Vehicle rear-end collision prevention method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention relates to the technical field of vehicles, and discloses a vehicle rear-end collision prevention method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring multi-sensor data, extracting a key feature array based on the multi-sensor data, calculating a spatial quality score according to the key feature array and a preset weight array, and judging a rear-end collision prevention level according to the spatial quality score and a score threshold array; and after a vehicle braking control strategy and a segmented avoidance strategy are determined at least according to the rear-end collision prevention level, corresponding strategy execution is completed. According to the embodiment of the invention, the method can at least dynamically evaluate the quality of the space where the vehicle is located, and facilitates the effective avoidance of the vehicle in a space-limited area. In addition, through the segmented avoidance strategy, the risk of secondary collision can be reduced while the probability of primary collision is reduced, and the personal and property safety of the user can be guaranteed.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and in particular to a method, device, electronic device, and storage medium for preventing rear-end collisions. Background Technology

[0002] Currently, vehicles may face the risk of rear-end collisions when stationary (such as at traffic light intersections, parking lots, highways, etc.). Existing technologies mainly employ the following two prevention methods:

[0003] 1. Electronic fence technology is used to actively move vehicles to avoid collisions before they occur. However, the electronic fence is a fixed range and cannot be dynamically adjusted according to the actual environment, which may result in ineffective avoidance in areas with limited space.

[0004] 2. It can achieve safe avoidance by setting a steering angle after being rear-ended, but the problem is that it only takes measures after the collision occurs and cannot avoid the first collision. Summary of the Invention

[0005] The purpose of this invention is to provide a vehicle rear-end collision prevention method, device, electronic device, and storage medium, which can at least perform dynamic assessment of the quality of the space in which the vehicle is located, facilitating effective collision avoidance by the vehicle in space-constrained areas; in addition, by using a segmented collision avoidance strategy, the probability of a primary collision can be reduced while mitigating the risk of a secondary collision, thus helping to protect the personal and property safety of users.

[0006] To address the aforementioned technical problems, in a first aspect, the present invention provides a method for preventing rear-end collisions of vehicles, applicable at least to scenarios where the vehicle is stationary, the method comprising at least:

[0007] Acquire multi-sensor data, extract key feature arrays based on the multi-sensor data, calculate space quality score based on the key feature array and preset weight array, and then determine the rear-end collision prevention level based on the space quality score and score threshold array.

[0008] After determining the vehicle braking control strategy and segmented avoidance strategy based on the aforementioned rear-end collision prevention level, the corresponding strategy execution is completed.

[0009] Optionally, the space quality score is calculated at least in the following ways:

[0010] Q=α1×(S / Smax)+α2×(1-R / Rmax)+α3×(1-F / Fmax)+α4×(1-P / Pmax);

[0011] In the above formula, Q represents the space quality score, α1 represents the first preset weight, α2 represents the second preset weight, α3 represents the third preset weight, α4 represents the fourth preset weight, S represents the area of ​​the vehicle's avoidable zone, Smax represents the maximum usable space area of ​​the vehicle, R represents the ground unevenness, Rmax represents the maximum ground unevenness allowed for vehicle rear-end collision prevention, F represents the obstacle density, Fmax represents the maximum obstacle density allowed for vehicle rear-end collision prevention, P represents the pedestrian activity intensity, Pmax represents the maximum pedestrian activity intensity allowed for vehicle rear-end collision prevention;

[0012] The first preset weight, the second preset weight, the third preset weight, and the fourth preset weight together constitute the preset weight array.

[0013] Optionally, the scoring threshold array includes at least a first scoring threshold and a second scoring threshold, wherein the first scoring threshold is less than the second scoring threshold;

[0014] The rear-end collision prevention level is determined by at least the following methods:

[0015] When the space quality score is lower than the first score threshold, the rear-end collision avoidance level is determined to be an unavoidable level.

[0016] When the space quality score is not lower than the first score threshold but lower than the second score threshold, the rear-end collision avoidance level is determined to be the cautious and slightly avoidable level.

[0017] When the space quality score is not lower than the second score threshold, the rear-end collision avoidance level is determined to be the safety minor avoidance level.

[0018] Optionally, when the rear-end collision avoidance level is an unavoidable level, the vehicle braking control strategy is at least to adjust the braking intensity of the vehicle's AEB system to the first AEB intensity and the braking intensity of the vehicle's RCTB system to the first RCTB intensity.

[0019] Optionally, when the rear-end collision level is the cautious evasive level, the vehicle braking control strategy is at least to adjust the braking intensity of the vehicle's AEB system to the second AEB intensity and the braking intensity of the vehicle's RCTB system to the second RCTB intensity.

[0020] Wherein, the second AEB intensity is less than the first AEB intensity, and the second RCTB intensity is less than the first RCTB intensity.

[0021] Optionally, when the rear-end collision level is a safety micro-avoidance level, the vehicle braking control strategy is at least to maintain the braking intensity of the vehicle's AEB system and the braking intensity of the vehicle's RCTB system at the current moment's intensity.

[0022] Optionally, the segmented avoidance strategy includes at least a warning preparation phase, a micro-avoidance phase, and a collision protection phase.

[0023] Based on the same concept, in a second aspect, the present invention also provides a vehicle rear-end collision prevention device for performing the vehicle rear-end collision prevention method described in any one of the first aspects;

[0024] The vehicle rear-end collision prevention device includes at least:

[0025] The level determination module is used to acquire multi-sensor data, extract key feature arrays based on the multi-sensor data, calculate space quality scores based on the key feature arrays and preset weight arrays, and then determine the rear-end collision prevention level based on the space quality scores and score threshold arrays.

[0026] The strategy execution module is used to determine the vehicle braking control strategy and the segmented avoidance strategy based at least on the rear-end collision avoidance level, and then complete the corresponding strategy execution.

[0027] Based on the same concept, in a third aspect, the present invention also provides an electronic device, including a memory and a processor, the memory storing a computer program executable on the processor, the processor executing the program to implement the steps of the vehicle rear-end collision prevention method according to any one of the first aspects.

[0028] Based on the same concept, in a fourth aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the vehicle rear-end collision prevention method according to any one of the first aspects.

[0029] The technical solution provided by this invention first acquires multi-sensor data, extracts key feature arrays based on the multi-sensor data, and calculates a spatial quality score based on the key feature array and a preset weight array. Then, it determines the rear-end collision prevention level based on the spatial quality score and a score threshold array. Afterward, it determines a vehicle braking control strategy and a segmented avoidance strategy based at least on the rear-end collision prevention level, and executes the corresponding strategies. Therefore, the vehicle rear-end collision prevention method of this invention can at least perform dynamic assessment of the spatial quality of the vehicle, which is beneficial for effective avoidance in space-constrained areas. Furthermore, the segmented avoidance strategy can reduce the probability of a primary collision while mitigating the risk of a secondary collision, thus protecting the personal and property safety of users. Attached Figure Description

[0030] Figure 1 This is a flowchart of a vehicle rear-end collision prevention method provided in an embodiment of the present invention;

[0031] Figure 2 This is a schematic diagram of the structure of a vehicle rear-end collision prevention device provided in an embodiment of the present invention;

[0032] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0034] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the application. The singular forms “a,” “said,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0035] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0036] It should be understood that although the terms first, second, third, etc., may be used in the embodiments of this application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, first may also be referred to as second without departing from the scope of the embodiments of this application, and similarly, second may also be referred to as first.

[0037] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0038] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0039] It should be noted that any symbols and / or numbers present in the specification that are not marked in the accompanying drawings are not reference numerals.

[0040] Figure 1 This is a flowchart of a vehicle rear-end collision prevention method provided by an embodiment of the present invention. This embodiment is at least applicable to rear-end collision prevention scenarios when the vehicle is stationary. This vehicle rear-end collision prevention method can be, but is not limited to, executed by the vehicle rear-end collision prevention device in this embodiment of the present invention as the execution subject, which can be implemented in software and / or hardware. Figure 1 As shown, this method for preventing rear-end collisions includes at least the following steps:

[0041] S1. Acquire multi-sensor data, extract key feature arrays based on multi-sensor data, calculate space quality score based on key feature arrays and preset weight arrays, and then determine rear-end collision prevention level based on space quality score and score threshold array.

[0042] The multi-sensor data can be obtained by at least one sensor, such as lidar or a camera. By employing multi-sensor data fusion, this embodiment can assess the spatial quality around the vehicle in real time and construct a 360° spatial quality map, including elements such as space size, ground smoothness, obstacle type, and pedestrian activity. The multi-sensor data can be collected every 100ms. It is understood that the key feature array can consist of various key feature data, such as the area of ​​the vehicle's avoidable zone, the maximum usable space area of ​​the vehicle, ground roughness, and the maximum permissible ground roughness for rear-end collision prevention.

[0043] In one specific implementation, the space quality score may optionally be calculated in at least the following ways:

[0044] Q=α1×(S / Smax)+α2×(1-R / Rmax)+α3×(1-F / Fmax)+α4×(1-P / Pmax);

[0045] In the above formula, Q represents the spatial quality score, α1 represents the first preset weight, α2 represents the second preset weight, α3 represents the third preset weight, α4 represents the fourth preset weight, S represents the area of ​​the vehicle's avoidable zone, Smax represents the maximum usable space area of ​​the vehicle, R represents the ground unevenness, Rmax represents the maximum ground unevenness allowed for vehicle rear-end collision prevention, F represents the obstacle density, Fmax represents the maximum obstacle density allowed for vehicle rear-end collision prevention, P represents the pedestrian activity intensity, Pmax represents the maximum pedestrian activity intensity allowed for vehicle rear-end collision prevention; the first preset weight, the second preset weight, the third preset weight, and the fourth preset weight together form the preset weight array.

[0046] For example, the first preset weight can be set to 0.3, the second preset weight can be set to 0.3, the third preset weight can be set to 0.3, and the fourth preset weight can be set to 0.3; the maximum usable space area of ​​the vehicle can be determined according to the vehicle model, for example, it can be taken as 3m × 2m (i.e., 6m). 2 The maximum permissible ground unevenness for vehicle rear-end collision prevention can be determined through pre-testing, for example, 5cm; the maximum permissible obstacle density for vehicle rear-end collision prevention can be configured according to the actual adaptability of the vehicle, for example, 0.5 obstacles / m². 2 The maximum pedestrian activity intensity allowed for rear-end collision prevention can also be set based on the actual vehicle settings, for example, it can be set to 0.3 people / (m²). 2 s).

[0047] In another specific implementation, optionally, the scoring threshold array includes at least a first scoring threshold and a second scoring threshold, wherein the first scoring threshold is less than the second scoring threshold;

[0048] The rear-end collision protection level can be determined by at least the following methods:

[0049] When the space quality score is lower than the first score threshold, the rear-end collision avoidance level is determined to be the unavoidable level.

[0050] When the space quality score is not lower than the first score threshold but lower than the second score threshold, the rear-end collision prevention level is determined to be the cautious and slightly avoidable level.

[0051] When the space quality score is not lower than the second score threshold, the rear-end collision avoidance level is determined to be the safe minor avoidance level.

[0052] For example, the first scoring threshold can be set to 0.4, and the second scoring threshold can be configured to 0.7.

[0053] More specifically, the above parameters and their physical meanings will be explained further through concrete examples and cases below.

[0054] 1. S - Area of ​​the vehicle's avoidance zone

[0055] S can refer to the area of ​​the safe zone where a vehicle can make minor avoidance maneuvers when stationary.

[0056] The weight of this parameter can be 0.3, indicating that the size of the space is the most important factor affecting the avoidance strategy.

[0057] The formula uses S / Smax, and the larger the value, the higher the spatial quality.

[0058] Example:

[0059] Assuming a small car is stationary in a parking lot, with a 1.5m clear space to the right and a 1m clear space in front, then the clearance area to avoid the car is S = 1.5m × 1m = 1.5m. 2 .

[0060] Smax = 3m × 2m = 6m 2 .

[0061] S / Smax = 1.5 / 6 = 0.25.

[0062] The contribution of this parameter is: 0.3 × 0.25 = 0.075.

[0063] 2. R - Ground unevenness

[0064] R can refer to the flatness of the ground in the area where the vehicle is stationary, usually expressed as the maximum height difference (cm).

[0065] The formula uses (1-R / Rmax) to indicate that the smoother the ground, the higher the score.

[0066] This parameter has a weight of 0.25 and is a key factor affecting obstacle avoidance safety.

[0067] Example:

[0068] In the parking lot, the ground has slight undulations, with a maximum height difference of 2cm (0.02m).

[0069] R=0.02m, Rmax=0.05m.

[0070] R / Rmax = 0.02 / 0.05 = 0.4.

[0071] (1-R / Rmax)=1-0.4=0.6.

[0072] The contribution of this parameter is: 0.25 × 0.6 = 0.15.

[0073] 3. F - Obstacle Density

[0074] F can refer to the number of obstacles per unit area around the vehicle (number / m²) 2 ).

[0075] The formula uses (1-F / Fmax) to indicate that the fewer obstacles there are, the higher the score.

[0076] This parameter has a weight of 0.25, the same as the ground unevenness.

[0077] Example:

[0078] In the parking lot, there are two obstacles (such as other vehicles, trash cans, etc.) within a 5m x 5m area around the vehicle.

[0079] F = 2 / (5×5) = 0.08 pieces / m 2 .

[0080] Fmax = 0.5 pieces / m 2 .

[0081] F / Fmax = 0.08 / 0.5 = 0.16.

[0082] (1-F / Fmax)=1-0.16=0.84.

[0083] The contribution of this parameter is: 0.25 × 0.84 = 0.21.

[0084] 4. P - Pedestrian activity intensity

[0085] P can refer to the intensity of pedestrian flow per unit area (people / (m²)). 2 s)).

[0086] The formula uses (1-P / Pmax) to indicate that the less pedestrian activity, the higher the score.

[0087] This parameter has a relatively low weight of 0.2, but it is crucial for pedestrian safety.

[0088] Example:

[0089] At a traffic light intersection, the pedestrian flow rate in the pedestrian crossing area is 0.2 people / (m²). 2 s).

[0090] P=0.2, Pmax=0.3.

[0091] P / Pmax = 0.2 / 0.3 ≈ 0.67.

[0092] (1-P / Pmax)=1-0.67=0.33.

[0093] The contribution of this parameter is: 0.2 × 0.33 = 0.066.

[0094] 5. Case Study

[0095] Case 1: Traffic light intersection scene

[0096] S=2.5m 2 (Can be moved slightly to the right by 1.25m × 2m);

[0097] R=0.03m (the ground is slightly uneven);

[0098] F = 0.2 pieces / m 2 (There are a few obstacles);

[0099] P = 0.15 people / (m 2 s) (few pedestrians);

[0100] Q=0.3×(2.5 / 6)+0.25×(1-0.03 / 0.05)+0.25×(1-0.2 / 0.5)+0.2×(1-0.15 / 0. 3)=0.3×0.417+0.25×0.4+0.25×0.6+0.2×0.5=0.125+0.1+0.15+0.1=0.475;

[0101] 0.4≤0.475<0.7→“Caution is required”, meaning the rear-end collision level is “caution is required”.

[0102] Case 2: Parking Lot Scene

[0103] S=1.8m 2 (Can be moved slightly to the left by 0.9m × 2m);

[0104] R=0.01m (ground is flat);

[0105] F = 0.4 pieces / m 2 (There are many obstacles);

[0106] P = 0.25 people / (m 2 s) (Many pedestrians);

[0107] Q=0.3×(1.8 / 6)+0.25×(1-0.01 / 0.05)+0.25×(1-0.4 / 0.5)+0.2×(1-0.25 / 0. 3)=0.3×0.3+0.25×0.8+0.25×0.2+0.2×0.167=0.09+0.2+0.05+0.033=0.373;

[0108] Q=0.373<0.4→“Unavoidable”, meaning the rear-end collision level is unavoidable.

[0109] Case 3: Highway Service Area

[0110] S=4.5m 2 (Can be moved slightly to the right by 2.25m × 2m);

[0111] R=0.02m (ground is flat);

[0112] F=0.1 pieces / m 2 (Few obstacles);

[0113] P = 0.05 people / (m 2 (s) (very few pedestrians);

[0114] Q=0.3×(4.5 / 6)+0.25×(1-0.02 / 0.05)+0.25×(1-0.1 / 0.5)+0.2×(1-0.05 / 0.3 )=0.3×0.75+0.25×0.6+0.25×0.8+0.2×0.833=0.225+0.15+0.2+0.167=0.742;

[0115] Q=0.742≥0.7→“Safe to avoid”, meaning the rear-end collision level is the safe to avoid level.

[0116] S2. After determining the vehicle braking control strategy and segmented avoidance strategy based on the rear-end collision prevention level, complete the corresponding strategy execution.

[0117] Different rear-end collision avoidance levels can correspond to different vehicle braking control strategies and segmented avoidance strategies.

[0118] In another specific implementation, optionally, when the rear-end collision avoidance level is the unavoidable level, the vehicle braking control strategy is at least to adjust the braking intensity of the vehicle's AEB system to the first AEB intensity (e.g., it can be set to 100%) and the braking intensity of the vehicle's RCTB system to the first RCTB intensity (e.g., it can be set to 100%).

[0119] In another specific implementation, optionally, when the rear-end collision level is the cautious avoidance level, the vehicle braking control strategy is at least to adjust the braking intensity of the vehicle's AEB system to the second AEB intensity (e.g., it can be set to 50%) and the braking intensity of the vehicle's RCTB system to the second RCTB intensity (e.g., it can be set to 50%).

[0120] The second AEB intensity is less than the first AEB intensity, and the second RCTB intensity is less than the first RCTB intensity.

[0121] In another specific implementation, optionally, when the rear-end collision level is a safety minor avoidance level, the vehicle braking control strategy is at least to maintain the braking intensity of the vehicle's AEB system and the braking intensity of the vehicle's RCTB system at the current moment's intensity.

[0122] In another specific implementation, the segmented avoidance strategy may optionally include at least a warning preparation phase, a micro-avoidance phase, and a collision protection phase.

[0123] For example, the segmented avoidance strategy can be specifically as follows:

[0124] Phase 1: Intelligent Early Warning and Preparation (i.e., the aforementioned early warning and preparation phase, for example, 5-10 seconds before a collision occurs)

[0125] - Determine the approaching speed and risk level of vehicles behind by fusing multiple sensors;

[0126] - Generate a risk heat map to show potential collision points and avoidance space;

[0127] - The driver should be informed of the system status at least through in-vehicle displays and audio prompts.

[0128] More specific explanation:

[0129] 1) Judging the approaching speed of vehicles behind

[0130] The approaching speed of vehicles behind can be determined in real time through data fusion from multiple sensors, including millimeter-wave radar, cameras, and ultrasonic radar.

[0131] Velocity calculation formula: V_rel=(D_prev-D_current) / t;

[0132] In the above formula, V_rel: relative velocity (m / s);

[0133] D_prev: Distance (m) of the vehicle behind at the previous moment;

[0134] D_current: Distance to vehicles behind at the current time (m);

[0135] t: time interval (e.g., 100ms).

[0136] 2) Risk level assessment

[0137] A multi-dimensional risk assessment model is used to calculate the risk level R:

[0138] R=β1×(V_rel / V_max)+β2×(D / D_safe)+β3×(V_rel×D)+β4×(S / Smax);

[0139] In the above formula, β1: the first evaluation weight, which can be 0.4;

[0140] β2: Second evaluation weight, which can be 0.3;

[0141] β3: The third evaluation weight, which can be 0.2;

[0142] β4: Fourth evaluation weight, which can be 0.1;

[0143] V_rel: Relative speed of vehicles behind (m / s);

[0144] V_max: The maximum possible speed of the vehicle behind (in m / s, usually taken as 20 m / s).

[0145] D: Current vehicle distance (m);

[0146] D_safe: Safety distance (unit: m, can be calculated based on vehicle speed, calculation formula: D_safe=0.5×V_rel+2.0).

[0147] Risk level classification:

[0148] R<0.3: Low risk;

[0149] 0.3≤R<0.6: Medium risk;

[0150] R≥0.6: High risk.

[0151] 3) Generation of risk heatmap

[0152] The rear area was divided into a 20×20 grid (each grid is 0.5m×0.5m), and the thermal value of each grid was calculated according to the risk level.

[0153] Heat_value=1-(R-R_min) / (R_max-R_min);

[0154] R: Risk level;

[0155] R_min: The minimum risk level in the current area;

[0156] R_max: The maximum risk level in the current area.

[0157] Heatmap color mapping:

[0158] Red (0.8-1.0): High-risk area;

[0159] Yellow (0.5-0.8): Medium-risk area;

[0160] Green (0.0-0.5): Low-risk area.

[0161] 4) Calculation of potential collision points

[0162] Potential collision points can be calculated using the following steps:

[0163] Calculate the collision time T based on the relative speed V_rel of the vehicle behind and the current distance D:

[0164] T = D / V_rel;

[0165] Based on the vehicle's braking performance, calculate the maximum distance D_move that the vehicle can move within time T:

[0166] D_move=0.5×a×T 2 ;

[0167] In the above formula, 'a' represents the maximum deceleration of the vehicle (m / s²). 2 For example, take 5m / s 2 );

[0168] Potential collision point coordinates (X_c, Y_c):

[0169] X_c=X_vehicle+(D_move×cos(θ));

[0170] Y_c=Y_vehicle+(D_move×sin(θ));

[0171] Where θ is the relative angle between the vehicle and the vehicle behind it.

[0172] Potential collision points can be marked on the risk heat map with red circles.

[0173] Phase 2: Intelligent Micro-Avoidance (i.e., the aforementioned micro-avoidance phase, for example, 3-5 seconds before a collision occurs)

[0174] -Activate only when the space quality assessment is "Safe Minor Avoidance Level";

[0175] - By controlling the steering system, perform micro-avoidance of 10cm-20cm;

[0176] - The avoidance process is smooth, does not startle the driver, and maintains vehicle stability.

[0177] illustrate:

[0178] 1) Micro-avoidance control principle

[0179] A steering fine-tuning algorithm based on PD control can be used to achieve smooth 10cm-20cm micro-avoidance:

[0180] Δθ = Kp × e + Kd × (e - e_prev);

[0181] In the above formula, Δθ: steering angle increment, the amount of steering angle change required by the vehicle to achieve micro-avoidance;

[0182] e: Deviation between the current vehicle position and the target avoidance position (lateral distance between the current lateral position of the vehicle and the ideal avoidance position).

[0183] e_prev: The deviation from the previous time step;

[0184] Kp: Proportional coefficient, the system's response strength to the current deviation, the "force" of steering control;

[0185] Kd: Differential coefficient, the system's response to the rate of change of deviation, the "smoothness" of the steering control;

[0186] To balance real-time performance, smoothness, system load, and actuator capability, the calculation frequency of Δθ is set to once every 10ms–20ms.

[0187] 2) Control Process

[0188] 2.1) Target location calculation:

[0189] The best avoidance direction (such as right, left or rear) can be determined based on the space quality score.

[0190] Calculate the target position: X_target = X_current + ΔX, Y_target = Y_current;

[0191] ΔX is the avoidance distance (10cm-20cm, which can be dynamically adjusted according to the Q value: ΔX=20cm when Q≥0.8, ΔX=15cm when 0.7≤Q<0.8, and ΔX=10cm when 0.4≤Q<0.7).

[0192] 2.2) Steering control:

[0193] The steering angle increment Δθ is calculated every 10ms.

[0194] Steering is achieved through the steering system, with a control precision of 0.1°.

[0195] Steering speed should be kept below 0.5° / s to ensure a smooth process.

[0196] 2.3) Stability Guarantee:

[0197] The vehicle stability control system monitors the vehicle's attitude in real time.

[0198] When the roll angle is detected to exceed 5°, the steering speed is automatically reduced;

[0199] Lateral acceleration is monitored using an accelerometer to ensure it does not exceed 0.3g.

[0200] 3) Case Study

[0201] Initial state: The vehicle is stationary, 30m away from the vehicle behind it;

[0202] The system determines that Q=0.78 (safe to slightly avoid), and a slight movement of 15cm to the right is required.

[0203] Control process:

[0204] 0ms-500ms: The steering angle smoothly increases from 0° to 0.8°;

[0205] 500ms-1000ms: Maintain a 0.8° steering angle;

[0206] 1000ms-1500ms: Steering angle smoothly decreases to 0°;

[0207] Result: The vehicle moved 15cm to the right, with a peak lateral acceleration of 0.25g, which the driver did not perceive.

[0208] Phase 3: Collision Response and Secondary Protection (i.e., the aforementioned unavoidable level, when a collision occurs)

[0209] -If micro-avoidance cannot be completed, the preset lateral sliding mechanism will be automatically activated;

[0210] - Automatically adjust the steering angle based on the collision force to achieve "collision force utilization";

[0211] - Works in conjunction with AEB and RCTB systems to form comprehensive protection.

[0212] illustrate:

[0213] 1) Lateral gliding preparation:

[0214] Activate the lateral coasting mode via the vehicle stability control system;

[0215] Adjust the braking pressure of the left and right wheels to induce lateral slippage of the vehicle;

[0216] The slip angle is controlled within 5° to ensure vehicle stability.

[0217] 2) Steering angle adjustment based on collision force:

[0218] Real-time monitoring of collision force F_collide (which can be calculated via IMU and wheel speed sensors);

[0219] Calculate the optimal steering angle θ_opt:

[0220] θ_opt=K×F_collide+θ_base;

[0221] In the above formula, K: collision force coefficient (which can be determined by the vehicle dynamics model, 0.005° / N);

[0222] F_collide: Collision force (unit: N);

[0223] θ_base: Base steering angle (can be determined based on road curvature, for example, 0°-3°).

[0224] 3) Utilization of collision force:

[0225] The lateral component generated by the collision force can be used to automatically steer the vehicle when a collision occurs.

[0226] By controlling the braking pressure, the collision force is maximized for steering.

[0227] 4) Right rear collision timing coordination logic

[0228] 4.1) Time point T = -3.0s (The vehicle to the right rear is approximately 50m away from this vehicle)

[0229] Lateral skid control:

[0230] Activate surround view perception fusion;

[0231] Identify high-risk targets to the right rear;

[0232] Predict the optimal avoidance direction: Slightly move to the left (to avoid the obstacle on the right);

[0233] Set θ_base=1.2° (to compensate for minor road curvatures).

[0234] AEB system:

[0235] A lateral risk was detected at the rear.

[0236] Broadcast RearObliqueRisk=True to the CAN bus;

[0237] Do not brake for now (vehicle is traveling at low speed).

[0238] RCTB system:

[0239] Initiate rearward radar + camera fusion;

[0240] Estimate the collision offset (confirmed to be in the right rear region);

[0241] Prepare to apply braking to the right rear wheel.

[0242] 4.2) T = -1.5s (The vehicle to the right rear is approximately 25m away from this vehicle)

[0243] Lateral skid control:

[0244] Perform a slight dodge to the left (10cm-18cm);

[0245] The vehicle slowly veered to the left, creating space on the right.

[0246] AEB system:

[0247] Remind the driver (HUD flashing + seat vibration);

[0248] Pre-charged main braking system.

[0249] RCTB system:

[0250] The collision area has been confirmed to be the right rear quadrant.

[0251] The upper limit of the braking pressure of the right rear wheel is preset.

[0252] 4.3) T = -0.4s (400ms before collision)

[0253] Lateral coasting control: (Initiation of lateral coasting pre-charge)

[0254] ESP applies pre-pressure braking to the left front wheel and left rear wheel;

[0255] Objective: To generate a counterclockwise yaw moment (turning the car left).

[0256] AEB system:

[0257] Lock the brake pedal;

[0258] Activate the pretensioning seat belts (front and rear seats).

[0259] RCTB system:

[0260] Provides the maximum available braking force for the right rear wheel;

[0261] Prepare to release the right rear wheel at the moment of impact (to reduce drag and assist in deflection).

[0262] 4.4) T=0s (the instant of the collision at the right rear)

[0263] Lateral skid control:

[0264] Apply full braking force to the left wheel (e.g., 70 bar).

[0265] Right wheel active brake release (RCTB coordination);

[0266] The vehicle developed a 4.2° slip angle, and the front of the vehicle veered to the left;

[0267] The lateral component of the collision force F_collide is used to enhance the steering effect.

[0268] AEB system:

[0269] Apply full braking force (to prevent forward lurching);

[0270] Record the collision G-value and direction.

[0271] RCTB system:

[0272] Release the brake on the right rear wheel;

[0273] To prevent the right rear wheel from "dragging" and hindering the turn;

[0274] If the steering is excessive, apply additional braking to the left rear wheel.

[0275] 4.5)T = +0.3s (after the collision)

[0276] Lateral skid control:

[0277] Monitor yaw rate and sideslip angle;

[0278] Gradually return to positive and turn.

[0279] AEB system:

[0280] Keep the brakes engaged until the vehicle comes to a complete stop;

[0281] Automatically turn on hazard lights.

[0282] RCTB system:

[0283] Restore regular monitoring;

[0284] If vehicle rotation is detected, activate the stabilization brakes.

[0285] The technical solution provided in this embodiment first acquires multi-sensor data, extracts key feature arrays based on the multi-sensor data, and calculates a spatial quality score based on the key feature array and a preset weight array. Then, it determines the rear-end collision prevention level based on the spatial quality score and a score threshold array. Afterward, it determines the vehicle braking control strategy and the segmented avoidance strategy based at least on the rear-end collision prevention level, and executes the corresponding strategies. Therefore, the vehicle rear-end collision prevention method in this embodiment can at least perform dynamic assessment of the spatial quality of the vehicle, which is beneficial for effective avoidance in space-constrained areas. Furthermore, the segmented avoidance strategy can reduce the probability of a primary collision while mitigating the risk of a secondary collision, thus protecting the personal and property safety of users.

[0286] It should be noted that, based on the above embodiments or implementation methods, this invention also provides an adaptive learning mechanism for a vehicle rear-end collision prevention method, as detailed below:

[0287] The adaptive learning mechanism consists of at least three sub-mechanisms: environmental adaptation, behavioral learning, and driver habit adaptation.

[0288] (1) Environmental Adaptation: Automatically adjust the avoidance strategy according to different scenarios such as urban roads, highways, and parking lots.

[0289] Scene recognition:

[0290] Identify the current environment using map data (such as high-precision maps) and sensor data (such as cameras, GPS, etc.);

[0291] Recognition algorithm: Deep learning-based scene classification model (such as ResNet-18), with an accuracy of 98.5%;

[0292] Identification categories: urban roads, highways, parking lots, school zones, construction zones, etc.

[0293] Strategy Adjustment:

[0294] Urban roads (low-speed scenario): Avoidance margin 10cm, warning lead time 5 seconds;

[0295] Highway (high-speed scenario): Avoidance margin 25cm, warning lead time 3 seconds;

[0296] Parking lot (limited space): Avoidance margin 8cm, warning lead time 7 seconds;

[0297] School area (high pedestrian density): Avoidance margin 15cm, warning lead time 6 seconds.

[0298] (2) Behavioral learning: Learn the best avoidance strategy in different scenarios through cloud data.

[0299] Data collection:

[0300] Detailed data for each avoidance event (including environmental parameters, space quality score, avoidance result, etc.) is stored in the cloud.

[0301] Data dimensions include: environment type, vehicle speed, distance between vehicles, avoidance range, and driver reaction time.

[0302] Learning Algorithms:

[0303] The avoidance strategy is optimized using a reinforcement learning (Q-learning) algorithm;

[0304] Each successful avoidance is automatically rewarded with +1; each failed avoidance is penalized with -1.

[0305] State space: environment type, vehicle speed, vehicle distance, and space quality score;

[0306] Action space: avoidance direction, avoidance range, and warning lead time.

[0307] Strategy optimization:

[0308] The strategy library is updated monthly and optimized based on cloud-based learning data.

[0309] Optimization objective: Maximize the success rate of obstacle avoidance while minimizing driver startle.

[0310] (3) Driver habit adaptation: Adjust the aggressiveness of warning and avoidance according to the driver's driving habits.

[0311] Habit recognition:

[0312] Identify driver style using driving data such as steering wheel angle, accelerator / brake pedal force, and steering habits;

[0313] Identification algorithm: Based on cluster analysis (K-means), drivers are divided into "conservative", "aggressive", and "moderate" types;

[0314] Identification dimensions: steering habits, braking habits, and obstacle avoidance response time.

[0315] Habit adaptation:

[0316] Conservative drivers: Increase the lead time for warnings and reduce the magnitude of avoidance maneuvers;

[0317] Aggressive drivers: Reduce the amount of advance warning and increase the avoidance range;

[0318] Stable driving: Maintain standard parameters.

[0319] Adaptive adjustment:

[0320] Parameters are dynamically adjusted based on the driver's real-time driving behavior.

[0321] After each avoidance maneuver, the system assesses the driver's reaction and fine-tunes the parameters.

[0322] It should also be noted that the following are several scenario examples for illustration.

[0323] Scenario Example 1: Urban traffic light intersection scenario

[0324] 1. Real-time assessment of the approaching speed of vehicles behind and the mass of the surrounding space;

[0325] 2. When a vehicle is detected approaching from behind at a speed of 80 km / h and the distance is less than 50 meters;

[0326] 3. The space quality is determined to be "safely slightly avoidable" (Q=0.75), initiating phase 2;

[0327] 4. Move the vehicle slightly to the right by 15 centimeters to provide safe passage for vehicles behind;

[0328] 5. Throughout the process, the in-vehicle display screen shows "Safety space has been adjusted," allowing the driver to clearly understand the system's operating status.

[0329] Scenario Example 2: Parking Lot Scenario

[0330] 1. Assess the parking lot environment and identify surrounding obstacles and pedestrians;

[0331] 2. If the parking lot space is limited, adopt a more conservative avoidance strategy;

[0332] 3. When a vehicle approaches from behind, prioritize slightly shifting to the left (assuming there is an obstacle on the right).

[0333] 4. The system provides an audio prompt: "There is an obstacle on the right; we have moved slightly to the left."

[0334] 5. The avoidance process is smooth and does not affect the passage of other vehicles in the parking lot.

[0335] Scenario Example 3: Highway

[0336] 1. Identified as a high-speed scene, a more aggressive avoidance strategy is adopted;

[0337] 2. When a vehicle approaches at a high speed from behind, initiate stage 2 in advance;

[0338] 3. The margin for maneuvering has been increased to 25 centimeters, providing more safety space for vehicles behind;

[0339] 4. A voice prompt will indicate "Vehicles are approaching from behind; position has been slightly adjusted."

[0340] 5. Works in conjunction with AEB to ensure safety in high-speed scenarios.

[0341] In view of this, the embodiments of the present invention can at least alleviate the technical problems existing in the prior art:

[0342] 1. Electronic fence technology is fixed and lacks environmental adaptability, making it impossible to dynamically adjust according to the actual quality of the space (such as ground flatness, obstacle type, pedestrian activity, etc.);

[0343] 2. Only a single strategy of "active avoidance" or "passive response" is adopted, without forming a full-chain security protection of "prevention-response-optimization";

[0344] 3. The lack of a driver interaction mechanism means that drivers are unaware of the system's operating status, which may lead to distrust.

[0345] 4. The inability to differentiate between different scenarios (city roads, highways, parking lots, etc.) and the adoption of a uniform strategy leads to poor performance in some scenarios;

[0346] 5. The first collision cannot be effectively avoided; only secondary protection after the collision is considered.

[0347] Meanwhile, the embodiments of the present invention can achieve at least the following beneficial effects:

[0348] 1. Enables dynamic assessment of spatial quality, improving spatial adaptability by 300%;

[0349] 2. By employing a phased avoidance strategy, the risk of a secondary collision can be reduced while avoiding the first collision.

[0350] 3. Enhance driver trust in the system through intelligent human-machine interaction;

[0351] 4. Through an adaptive learning mechanism, the system can automatically adjust its strategies according to different scenarios;

[0352] 5. When stationary at intersections with traffic lights, the system can automatically identify the risk of vehicles behind and make precise avoidance maneuvers before a collision occurs.

[0353] Of course, the embodiments of the present invention also provide the following alternative solutions:

[0354] 1. Only dynamic spatial assessment is used, but no phased avoidance strategy is adopted;

[0355] 2. Only a phased avoidance strategy is adopted, but dynamic spatial assessment is not used;

[0356] 3. Adopt avoidance strategies with different phase divisions (such as four-phase or two-phase);

[0357] 4. Only an adaptive learning mechanism is used, without combining dynamic space evaluation and phased avoidance.

[0358] Figure 2 This is a schematic diagram of a vehicle rear-end collision prevention device provided in an embodiment of the present invention. This embodiment is at least applicable to rear-end collision prevention scenarios when the vehicle is stationary. The vehicle rear-end collision prevention device can be implemented using software and / or hardware. Figure 2 As shown, the vehicle rear-end collision prevention device is used to perform the vehicle rear-end collision prevention method of any of the foregoing embodiments or implementations.

[0359] Vehicle rear-end collision prevention devices include at least:

[0360] The level determination module 110 is used to acquire multi-sensor data, extract key feature arrays based on multi-sensor data, calculate space quality score based on key feature arrays and preset weight arrays, and then determine the rear-end collision prevention level based on space quality score and score threshold array.

[0361] The strategy execution module 120 is used to determine the vehicle braking control strategy and the segmented avoidance strategy based at least on the rear-end collision prevention level, and then complete the corresponding strategy execution.

[0362] Optionally, the space quality score may be calculated at least in the following ways:

[0363] Q=α1×(S / Smax)+α2×(1-R / Rmax)+α3×(1-F / Fmax)+α4×(1-P / Pmax);

[0364] In the above formula, Q represents the space quality score, α1 represents the first preset weight, α2 represents the second preset weight, α3 represents the third preset weight, α4 represents the fourth preset weight, S represents the area of ​​the vehicle's avoidable zone, Smax represents the maximum usable space area of ​​the vehicle, R represents the ground unevenness, Rmax represents the maximum ground unevenness allowed for the vehicle to avoid rear-end collisions, F represents the obstacle density, Fmax represents the maximum obstacle density allowed for the vehicle to avoid rear-end collisions, P represents the pedestrian activity intensity, Pmax represents the maximum pedestrian activity intensity allowed for the vehicle to avoid rear-end collisions;

[0365] The first preset weight, the second preset weight, the third preset weight, and the fourth preset weight together form a preset weight array.

[0366] Optionally, the scoring threshold array includes at least a first scoring threshold and a second scoring threshold, wherein the first scoring threshold is less than the second scoring threshold;

[0367] The rear-end collision protection level can be determined by at least the following methods:

[0368] When the space quality score is lower than the first score threshold, the rear-end collision avoidance level is determined to be the unavoidable level.

[0369] When the space quality score is not lower than the first score threshold but lower than the second score threshold, the rear-end collision prevention level is determined to be the cautious and slightly avoidable level.

[0370] When the space quality score is not lower than the second score threshold, the rear-end collision avoidance level is determined to be the safe minor avoidance level.

[0371] Optionally, when the rear-end collision avoidance level is the unavoidable level, the vehicle braking control strategy is at least to adjust the braking intensity of the vehicle's AEB system to the first AEB intensity and the braking intensity of the vehicle's RCTB system to the first RCTB intensity.

[0372] Optionally, when the rear-end collision level is the cautious avoidance level, the vehicle braking control strategy shall at least be to adjust the braking intensity of the vehicle's AEB system to the second AEB intensity and the braking intensity of the vehicle's RCTB system to the second RCTB intensity.

[0373] The second AEB intensity is less than the first AEB intensity, and the second RCTB intensity is less than the first RCTB intensity.

[0374] Optionally, when the rear-end collision level is the safety minor avoidance level, the vehicle braking control strategy is at least to maintain the braking intensity of the vehicle's AEB system and the braking intensity of the vehicle's RCTB system at the current moment.

[0375] Optionally, the segmented avoidance strategy includes at least a warning preparation phase, a micro-avoidance phase, and a collision protection phase.

[0376] The technical solution provided in this embodiment first acquires multi-sensor data through a level determination module, extracts key feature arrays based on the multi-sensor data, and calculates a spatial quality score based on the key feature array and a preset weight array. Then, it determines the rear-end collision prevention level based on the spatial quality score and a score threshold array. Afterward, the strategy execution module determines at least the vehicle braking control strategy and the segmented avoidance strategy based on the rear-end collision prevention level, and completes the corresponding strategy execution. Therefore, the vehicle rear-end collision prevention method of this embodiment can at least perform dynamic evaluation of the spatial quality of the vehicle, which is beneficial for achieving effective avoidance of the vehicle in space-constrained areas. In addition, the segmented avoidance strategy can reduce the probability of a primary collision while mitigating the risk of a secondary collision, thus helping to protect the personal and property safety of users.

[0377] This embodiment provides an electronic device. Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. See also: Figure 3 The electronic device 1000 includes a processor 1001 and a memory 1002. The memory 1002 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 1001, the steps in any of the above-described vehicle rear-end collision prevention methods are performed. Through the above technical solution, the processor 1001 and the memory 1002 are interconnected and communicate with each other through a communication bus and / or other forms of connection mechanism (not shown). The memory 1002 stores a computer program executable by the processor. When the electronic device 1000 is running, the processor 1001 executes the computer program to perform the vehicle rear-end collision prevention method in any optional implementation of the above embodiments, so as to at least achieve the following functions: acquiring multi-sensor data, extracting key feature arrays based on multi-sensor data, calculating a spatial quality score based on the key feature array and a preset weight array, and then determining the rear-end collision prevention level based on the spatial quality score and a score threshold array; determining the vehicle braking control strategy and the segmented avoidance strategy based at least on the rear-end collision prevention level, and then completing the corresponding strategy execution.

[0378] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the vehicle rear-end collision prevention method provided in all embodiments of this application: acquiring multi-sensor data, extracting key feature arrays based on the multi-sensor data, calculating a spatial quality score based on the key feature array and a preset weight array, and then determining the rear-end collision prevention level based on the spatial quality score and a score threshold array; determining the vehicle braking control strategy and segmented avoidance strategy based at least on the rear-end collision prevention level, and then completing the corresponding strategy execution.

[0379] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0380] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including—but not limited to—electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0381] The program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0382] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—as well as conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0383] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for preventing rear-end collisions of vehicles, characterized in that, The vehicle rear-end collision prevention method is applicable at least to scenarios where the vehicle is stationary, and includes at least the following: Acquire multi-sensor data, extract key feature arrays based on the multi-sensor data, calculate space quality score based on the key feature array and preset weight array, and then determine the rear-end collision prevention level based on the space quality score and score threshold array. After determining the vehicle braking control strategy and segmented avoidance strategy based on the aforementioned rear-end collision prevention level, the corresponding strategy execution is completed.

2. The vehicle rear-end collision prevention method according to claim 1, characterized in that, The space quality score is calculated in at least the following ways: Q=α1×(S / Smax)+α2×(1-R / Rmax)+α3×(1-F / Fmax)+α4×(1-P / Pmax); In the above formula, Q represents the space quality score, α1 represents the first preset weight, α2 represents the second preset weight, α3 represents the third preset weight, α4 represents the fourth preset weight, S represents the area of ​​the vehicle's avoidable zone, Smax represents the maximum usable space area of ​​the vehicle, R represents the ground unevenness, Rmax represents the maximum ground unevenness allowed for vehicle rear-end collision prevention, F represents the obstacle density, Fmax represents the maximum obstacle density allowed for vehicle rear-end collision prevention, P represents the pedestrian activity intensity, Pmax represents the maximum pedestrian activity intensity allowed for vehicle rear-end collision prevention; The first preset weight, the second preset weight, the third preset weight, and the fourth preset weight together constitute the preset weight array.

3. The vehicle rear-end collision prevention method according to claim 1, characterized in that, The scoring threshold array includes at least a first scoring threshold and a second scoring threshold, wherein the first scoring threshold is less than the second scoring threshold; The rear-end collision prevention level is determined by at least the following methods: When the space quality score is lower than the first score threshold, the rear-end collision avoidance level is determined to be an unavoidable level. When the space quality score is not lower than the first score threshold but lower than the second score threshold, the rear-end collision avoidance level is determined to be the cautious and slightly avoidable level. When the space quality score is not lower than the second score threshold, the rear-end collision avoidance level is determined to be the safety minor avoidance level.

4. The vehicle rear-end collision prevention method according to claim 1, characterized in that, When the rear-end collision avoidance level is the unavoidable level, the vehicle braking control strategy is at least to adjust the braking intensity of the vehicle's AEB system to the first AEB intensity and the braking intensity of the vehicle's RCTB system to the first RCTB intensity.

5. The vehicle rear-end collision prevention method according to claim 4, characterized in that, When the rear-end collision level is the cautious avoidance level, the vehicle braking control strategy is at least to adjust the braking intensity of the vehicle's AEB system to the second AEB intensity and the braking intensity of the vehicle's RCTB system to the second RCTB intensity. Wherein, the second AEB intensity is less than the first AEB intensity, and the second RCTB intensity is less than the first RCTB intensity.

6. The vehicle rear-end collision prevention method according to claim 4, characterized in that, When the rear-end collision level is the safety minor avoidance level, the vehicle braking control strategy is at least to maintain the braking intensity of the vehicle's AEB system and the braking intensity of the vehicle's RCTB system at the current moment's intensity.

7. The vehicle rear-end collision prevention method according to claim 1, characterized in that, The segmented avoidance strategy includes at least a warning preparation phase, a micro-avoidance phase, and a collision protection phase.

8. A vehicle rear-end collision prevention device, characterized in that, Used to perform the vehicle rear-end collision prevention method according to any one of claims 1-7; The vehicle rear-end collision prevention device includes at least: The level determination module is used to acquire multi-sensor data, extract key feature arrays based on the multi-sensor data, calculate space quality scores based on the key feature arrays and preset weight arrays, and then determine the rear-end collision prevention level based on the space quality scores and score threshold arrays. The strategy execution module is used to determine the vehicle braking control strategy and the segmented avoidance strategy based at least on the rear-end collision avoidance level, and then complete the corresponding strategy execution.

9. An electronic device comprising a memory and a processor, the memory storing a computer program executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the vehicle rear-end collision prevention method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the vehicle rear-end collision prevention method according to any one of claims 1 to 7.

Images