree of membership, while the communication quality fuzzy level is stable with a low degree of membership.

[0086] For example, if at least one of the following conditions is met: large longitudinal acceleration, large yaw rate, large vehicle pitch angle, and large tire slip ratio, the motion stability score is small. Correspondingly, the motion stability fuzzy level is unstable and has a high degree of membership, while the motion more stable fuzzy level is stable and has a low degree of membership.

[0087] S20. Based on the fuzzy results of environmental degradation, perception quality, communication quality, and motion stability, perform fuzzy inference to obtain the credibility discount factor.

[0088] Fuzzy reasoning is an uncertainty reasoning method that uses pre-established fuzzy rules to reason about multiple fuzzy input variables and obtain an approximate result with fuzziness.

[0089] The confidence discount factor is used to characterize the overall confidence level of a vehicle across multiple dimensions, including environment, perception, communication, and motion stability. The confidence discount factor can be represented by a continuous value between 0 and 1. The larger the value, the higher the confidence level of the current operating state, and the smaller the value, the lower the confidence level of the current operating state.

[0090] In practice, a fuzzy rule base is pre-built, which includes multiple preset fuzzy rules. The input of each preset fuzzy rule is a combination of environmental severity fuzziness level, perception quality fuzziness level, communication quality fuzziness level and motion stability fuzziness level, and the output is the credibility fuzziness level.

[0091] For example, the input to a preset fuzzy rule is: the fuzzy level of the harsh environment is high, the fuzzy level of the perception quality is low, the fuzzy level of the communication quality is unstable, and the fuzzy level of the motion stability is unstable. The output of the preset fuzzy rule is: the fuzzy level of the credibility is low.

[0092] Specifically, based on the fuzzification results of environmental degradation, perception quality, communication quality, and motion stability, the fuzzy inference engine calculates the activation intensity of each preset fuzzy rule. It then defuzzifies the rules based on the activation intensity of each preset fuzzy rule and the preset center value corresponding to the credibility level of the preset fuzzy rule, obtaining a credibility discount factor. In practical applications, the fuzzy inference engine can be implemented using Mamdani fuzzy inference.

[0093] In a specific example, the vehicle is an unmanned mining truck, which performs transportation tasks in an open-pit mine; Reference Figure 3 The credibility discount factor is determined through a fuzzy inference system, which includes a fuzzer, a fuzzy inference engine, and a defuzzifier.

[0094] During the operation of the unmanned mining truck, environmental data, perception data, communication data, and motion stability data are acquired in real time. An environmental severity score is determined based on the environmental data; a perception quality score is determined based on the vehicle's perception data; a communication quality score is determined based on the vehicle's communication data; and a motion stability score is determined based on the motion stability data. A fuzzy logic device (FLP) uses a triangular membership function to fuzzify the environmental severity score, perception quality score, communication quality score, and motion stability score, resulting in fuzzified environmental severity score, perception quality score, communication quality score, and motion stability score. A fuzzy inference engine then calls a fuzzy rule base to perform fuzzy inference on these results, followed by defuzzification to obtain a credibility discount factor.

[0095] This step uses fuzzy reasoning to determine the credibility discount factor based on the fuzzification results of environmental degradation, perception quality, communication quality, and motion stability. This achieves the transformation from multi-dimensional state perception to the quantification of system reliability. The credibility discount factor reflects the comprehensive impact of environmental disturbances, perception disturbances, communication anomalies, and vehicle instability on safety control, thus improving the comprehensiveness of the credibility discount factor.

[0096] S30. Determine the initial safe distance based on the target operation data corresponding to the vehicle platooning operation status.

[0097] The formation operation status refers to the operation status of vehicles in traffic flow, including but not limited to: following vehicles, the lead vehicle being blocked, and the lead vehicle passing through.

[0098] For example, when the vehicle is an unmanned mining truck, the unmanned mining truck performs transportation operations in an open-pit mine. Multiple unmanned mining trucks transport in a convoy. The following vehicles in the convoy are in a following state, while the lead vehicle in the convoy may be in a state of obstruction (e.g., the lead vehicle is about to reach an intersection and needs to stop) or a state of passage (e.g., the lead vehicle is passing normally in the road).

[0099] Among them, target operation data refers to dynamic data that matches the current formation operation status and is used to calculate the safe distance.

[0100] The initial safety distance refers to the basic safety distance determined solely based on dynamic data without considering the confidence discount factor correction.

[0101] In practice, the platooning status of vehicles is determined based on map data, the status of the vehicle in front, and the status of the vehicle itself.

[0102] For example, if the vehicle and the vehicle in front are in the same lane, and the heading angle and lateral deviation of the vehicle and the vehicle in front are both within a preset range, the formation operation state is determined to be the following vehicle state; if there are no vehicles in front of the vehicle and there is an intersection within a preset distance in front, the formation operation state is determined to be the lead vehicle obstructed state; if there are no vehicles in front of the vehicle and there is no intersection within a preset distance in front, the formation operation state is determined to be the lead vehicle passing state.

[0103] After determining the platooning status of the vehicles, the corresponding target operating data is obtained. When the platooning status is that the following vehicles are in a following state, the target operating data includes: the speed of the leading vehicle, the speed of the following vehicle, the load factor, and the road gradient. When the platooning status is that the leading vehicle is blocked or the leading vehicle is passing, the target operating data includes: the speed of the leading vehicle, the speed of the following vehicle, the load factor, and the road gradient.

[0104] The load factor is calculated by feedback from the vehicle weighing system or hydraulic suspension, and the road gradient can be obtained from map data or determined by fusing map data and inertial navigation attitude information.

[0105] After obtaining the target running data, the basic safe distance model corresponding to the formation running state is called to calculate the initial safe distance.

[0106] Optionally, the basic safety distance model can be obtained by combining the safety time-distance model and the braking distance model. The reaction distance is calculated by the safety time-distance model, and the braking distance is calculated by the braking distance model. The sum of the reaction distance, braking distance and static margin is used as the initial safety distance.

[0107] It should be noted that the initial safe distance is determined in real time by the vehicle formation operation status and target operation data. This initial safe distance is not a fixed threshold, so that the subsequent confidence discount factor correction is based on the actual vehicle dynamics and driving scenario, thereby improving the adaptability of the target safe distance to actual driving conditions.

[0108] S40. Adjust the initial safety distance according to the credibility discount factor to obtain the target safety distance.

[0109] The target safety distance is obtained by adjusting the initial safety distance based on the credibility discount factor, that is, by comprehensively considering vehicle dynamics constraints and system credibility level; the target safety distance is used for vehicle car-following control, braking control or speed planning control.

[0110] In some embodiments, adjusting the initial security distance based on a credibility discount factor to obtain a target security distance includes: determining the target credibility interval to which the credibility discount factor belongs; determining a first compensation coefficient based on the target credibility interval, wherein if the target credibility interval is a low credibility interval, the first compensation coefficient is greater than 1; if the target credibility interval is a low credibility interval, the first compensation coefficient is equal to 1; if the target credibility interval is a high credibility interval, the first compensation coefficient is less than 1; and using the product of the first compensation coefficient and the initial security distance as the target security distance.

[0111] The confidence discount factor ranges from [0,1]; the low confidence interval, medium confidence interval, and high confidence interval are sub-intervals within this range; the actual values ​​of the low confidence interval, medium confidence interval, and high confidence interval can be set according to requirements; for example, the low confidence interval is [0, 0.6), the medium confidence interval is [0.6, 0.8), and the high confidence interval is [0.8, 1].

[0112] Specifically, within the low-confidence, medium-confidence, and high-confidence intervals, a target confidence interval is determined to which the confidence discount factor belongs. A first compensation coefficient is then determined based on this target confidence interval. Specifically, if the confidence discount factor belongs to the high-confidence interval, the first compensation coefficient is less than 1; if it belongs to the medium-confidence interval, the first compensation coefficient is 1; and if it belongs to the low-confidence interval, the first compensation coefficient is greater than 1. The product of the first compensation coefficient and the initial safety distance is used as the target safety distance. Thus, when the confidence discount factor belongs to the high-confidence interval, the initial safety distance is reduced; when it belongs to the medium-confidence interval, the initial safety distance is not adjusted; and when it belongs to the low-confidence interval, the initial safety distance is increased.

[0113] Specifically, when the confidence discount factor belongs to the high confidence interval, the first compensation coefficient can be a pre-set fixed value corresponding to the high confidence interval. For example, the first compensation coefficient corresponding to the high confidence interval can be set to 0.9. The first compensation coefficient can also be further determined based on the vehicle speed. For example, when the confidence discount factor belongs to the high confidence interval, the first compensation coefficient is set to 0.9 when the vehicle speed is greater than the preset speed, and to 0.85 when the vehicle speed is not greater than the preset speed. In the above example, by setting the first compensation coefficient corresponding to the high confidence interval, the initial safety distance is appropriately reduced.

[0114] When the confidence discount factor belongs to the low confidence interval, the first compensation coefficient can be a pre-set fixed value corresponding to the low confidence interval. For example, the first compensation coefficient corresponding to the low confidence interval can be set to 1.2. The first compensation coefficient can also be further determined according to the vehicle speed. For example, when the confidence discount factor belongs to the low confidence interval, the first compensation coefficient is set to 1.4 when the vehicle speed is greater than the preset vehicle speed, and to 1.2 when the vehicle speed is not greater than the preset vehicle speed. In the above example, the initial safety distance is appropriately increased by setting the first compensation coefficient corresponding to the low confidence interval.

[0115] In the above embodiments, a second compensation coefficient is determined based on the confidence interval to which the confidence discount factor belongs. The initial safety distance is adjusted using the second compensation coefficient to appropriately reduce the safety distance when the system has high confidence, thereby increasing operating efficiency. When the system has low confidence, the safety distance is appropriately increased to increase safety redundancy, thus improving the dynamic adaptability of safety distance control to complex driving conditions.

[0116] In some embodiments, adjusting the initial safety distance based on a credibility discount factor to obtain a target safety distance includes: taking the reciprocal of the credibility discount factor to obtain a second compensation coefficient; and using the product of the second compensation coefficient and the initial safety distance as the target safety distance.

[0117] Specifically, the credibility discount factor is The second compensation coefficient is The product of the second compensation coefficient and the initial safety distance is taken as the target safety distance. Since the confidence discount factor ranges from [0,1], the second compensation coefficient is greater than or equal to 1. When the confidence discount factor is less than 1, the second compensation coefficient is greater than 1. The initial safety distance is amplified by the second compensation coefficient to obtain the target safety distance. When the confidence discount factor is equal to 1, it indicates that the confidence is high. The second compensation coefficient is 1, so the initial safety distance is not adjusted.

[0118] Optionally, the second compensation coefficient can be restricted by setting a second maximum value and a second minimum value to constrain the degree of adjustment of the initial safety distance. For example, the second maximum value is set to 1.5 and the second minimum value is 1, that is, the value range of the second compensation coefficient is [1, 1.5]. When the confidence discount factor is less than 1, the second compensation coefficient is greater than 1 and less than 1.5, so that the initial safety distance is appropriately amplified.

[0119] In the above embodiments, the second compensation coefficient is the reciprocal of the reliability discount factor. When the system is highly reliable, the initial safety distance is not changed, and when the system is reliable or low reliable, the initial safety distance is increased. This can increase the safety distance and improve rear-end collision protection capability and ensure driving safety when the environment is harsh, perception is disturbed, communication is abnormal, and motion stability is reduced.

[0120] Optionally, to prevent the adjusted target safety distance from being unreasonable, a lower limit and an upper limit value can be set to constrain the target safety distance. The lower limit value is used to ensure that the target safety distance is not less than the minimum controllable stopping distance and the minimum stable following distance under any circumstances, while the upper limit value is used to avoid redundancy and excessive dispersion of the formation due to an excessively large target safety distance.

[0121] Optionally, the target safe distance is updated continuously in a loop. The update cycle can be consistent with the perception and control cycle, for example, once every 100 to 200 milliseconds. In each loop, multi-dimensional data is re-collected, multi-dimensional fuzzing results are updated, confidence discount factor and initial safe distance are recalculated, and a new target safe distance is output to meet the dynamically changing driving scenarios.

[0122] This application provides a method for dynamically determining safe distance. By fuzzifying multi-dimensional vehicle data, it obtains fuzzified results for environmental adverseness, perception quality, communication quality, and motion stability. Then, through fuzzy inference, these results are converted into a credibility discount factor that quantifies the system's reliability. The credibility discount factor characterizes the comprehensive confidence level of the vehicle across multiple dimensions (environment, perception, communication, and motion stability). The initial safe distance is determined in real-time using target operation data corresponding to the vehicle's platooning status. The initial safe distance is not a fixed threshold, which improves the adaptability of the initial safe distance to vehicle dynamics and actual driving scenarios. The initial safe distance is adjusted by a credibility discount factor, which incorporates adverse environmental conditions, perception quality, communication quality and motion stability into the safe distance calculation. This means that the safe distance is no longer determined solely by fixed dynamic parameters, but is adjusted in real time based on the system credibility under the current driving conditions. This improves the problem of insufficient adaptability of the safe distance determined by the fixed parameter model to complex driving conditions, and realizes the dynamic adjustment of the safe distance according to the driving conditions, thereby improving the dynamic adaptability of the safe distance control to complex driving conditions.

[0123] It should be noted that when the dynamic determination method for safe distance provided in this application is applied to unmanned mining trucks, since the unmanned mining trucks operate in open-pit mines, there are complex driving scenarios in open-pit mines, such as fluctuations in dust concentration, sudden changes in light, communication obstruction, link delay, load changes, and slope fluctuations. By using a credibility discount factor, the impact of perception quality, communication quality, harsh environmental conditions, and motion stability on safety control is quantified. The initial safe distance is determined by the platooning operation status of the unmanned mining trucks, and then the initial safe distance is adaptively corrected according to the credibility discount factor. This achieves comprehensive and adaptive adjustment of the safe distance, reducing the risk of rear-end collisions caused by perception deviations, communication failures, and changes in braking capabilities, while also mitigating the problems of increased platooning spacing and decreased transportation efficiency caused by excessive conservatism.

[0124] In one possible implementation, S10, fuzzification processing is performed based on the vehicle's multi-dimensional data to obtain multi-dimensional fuzzification results, including: S101, determining an environmental severity score based on the vehicle's environmental data; S102, determining a perception quality score based on the vehicle's perception data; S103, determining a communication quality score based on the vehicle's communication data; S104, determining a motion stability score based on the vehicle's motion stability data; S105, fuzzification processing is performed on the environmental severity score, perception quality score, communication quality score, and motion stability score respectively to obtain membership degrees under multiple environmental severity fuzziness levels, multiple perception quality fuzziness levels, multiple communication quality fuzziness levels, and multiple motion stability fuzziness levels.

[0125] Among them, the environmental severity score is used to characterize the degree to which the environment in which the vehicle is located has an adverse impact on the vehicle's operation; the perception quality score is used to characterize the reliability of the on-board sensors in identifying targets; the communication quality score is used to characterize the interaction stability of the vehicle-to-ground and vehicle-to-vehicle links; and the motion stability score is used to characterize the attitude stability of the vehicle's motion.

[0126] The environmental data includes, but is not limited to: weather information, dust concentration, light intensity, slope distance, and obstacle distance. In practice, weather information can be obtained through vehicle-mounted meteorological sensors or through the network; dust concentration can be obtained through vehicle-mounted dust sensors; light intensity can be measured through vehicle-mounted illuminance sensors; and slope distance and obstacle distance can be obtained from radar or fusion perception results.

[0127] Perception data includes, but is not limited to: point cloud data collected by vehicle-mounted radar and images collected by vehicle-mounted cameras.

[0128] Communication data includes, but is not limited to: communication latency and packet loss rate; in specific implementation, the vehicle communication unit determines the communication latency and packet loss rate in real time based on the message transmission and reception data.

[0129] Motion stability data includes, but is not limited to: vehicle pitch angle and tire slip ratio; in practice, the vehicle pitch angle is obtained through an inertial measurement unit and the tire slip ratio is obtained through wheel speed sensors.

[0130] Specifically, weather scores are obtained based on weather information, and the correspondence between weather information and weather scores is predetermined. For example, the weather score for a sunny day is lower than that for a rainy day. Dust scores are determined based on dust concentration, and the mapping relationship between dust concentration and dust scores is predetermined; the higher the dust concentration, the higher the corresponding dust score. Light scores are determined based on light intensity, and the correspondence between light intensity and light scores is predetermined. If the light intensity is too strong or too weak, the light score is higher; if the light intensity is moderate, the light score is lower. Slope scores are determined based on slope distance, and the mapping relationship between slope distance and slope scores is predetermined. If the slope distance is large, the slope score is low; if the slope distance is small, the slope score is high. Obstacle scores are determined based on obstacle distance, and the mapping relationship between obstacle distance and obstacle scores is predetermined. If the obstacle distance is large, the obstacle score is low; if the obstacle distance is small, the obstacle score is high.

[0131] The environmental severity score is obtained by weighted summation of the weather score, dust score, light score, slope score, and obstacle score. The weights of the weather score, dust score, light score, slope score, and obstacle score can be set according to actual needs.

[0132] The point cloud missing rate is determined based on point cloud data collected by the vehicle-mounted radar, and the image sharpness is determined based on images collected by the vehicle-mounted camera. The radar quality score is determined based on the point cloud missing rate, and the visual quality score is determined based on the image sharpness. The radar quality score and the visual quality score are then weighted and summed to obtain the perception quality score. The mapping relationship between the point cloud missing rate and the radar quality score is predetermined; a lower point cloud missing rate results in a higher radar quality score. Similarly, the mapping relationship between image sharpness and the visual quality score is predetermined; higher image sharpness results in a higher visual quality score. The weights of the radar quality score and the visual quality score can be set according to actual needs.

[0133] A first communication score is determined based on communication latency, and a second communication score is determined based on packet loss rate. The first and second communication scores are then weighted and summed to obtain a communication quality score. The mapping relationship between communication latency and the first communication score is predetermined; the higher the communication latency, the lower the first communication score. Similarly, the mapping relationship between packet loss rate and the second communication score is predetermined; the higher the packet loss rate, the lower the second communication score. The weights of the first and second communication scores can be set according to actual needs.

[0134] The bump score is determined based on the vehicle pitch angle, and the slip score is determined based on the tire slip ratio. The motion stability score is obtained by weighted summation of the bump score and the slip score. The mapping relationship between the vehicle pitch angle and the bump score is predetermined; the lower the vehicle pitch angle, the lower the bump score. The mapping relationship between the tire slip ratio and the slip score is also predetermined; the lower the tire slip ratio, the lower the slip score. The weights of the bump score and the slip score can be set according to actual needs.

[0135] For example, the factors affecting the environmental severity score, perceived quality score, communication stability score, and motion stability score are shown in Table 1.

[0136] Table 1

[0137]

[0138] Using triangular membership functions, the environmental severity score, perception quality score, communication stability score, and motion stability score are fuzzified respectively, to obtain the membership degrees of the environmental severity score under multiple environmental severity fuzziness levels, the perception quality score under multiple perception quality fuzziness levels, the communication stability score under multiple communication stability fuzziness levels, and the motion stability score under multiple motion stability fuzziness levels.

[0139] The following example illustrates how the membership degree of the environmental severity score under multiple environmental severity fuzzy levels is obtained by fuzzifying the environmental severity score using the triangular membership function. The universe of discourse of the environmental severity score is [0,10]. The multiple environmental severity fuzzy levels include: E_L (low), E_M (medium), and E_H (high). The membership degree of the environmental severity score under multiple environmental severity fuzzy levels is calculated according to formula (1).

[0140] Formula (1):

[0141]

[0142] in, It is a score based on the harsh environment. It represents the membership degree of the environmental severity score under the fuzzy environmental severity level E_L (low). It is the membership degree of the environmental severity score under the environmental severity fuzzy level E_M (medium). It is the membership degree of the environmental severity score under the environmental severity fuzzy level E_H (high).

[0143] For example, taking an unmanned mining truck as an example, the relationship between the environmental severity score, perception quality score, communication stability score, motion stability score and their respective fuzzy levels is shown in Table 2.

[0144] Table 2

[0145]

[0146] In the above embodiments, multi-dimensional data is quantified into scores, and then the fuzziness level of the multi-dimensional data is determined through fuzzification processing. This avoids the impact of instantaneous anomalies of single data on the control results, improves the robustness of safety control, and the multi-dimensional fuzziness level is dynamically updated as the vehicle driving scenario changes, improving the adaptability of safety control under dynamic and complex driving conditions.

[0147] In one possible implementation, the environmental severity fuzzification result includes membership degrees corresponding to multiple environmental severity fuzziness levels, the perception quality fuzzification result includes membership degrees corresponding to multiple perception quality fuzziness levels, the communication quality fuzzification result includes membership degrees corresponding to multiple communication quality fuzziness levels, and the motion stability fuzzification result includes membership degrees corresponding to multiple motion stability fuzziness levels; S20, perform fuzzy inference based on the environmental severity fuzzification result, perception quality fuzzification result, communication quality fuzzification result, and motion stability fuzzification result to obtain a credibility discount factor, including: S201, determining the activation intensity of multiple preset fuzzy rules based on the membership degrees corresponding to multiple environmental severity fuzziness levels, multiple perception quality fuzziness levels, multiple communication quality fuzziness levels, and multiple motion stability fuzziness levels; S202, determining the activation fuzzy rule among the multiple preset fuzzy rules based on the activation intensity; S203, performing defuzzification processing based on the activation intensity of the activation fuzzy rule and the preset center value corresponding to the credibility fuzziness level of the activation fuzzy rule to obtain the credibility discount factor.

[0148] The preset fuzzy rules define the mapping rules between the input and the output. The inputs are: environmental severity fuzzy level, perception quality fuzzy level, communication quality fuzzy level, and motion stability fuzzy level. The output is: credibility fuzzy level. Optionally, the credibility fuzzy level includes low α_L, medium α_M, and high α_H.

[0149] The activation strength of the preset fuzzy rule is calculated based on the membership degree corresponding to the input of the preset fuzzy rule.

[0150] Activate fuzzy rules: among multiple preset fuzzy rules, activate the preset fuzzy rules with an intensity greater than 0.

[0151] The preset center value corresponding to the credibility fuzzy level is pre-set. In the membership function of the credibility fuzzy level, the preset center value is the horizontal coordinate corresponding to the maximum value of the membership degree of the credibility fuzzy level.

[0152] Specifically, a preset fuzzy rule is constructed in advance based on the following constraints: if two or more of the environmental severity fuzzy level, perception quality fuzzy level, communication quality fuzzy level, and motion stability fuzzy level deteriorate, the credibility fuzzy level α is downgraded; if three or more of the fuzzy levels deteriorate, the credibility fuzzy level α is low.

[0153] For each preset fuzzy rule, the inputs are: environmental severity fuzziness level E, perception quality fuzziness level P, communication quality fuzziness level C, and motion stability fuzziness level V, and the output is the credibility fuzziness level A. Based on the membership degrees corresponding to environmental severity fuzziness level E, perception quality fuzziness level P, communication quality fuzziness level C, and motion stability fuzziness level V, the minimum value is obtained by performing a bitwise AND operation to obtain the activation intensity w of the preset fuzzy rule. The activation intensity of each preset fuzzy rule is determined in the same way.

[0154] For example, the activation intensity of the preset fuzzy rule is determined according to formula (2).

[0155] Formula (2): ;

[0156] in, Ei is the activation strength of the i-th preset fuzzy rule, and Ei is the fuzziness level corresponding to the severe environment of the i-th preset fuzzy rule. Ei represents the membership degree, and Pi represents the perceptual quality fuzziness level corresponding to the i-th preset fuzzy rule. Pi represents the membership degree, and Ci represents the communication quality fuzzy level corresponding to the i-th preset fuzzy rule. Ci represents the membership degree; Vi represents the motion-stability fuzziness level corresponding to the i-th preset fuzzy rule. It is the membership degree corresponding to Vi.

[0157] Among multiple preset fuzzy rules, the preset fuzzy rules with activation strength greater than 0 are taken as the activated fuzzy rules; the preset center value corresponding to the output confidence fuzziness level A of the activated fuzzy rules is obtained; based on the activation strength of the activated fuzzy rules and the preset center value corresponding to the confidence fuzziness level A, the center averaging method is used for defuzzification to obtain the confidence discount factor.

[0158] For example, the credibility discount factor is calculated according to formula (3).

[0159] Formula (3): ;

[0160] in, It is a credibility discount factor. It is the total number of fuzzy rules activated. It is the first The activation strength of a fuzzy activation rule. It is the first The preset center value corresponding to the credibility fuzziness level of the activated fuzzy rule.

[0161] In a specific example, the preset fuzzy rules contain 15 rules, described using if-then statements:

[0162] R1: If E is E_L and P is P_G and C is C_S and V is V_S Then α is α_H;

[0163] R2: If E is E_L and P is P_G and C is C_M and V is V_S Then α is α_H;

[0164] R3: If E is E_M and P is P_G and C is C_S and V is V_S Then α is α_H;

[0165] R4: If E is E_L and P is P_N and C is C_M and V is V_M Then α is α_M;

[0166] R5: If E is E_M and P is P_N and C is C_M and V is V_M Then α is α_M;

[0167] R6: If E is E_M and P is P_G and C is C_U and V is V_U Then α is α_M;

[0168] R7: If E is E_H and P is P_G and C is C_S and V is V_U Then α is α_M;

[0169] R8: If E is E_H and P is P_N and C is C_M and V is V_M Then α is α_L;

[0170] R9: If E is E_M and P is P_N and C is C_U and V is V_M Then α is α_L;

[0171] R10: If E is E_H and P is P_B and C is C_U and V is V_M Then α is α_L;

[0172] R11: If E is E_H and P is P_B and C is C_U and V is V_U Then α is α_L;

[0173] R12: If E is E_L and P is P_B and C is C_U and V is V_U Then α is α_L;

[0174] R13: If E is E_H and P is P_G and C is C_M and V is V_M Then α is α_M;

[0175] R14: If E is E_M and P is P_B and C is C_M and V is V_U Then α is α_L;

[0176] R15: If E is E_L and P is P_G and C is C_U and V is V_M Then α is α_M;

[0177] Among them, the preset fuzzy rule R1 means that if the environmental fuzziness level is low (E_L), the perception quality fuzziness level is high (P_G), the communication quality fuzziness level is stable (C_S), and the motion stability fuzziness level is stable (V_S), then the credibility fuzziness level is high (α_H). In the same way, the meanings represented by the preset fuzzy rules R2-R15 can be determined, which will not be elaborated here.

[0178] Fuzzy reasoning is implemented using the Mamdani reasoning method, and the pseudocode for fuzzy reasoning is shown in Table 3.

[0179] Table 3

[0180]

[0181] Among them, the membership curves corresponding to the environmental severity score, perception quality score, communication quality score, motion stability score, and credibility discount factor are as follows: Figure 2 As shown.

[0182] Specifically, taking R1 as an example, the input is a condition connected by the and operator, which satisfies formula (4).

[0183] Formula (4): ;

[0184] in, It is the membership degree of E_L (the fuzzy level of the harsh environment is low). It is the membership degree of P_G (the level of perceptual quality fuzziness is high). It is the membership degree of C_S (communication quality fuzzy level is stable). It is the membership degree of V_S (motion-stable fuzzy level is stable). , represents the membership degree when E_L, P_G, C_S, and V_S are simultaneously satisfied, i.e., the activation strength of R1.

[0185] The fuzzy rules are represented by the cross product of the conditions and the conclusions. The minimum value of the membership function of the conditions and the conclusions is the membership function of the cross product. R1 can be expressed as formula (5).

[0186] Formula (5): ;

[0187] in, The inputs to the preset fuzzy rule R1 are: E_L, P_G, C_S, V_S, and the output is: cross product, It is a score for harsh environment, It is the perceived quality score, It is the communication quality score, It is the motion stability fraction; It is the membership degree of the fuzzy level of environmental severity E_L, determined based on the environmental severity score. It is the membership degree of the environmental severity fuzzy level P_G determined based on the perceived quality score. It is the membership degree of the environmental severity fuzzy level C_S determined based on the perceived quality score. It is the credibility fuzziness level corresponding to the credibility discount factor. membership degree This is the activation strength of R1, which takes values ​​of E_L, P_G, C_S, V_S, and The minimum membership degree.

[0188] In the same way, the expressions of the preset fuzzy rules R1-R15 can be determined. The preset fuzzy rules R1-R15 are connected by the OR operator. Therefore, the preset fuzzy rules R1-R15 can be expressed as formula (6):

[0189] Formula (6): ;

[0190] R represents all preset fuzzy rules, connected by a union of R1-R15. It is the activation intensity of R, which is the maximum value of the activation intensity of the preset fuzzy rule.

[0191] In this example, among R1-R15, there are M preset fuzzy rules with activation strength greater than 0, i.e., there are M activation fuzzy rules. The preset center value of the confidence fuzziness level in the j-th activation fuzzy rule is... The fuzziness is resolved using the central mean method, as shown in formula (7):

[0192] Formula (7): ;

[0193] in, It is a credibility discount factor. It is the center value of the credibility fuzziness level in the j-th activation fuzzy rule. It is the activation strength of the j-th activation fuzzy rule; The minimum value for setting the credibility discount factor is 0.2 to prevent the denominator from approaching 0 and causing distance explosion;

[0194] Substituting the membership functions of E, P, C and V into formula (7), we can obtain the expression shown in formula (8).

[0195] Formula (8):

[0196] ;

[0197] in, , , , , which are the inputs in the j-th activation fuzzy rule respectively. Corresponding membership degree; Input The corresponding membership degree satisfies the following constraints:

[0198] ;

[0199] Where E represents a vague level of environmental severity. It is the half-width of the fuzzy subset corresponding to E in the j-th preset fuzzy rule; In the j-th preset fuzzy rule, E is the center value corresponding to E, and e is the input value of the environmental severity fuzziness level E. , indicating that for the j-th preset fuzzy rule, when the input value e is... The absolute value of the difference is less than When the degree of membership of input value e is greater than 0, the degree of membership of input value e is 0; otherwise, the degree of membership of input value e is 0.

[0200] In the above embodiments, fuzzy reasoning is used to convert the fuzzification results of environmental adverse conditions, perception quality, communication quality, and motion stability into a credibility discount factor that quantifies the reliability of the system. This credibility discount factor can characterize the comprehensive confidence level of the vehicle under multiple dimensions of environment, perception, communication, and motion stability. Subsequently, the credibility discount factor is used to adjust the initial safe distance, so that environmental adverse conditions, perception quality, communication quality, and motion stability are incorporated into the safe distance calculation, thereby improving the adaptability of safe distance control to dynamic driving conditions.

[0201] In one possible implementation, S30, determining the initial safe distance based on target operation data corresponding to the vehicle platooning operation state includes: S301, when the vehicle platooning operation state is a following vehicle state, determining the initial safe distance based on the speed of the preceding vehicle, the speed of the following vehicle, the load factor, and the road gradient; S302, when the vehicle platooning operation state is a lead vehicle state, determining the initial safe distance based on the speed of the following vehicle, the load factor, and the road gradient.

[0202] It should be noted that in this embodiment, S301 is the step of determining the initial safe distance when the vehicle is in the following state, and S302 is the step of determining the initial safe distance when the vehicle is in the leading state. S301 and S302 are parallel solutions.

[0203] Among them, the speed of the vehicle in front represents the speed of the vehicle in front, the speed of the vehicle itself is the speed of the vehicle, the load factor is used to represent the degree of change of the actual load of the vehicle relative to the rated load, and the road gradient is the gradient of the road where the vehicle is currently located.

[0204] It should be noted that, in the case of following a vehicle, the speed of the vehicle in front and the speed of the vehicle itself can reflect the relative motion relationship between the two vehicles, and reflect the impact of the vehicle in front decelerating, moving at a constant speed or accelerating on the vehicle following. By introducing the load factor and road gradient, the actual braking capacity and inertial response of the vehicles can be reflected on the impact of following on the vehicle following.

[0205] For the lead vehicle's operating status, the initial safe distance is determined based on the vehicle's speed, combined with the load factor and road gradient. By introducing the load factor and road gradient, the influence of the vehicle's actual braking capacity and inertial response on the lead vehicle's braking and acceleration is reflected.

[0206] In the above embodiments, the initial safe distance is determined by using corresponding target operation data under different platooning operation states, which improves the adaptability of the initial safe distance to the platooning operation state. In addition, the introduction of load factor and road slope makes the initial safe distance take into account the vehicle load and road slope, thus improving the accuracy of the initial safe distance.

[0207] In one possible implementation, in step S301, an initial safe distance is determined based on the speed of the preceding vehicle, the speed of the following vehicle, the load factor, and the road gradient, including: S3011, adjusting the deceleration threshold according to the load factor and the road gradient to obtain an equivalent deceleration; S3012, determining a first braking distance based on the speed of the preceding vehicle, the speed of the following vehicle, and the equivalent deceleration; S3013, determining a reaction distance based on the speed of the following vehicle and a safe travel distance; and S3014, determining an initial safe distance based on the first braking distance, the reaction distance, and a preset safety margin.

[0208] The deceleration threshold limits the maximum permissible deceleration (maximum braking deceleration) for the vehicle. The equivalent deceleration is obtained by adjusting the deceleration threshold using a load factor and road gradient. The first braking distance is the distance required for the vehicle to brake when following another vehicle. The safe braking distance is the time margin between recognizing the preceding vehicle's condition and completing the braking response. The reaction distance is the distance required between recognizing the preceding vehicle's condition and completing the braking response. The deceleration threshold, safe braking distance, and preset safety margin are all pre-set according to requirements.

[0209] Specifically, when the vehicles are in a convoy state where the following vehicle is in a following state, the slope correction term is determined based on the road gradient. Based on the slope correction term, load coefficient, and deceleration threshold, the equivalent deceleration is calculated. The relative speed is calculated based on the speed of the preceding vehicle and the speed of the following vehicle. The first braking distance is determined based on the relative speed and the equivalent deceleration. The product of the following vehicle speed and the safe time distance is used as the reaction distance. The sum of the first braking distance, the reaction distance, and the preset safety margin is used as the initial safety distance.

[0210] For example, in the case of a following vehicle, the initial safe distance is calculated according to formula (9).

[0211] Formula (9): ;

[0212] in, It is the initial safe distance when following another vehicle. It is the first braking distance. It is the reaction distance. It is a preset safety margin; It's the speed of the car in front. It is the vehicle's speed. It is the load factor. It is the road slope (+ indicates uphill, - indicates downhill). It is the deceleration threshold. It is a safe time interval.

[0213] For example, the vehicle is an unmanned mining truck performing transportation tasks in an open-pit mine. According to relevant specifications, the deceleration threshold can be set to 4 m / s², the safe distance is 2 seconds, and the preset safety margin is 5 m. Both the speed of the preceding vehicle and the speed of the driving vehicle are less than the maximum speed limit, and the road gradient is less than the maximum gradient. In the case of an open-pit mine, the maximum speed limit for the unmanned mining truck is 11.11 m / s, and the maximum gradient is 4.57°. This applies when the unmanned mining truck is heavily loaded. Take 0.7.

[0214] In the above embodiments, when the vehicle is in a following state, the deceleration requirement is determined by the relative relationship between the speed of the preceding vehicle and the speed of the vehicle itself. The vehicle inertia is corrected by the load coefficient, and the braking increment or decrement caused by gravity component is compensated according to the road slope to obtain a first braking distance that conforms to the current driving scenario. Then, the initial safety distance is determined based on the first braking distance, the reaction distance, and the preset safety margin. This reduces the distance estimation deviation caused by heavy load, slope changes, or response lag, and improves the accuracy of the initial safety distance.

[0215] In one possible implementation, S302, when the vehicle platooning is in the lead vehicle state, an initial safe distance is determined based on the vehicle speed, load factor, and road gradient, including: S302A1, when the vehicle platooning is in the lead vehicle obstruction state, adjusting the deceleration threshold according to the load factor and road gradient to obtain an equivalent deceleration; S302A2, determining a second braking distance based on the vehicle transmission and equivalent deceleration; S302A3, determining a reaction distance based on the vehicle speed and safe travel distance; S302A4, determining the initial safe distance based on the second braking distance and reaction distance.

[0216] Among them, the first vehicle being blocked refers to the vehicle being the lead vehicle in the formation, and the vehicle needs to brake due to the situation ahead (such as the detection of an obstacle or an intersection ahead); the second braking distance is the distance required for the vehicle to brake when the lead vehicle is blocked.

[0217] Specifically, when the vehicle platoon is in a state where the lead vehicle is obstructed, a slope correction term is determined based on the road gradient. Based on the slope correction term, load coefficient, and deceleration threshold, the equivalent deceleration is calculated. The second braking distance is calculated based on the vehicle speed and the equivalent deceleration. The product of the vehicle speed and the safe time distance is used as the reaction distance. The sum of the second braking distance, the reaction distance, and the preset safety margin is used as the initial safety distance.

[0218] For example, when the lead vehicle is blocked, the initial safe distance is calculated according to formula (10).

[0219] Formula (10): ;

[0220] in, This is the initial safe distance when the lead vehicle is obstructed. It is the reaction distance. It is the second braking distance; It is a preset safety margin. It is the vehicle's speed. It is the load factor. It is the road slope (+ indicates uphill, - indicates downhill). It is the deceleration threshold. It is a safe time interval.

[0221] In the above embodiment, when the vehicle is in a state where the lead vehicle is obstructed, the deceleration threshold is adjusted by the load coefficient and the road slope to obtain an equivalent deceleration that conforms to the current driving scenario. The second braking distance is determined based on the equivalent deceleration and the vehicle speed. The initial safety distance is determined based on the second braking distance, the reaction distance, and the preset safety margin. This reduces the distance estimation deviation caused by heavy load, slope changes, or response lag, and improves the accuracy of the initial safety distance.

[0222] In one possible implementation, S302, when the vehicle platooning is in the lead vehicle mode, an initial safety distance is determined based on the vehicle speed, load factor, and road gradient, including: S302B1, when the vehicle platooning is in the lead vehicle mode, adjusting the acceleration threshold according to the load factor and road gradient to obtain an equivalent acceleration; S302B2, determining the acceleration distance based on the equivalent acceleration, vehicle speed, and acceleration time; S302B3, determining the reaction distance based on the vehicle speed and safety time distance; S302B4, determining the initial safety distance based on the acceleration distance, reaction distance, and preset safety margin.

[0223] Among these, "lead vehicle passage status" refers to a vehicle that is the lead vehicle in a convoy and is in a normal driving state that does not require braking; "acceleration threshold" limits the maximum allowable acceleration of the vehicle; "acceleration time" is the preset duration of continuous acceleration; "acceleration distance" is the distance the vehicle travels during the acceleration time from the start of acceleration; and "equivalent acceleration" is obtained by adjusting the acceleration threshold using a load factor and road gradient. Both the acceleration threshold and acceleration time can be preset according to actual needs.

[0224] Specifically, when the vehicle platoon is in a lead vehicle passage state, a slope correction term is determined based on the road gradient. Based on the slope correction term, load coefficient, and acceleration threshold, the equivalent acceleration is calculated. The acceleration distance is calculated using the equivalent acceleration, the vehicle speed, and the acceleration time. The product between the vehicle speed and the safe distance is used as the reaction distance. The sum of the acceleration distance, the reaction distance, and the preset safety margin is used as the initial safety distance.

[0225] For example, when the lead vehicle is in the passable state, the initial safe distance is calculated according to formula (11).

[0226] Formula (11): ;

[0227] in, This is the initial safe distance when the lead vehicle is passing. It is the reaction distance. It is the acceleration distance. It is a preset safety margin. It is the vehicle's speed. It is a safe time interval. It accelerates time. It is the acceleration threshold. It is the load factor. It is the road slope (+ indicates uphill, - indicates downhill). It is the acceleration threshold. It is a safe time interval.

[0228] In the above embodiments, when the vehicle is in the state of the lead vehicle passing, the acceleration threshold is adjusted by the load coefficient and the road slope to obtain the equivalent acceleration that conforms to the current driving scenario. The acceleration distance is determined based on the equivalent acceleration, the vehicle speed and the acceleration time. The initial safety distance is determined based on the acceleration distance, the reaction distance and the preset safety margin. This reduces the distance estimation deviation caused by heavy load, slope change or response lag and improves the accuracy of the initial safety distance.

[0229] In one possible implementation, S40, adjusting the initial safety distance based on a credibility discount factor to obtain a target safety distance includes: S401, adjusting the initial safety distance based on a credibility discount factor to obtain a candidate safety distance; S402A, when the vehicle platooning is in a following-vehicle mode, using the candidate safety distance as the target safety distance; S402B, when the vehicle platooning is in a lead-vehicle mode, using the larger of the reaction distance and the candidate safety distance as the target safety distance; the reaction distance is determined based on the vehicle's speed and safe headway.

[0230] Among them, the candidate safety distance is an intermediate result of the confidence discount factor correcting the initial safety distance.

[0231] Specifically, the credibility discount factor is mapped to an adjustment coefficient, and the initial safety distance is linearly corrected according to the adjustment coefficient to obtain the candidate safety distance. When the credibility discount factor is low, the candidate safety distance is amplified relative to the initial safety distance to compensate for the risks caused by perception error, communication delay, or braking capability fluctuation. When the credibility discount factor is high, the candidate safety distance is less than or equal to the initial safety distance, or the candidate safety distance is equal to the initial safety distance.

[0232] When the vehicles are in a convoy and the following vehicle is in a trailing state, the candidate safe distance is used as the target safe distance.

[0233] For example, in the case of a following vehicle, the target safe distance is determined according to formula (12).

[0234] Formula (12): ;

[0235] in, It is the target safe distance when following another vehicle. It is the initial safe distance when following another vehicle. It is a credibility discount factor. It is an adjustment factor.

[0236] When the vehicle platoon is in a state where the lead vehicle is blocked or the lead vehicle is passing, the greater of the reaction distance and the candidate safe distance shall be taken as the target safe distance.

[0237] For example, when the lead vehicle is blocked, the target safe distance is determined according to formula (13).

[0238] Formula (13):

[0239] ;

[0240] in, This refers to the target safe distance when the lead vehicle is obstructed. This is the initial safe distance when the lead vehicle is obstructed. It is a credibility discount factor. It is an adjustment factor. It is the reaction distance.

[0241] For example, when the lead vehicle is in the passage state, the target safe distance is determined according to formula (14).

[0242] Formula (14):

[0243] ;

[0244] in, This is the target safe distance when the lead vehicle is passing. This is the initial safe distance when the lead vehicle is passing. It is a credibility discount factor. It is an adjustment factor. It is the reaction distance.

[0245] In the above embodiments, the initial safe distance is intermediately corrected by a confidence discount factor, and the candidate safe distance is then constrained a second time by the formation operation state. In the following vehicle state, the candidate safe distance is used as the target safe distance to satisfy the following continuity. In the lead vehicle state, the reaction distance constraint is introduced to determine the target safe distance, which can reserve more response space for the vehicle when the risk ahead occurs. In this way, the target safe distance can be dynamically adjusted when there are fluctuations in the environment, perception, communication and motion stability, thus improving the adaptability of safe distance control to dynamic driving conditions.

[0246] In a specific example, the vehicle is an unmanned mining truck that performs transportation operations in an open-pit mine. The unmanned mining truck uses a safety distance system to dynamically determine the safety distance.

[0247] refer to Figure 4 The unmanned mining truck safety distance system includes an input layer, a decision layer, a safety constraint layer, and an output layer.

[0248] Input layer: Input environmental data, perception data, communication data, and motion stabilization data;

[0249] Decision layer: Based on environmental data, perception data, communication data, and motion stability data, fuzzification processing is performed to obtain fuzzification results for environmental degradation, perception quality, communication quality, and motion stability. Fuzzy reasoning is then used to process these fuzzification results and output a credibility discount factor.

[0250] Safety constraint layer: includes safety distance models corresponding to the following vehicle's running state, the leading vehicle's obstructed state, and the leading vehicle's passing state; the corresponding safety distance model is called according to the current platooning operation state of the unmanned mining trucks, the initial safety distance is calculated through the safety distance model, and the initial safety distance is adjusted through the credibility discount factor to obtain the target safety distance.

[0251] Among them, the safety distance models all incorporate load factor and road slope; the safety distance models for following vehicles and the obstructed leading vehicle incorporate load factor and road slope to adjust deceleration threshold, and the safety distance model for the leading vehicle passing vehicle incorporates load factor and road slope to adjust acceleration threshold, thus improving the accuracy of the initial safety distance.

[0252] Output layer: Outputs the target safe distance.

[0253] The dynamic safety distance determination method provided in this application embodiment fuzzifies multi-dimensional vehicle data to obtain fuzzification results for environmental adverseness, perception quality, communication quality, and motion stability. Then, through fuzzy inference, these results are converted into a credibility discount factor that quantifies system reliability. This credibility discount factor characterizes the comprehensive confidence level of the vehicle across multiple dimensions of environment, perception, communication, and motion stability. The initial safety distance is determined in real-time using target operation data corresponding to the vehicle's platooning operation state. The initial safety distance is not a fixed threshold, which improves the adaptability of the initial safety distance to vehicle dynamics and actual driving scenarios. The initial safety distance is adjusted by a credibility discount factor, which incorporates adverse environmental conditions, perception quality, communication quality and motion stability into the safety distance calculation. This means that the safety distance is no longer determined solely by fixed dynamic parameters, but is adjusted in real time based on the system credibility under the current driving conditions. This improves the problem of insufficient adaptability of the safety distance determined by the fixed parameter model to complex driving conditions, and realizes the dynamic adjustment of the safety distance according to the driving conditions, thereby improving the dynamic adaptability of the safety distance control to complex driving conditions.

[0254] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0255] Figure 5 A schematic diagram of the structure of the dynamic determination device for safe distance provided in this application is shown below. Figure 5 As shown, the dynamic distance determination device provided in this embodiment includes:

[0256] The fuzzing processing module 501 is used to perform fuzzing processing on multi-dimensional vehicle data to obtain multi-dimensional fuzzing results, including: environmental degradation fuzzing results, perception quality fuzzing results, communication quality fuzzing results, and motion stability fuzzing results.

[0257] The fuzzy reasoning module 502 is used to perform fuzzy reasoning based on the fuzzification results of environmental severity, perception quality, communication quality, and motion stability to obtain a credibility discount factor.

[0258] The distance determination module 503 is used to determine the initial safe distance based on the target running data corresponding to the vehicle formation running status;

[0259] The distance adjustment module 504 is used to adjust the initial safety distance according to the confidence discount factor to obtain the target safety distance.

[0260] In one possible embodiment, the distance adjustment module is used to adjust the initial safe distance according to a confidence discount factor to obtain a target safe distance, including: adjusting the initial safe distance according to the confidence discount factor to obtain a candidate safe distance; when the vehicle platooning state is a following vehicle state, using the candidate safe distance as the target safe distance; when the vehicle platooning state is a lead vehicle state, using the larger of the reaction distance and the candidate safe distance as the target safe distance; the reaction distance is determined based on the vehicle's speed and safe headway.

[0261] In one possible embodiment, the distance adjustment module is used to determine the target confidence interval to which the confidence discount factor belongs; determine a first compensation coefficient based on the target confidence interval, wherein if the target confidence interval is a low confidence interval, the first compensation coefficient is greater than 1, if the target confidence interval is a low confidence interval, the first compensation coefficient is equal to 1, and if the target confidence interval is a high confidence interval, the second compensation coefficient is less than 1; and use the product of the first compensation coefficient and the initial safety distance as the target safety distance.

[0262] In one possible embodiment, the fuzzification processing module is used to determine an environmental severity score based on the vehicle's environmental data; determine a perception quality score based on the vehicle's perception data; determine a communication quality score based on the vehicle's communication data; determine a motion stability score based on the vehicle's motion stability data; and perform fuzzification processing on the environmental severity score, perception quality score, communication quality score, and motion stability score respectively to obtain membership degrees under multiple environmental severity fuzziness levels, multiple perception quality fuzziness levels, multiple communication quality fuzziness levels, and multiple motion stability fuzziness levels.

[0263] In one possible embodiment, the environmental degradation fuzzification result includes membership degrees corresponding to multiple environmental degradation fuzzification levels, the perception quality fuzzification result includes membership degrees corresponding to multiple perception quality fuzzification levels, the communication quality fuzzification result includes membership degrees corresponding to multiple communication quality fuzzification levels, and the motion stability fuzzification result includes membership degrees corresponding to multiple motion stability fuzzification levels.

[0264] The fuzzy inference module is used to determine the activation strength of multiple preset fuzzy rules based on the membership degrees corresponding to multiple environmental severity fuzzy levels, multiple perception quality fuzzy levels, multiple communication quality fuzzy levels, and multiple motion stability fuzzy levels; determine the active fuzzy rule among the multiple preset fuzzy rules based on the activation strength of the active fuzzy rule; and perform defuzzification processing based on the activation strength of the active fuzzy rule and the preset center value corresponding to the credibility fuzzy level of the active fuzzy rule to obtain the credibility discount factor.

[0265] In one possible embodiment, the distance determination module is used to determine an initial safe distance based on the speed of the preceding vehicle, the speed of the following vehicle, the load factor, and the road gradient when the vehicles are in a platooning state of following other vehicles; and to determine an initial safe distance based on the speed of the following vehicle, the load factor, and the road gradient when the vehicles are in a platooning state of leading vehicle.

[0266] In one possible embodiment, the distance determination module is used to adjust the deceleration threshold according to the load coefficient and road slope to obtain an equivalent deceleration; determine a first braking distance according to the speed of the preceding vehicle, the speed of the following vehicle, and the equivalent deceleration; determine a reaction distance according to the speed of the following vehicle and a safe travel distance; and determine an initial safe distance according to the first braking distance, the reaction distance, and a preset safety margin.

[0267] In one possible embodiment, the distance determination module is used to adjust the deceleration threshold according to the load coefficient and road gradient to obtain an equivalent deceleration when the vehicle platooning state is such that the lead vehicle is blocked; determine a second braking distance based on the vehicle speed and the equivalent deceleration; determine a reaction distance based on the vehicle speed and the safe distance; and determine an initial safety distance based on the second braking distance, the reaction distance and the preset safety margin.

[0268] In one possible embodiment, the distance determination module is used to adjust the acceleration threshold according to the load coefficient and road gradient to obtain the equivalent acceleration when the vehicle platooning operation is in the case of the lead vehicle passing state; determine the acceleration distance based on the equivalent acceleration, the vehicle speed and acceleration time; determine the reaction distance based on the vehicle speed and safe distance; and determine the initial safety distance based on the acceleration distance, reaction distance and preset safety margin.

[0269] The dynamic safety distance determination device provided in this embodiment can execute the dynamic safety distance determination method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0270] Figure 6 This is a structural diagram of the vehicle provided in this application. Figure 6 As shown, the vehicle 60 provided in this embodiment includes at least one processor 601 and a memory 602. Optionally, the device 60 further includes a communication component 603. The processor 601, memory 602, and communication component 603 are connected via a bus.

[0271] In a specific implementation, at least one processor 601 executes computer execution instructions stored in memory 602, causing at least one processor 601 to perform the above-described method.

[0272] The specific implementation process of processor 601 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0273] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0274] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0275] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0276] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0277] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0278] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0279] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0280] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0281] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0282] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0283] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0284] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0285] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for dynamically determining a safe distance, characterized in that, include: The multi-dimensional data of the vehicle is fuzzed to obtain multi-dimensional fuzzification results, which include environmental degradation fuzzification results, perception quality fuzzification results, communication quality fuzzification results, and motion stability fuzzification results. Based on the environmental degradation fuzzification result, the perception quality fuzzification result, the communication quality fuzzification result, and the motion stability fuzzification result, fuzzy inference is performed to obtain the credibility discount factor; The initial safe distance is determined based on the target operation data corresponding to the vehicle formation operation status; The initial safety distance is adjusted based on the credibility discount factor to obtain the target safety distance.

2. The method according to claim 1, characterized in that, The step of adjusting the initial safety distance based on the credibility discount factor to obtain the target safety distance includes: The initial safety distance is adjusted according to the confidence discount factor to obtain a candidate safety distance; When the vehicles are in a convoy and the following vehicle is in a tail-to-tail state, the candidate safe distance is taken as the target safe distance; When the vehicles are in platooning with the lead vehicle in the lead vehicle position, the larger of the reaction distance and the candidate safe distance is taken as the target safe distance; the reaction distance is determined based on the vehicle's speed and safe travel distance.

3. The method according to claim 1, characterized in that, The step of adjusting the initial safety distance based on the credibility discount factor to obtain the target safety distance includes: Determine the target credibility interval to which the credibility discount factor belongs; A first compensation coefficient is determined based on the target confidence interval, wherein if the target confidence interval is a low confidence interval, the first compensation coefficient is greater than 1; if the target confidence interval is a low confidence interval, the first compensation coefficient is equal to 1; and if the target confidence interval is a high confidence interval, the first compensation coefficient is less than 1. The product of the first compensation coefficient and the initial safety distance is taken as the target safety distance.

4. The method according to any one of claims 1 to 3, characterized in that, The multi-dimensional data based on the vehicle is fuzzified to obtain a multi-dimensional fuzzification result, including: Determine the environmental severity score based on the vehicle's environmental data; The perceived quality score is determined based on the vehicle's perception data. The communication quality score is determined based on the vehicle's communication data; The motion stability score is determined based on the vehicle's motion stability data. The environmental severity score, perception quality score, communication quality score, and motion stability score are fuzzified to obtain membership degrees under multiple environmental severity fuzzy levels, multiple perception quality fuzzy levels, multiple communication quality fuzzy levels, and multiple motion stability fuzzy levels.

5. The method according to any one of claims 1 to 3, characterized in that, The environmental severity fuzzification result includes membership degrees corresponding to multiple environmental severity fuzziness levels; the perception quality fuzzification result includes membership degrees corresponding to multiple perception quality fuzziness levels; the communication quality fuzzification result includes membership degrees corresponding to multiple communication quality fuzziness levels; and the motion stability fuzzification result includes membership degrees corresponding to multiple motion stability fuzziness levels. The process of performing fuzzy inference based on the environmental degradation fuzzification result, the perception quality fuzzification result, the communication quality fuzzification result, and the motion stability fuzzification result to obtain a credibility discount factor includes: Based on the membership degrees corresponding to the multiple environmental severity fuzzy levels, the multiple perception quality fuzzy levels, the multiple communication quality fuzzy levels, and the multiple motion stability fuzzy levels, the activation intensity of multiple preset fuzzy rules is determined. Based on the activation intensity, an activation fuzzy rule is determined from the plurality of preset fuzzy rules; Defuzzification is performed based on the activation intensity of the activation fuzzy rule and the preset center value corresponding to the credibility fuzziness level of the activation fuzzy rule to obtain the credibility discount factor.

6. The method according to any one of claims 1 to 3, characterized in that, The determination of the initial safe distance based on the target operation data corresponding to the vehicle platooning status includes: When the vehicles are in a convoy and the following vehicle is in a following state, the initial safe distance is determined based on the speed of the preceding vehicle, the speed of the following vehicle, the load factor, and the road gradient. When the vehicles are in platooning with the lead vehicle in the lead vehicle position, the initial safe distance is determined based on the vehicle speed, load factor, and road gradient.

7. The method according to claim 6, characterized in that, The determination of the initial safe distance based on the speed of the vehicle in front, the speed of the vehicle behind, the load factor, and the road gradient includes: The deceleration threshold is adjusted based on the load factor and road slope to obtain the equivalent deceleration; The first braking distance is determined based on the speed of the vehicle in front, the speed of the vehicle itself, and the equivalent deceleration. The reaction distance is determined based on the vehicle speed and safe travel distance. The initial safety distance is determined based on the first braking distance, the reaction distance, and the preset safety margin.

8. The method according to claim 6, characterized in that, When the vehicle platoon is in a lead vehicle mode, the initial safe distance is determined based on the vehicle's speed, load factor, and road gradient, including: When the lead vehicle is obstructed during convoy operation, the deceleration threshold is adjusted based on the load factor and road gradient to obtain the equivalent deceleration. The second braking distance is determined based on the vehicle speed and the equivalent deceleration. The reaction distance is determined based on the vehicle speed and safe travel distance. The initial safety distance is determined based on the second braking distance, the reaction distance, and the preset safety margin.

9. The method according to claim 6, characterized in that, When the vehicle platoon is in a lead vehicle mode, the initial safe distance is determined based on the vehicle's speed, load factor, and road gradient, including: When the vehicle platoon is in the lead vehicle passing state, the acceleration threshold is adjusted according to the load factor and road gradient to obtain the equivalent acceleration. The acceleration distance is determined based on the equivalent acceleration, vehicle speed, and acceleration time. The reaction distance is determined based on the vehicle's speed and safe travel distance. The initial safety distance is determined based on the acceleration distance, the reaction distance, and the preset safety margin.

10. A device for dynamically determining a safe distance, characterized in that, The device includes: The fuzzing processing module is used to perform fuzzing processing based on multi-dimensional vehicle data to obtain multi-dimensional fuzzing results, including environmental degradation fuzzing results, perception quality fuzzing results, communication quality fuzzing results, and motion stability fuzzing results. The fuzzy inference module is used to perform fuzzy inference based on the environmental degradation fuzzification result, the perception quality fuzzification result, the communication quality fuzzification result, and the motion stability fuzzification result to obtain a confidence discount factor. The distance determination module is used to determine the initial safe distance based on the target running data corresponding to the vehicle formation running status; The distance adjustment module is used to adjust the initial safety distance according to the confidence discount factor to obtain the target safety distance.

11. A vehicle, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 9.

13. A computer program product, characterized in that, Includes computer execution instructions, which, when executed by a processor, implement the method as described in any one of claims 1 to 9.

Images