A method, system, electronic device, and storage medium for generating braking decision information.

Abstract

This application provides a braking decision information generation method, system, electronic device, and storage medium, relating to the field of vehicle control technology. The method includes: performing target detection on current driving environment data to determine a candidate target set; conducting collision risk assessment on each candidate target in the candidate target set to obtain a collision risk score for each candidate target; determining a high-collision-risk target set based on the collision risk scores of each candidate target; performing collision verification on each high-collision-risk target in the high-collision-risk target set; and generating braking decision information for the current moment based on the collision verification results of each high-collision-risk target. This can improve the accuracy of hazardous target identification, reduce misjudgments caused by perception fluctuations, multi-target interference, etc., and thus improve the accuracy of braking decisions.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and in particular to a method, system, electronic device, and storage medium for generating braking decision information. Background Technology

[0002] With the rapid development of intelligent connected vehicle technology, AEB (Automatic Emergency Braking) has become an important support for improving the active safety performance of vehicles. In actual driving, different types of risk targets may appear in front of the vehicle, such as stationary vehicles, vehicles that suddenly decelerate, pedestrians crossing the road, and two-wheeled vehicles. The collision risk posed by different risk targets varies significantly with different vehicle speeds, relative distances, and motion states.

[0003] However, existing AEB systems typically assess target risk based on fixed thresholds or weights and determine braking targets based on the current frame perception results. When sensor detection results fluctuate, targets are temporarily occluded, or multiple risky targets exist simultaneously, problems such as inaccurate risk assessment and frequent switching of braking targets may occur.

[0004] Therefore, this application proposes a new method for generating braking decision information. Summary of the Invention

[0005] In view of the above problems, embodiments of this application provide a braking decision information generation method, system, electronic device, and storage medium to overcome or at least partially solve the above problems.

[0006] A first aspect of this application provides a method for generating braking decision information, the method comprising: Target detection is performed on the current driving environment data to determine a set of candidate targets; Each candidate target in the candidate target set is assessed for collision risk to obtain a collision risk score for each candidate target. Based on the collision risk scores of each candidate target, a set of high-collision-risk targets is determined. Collision verification is performed on each high-collision-risk target in the set of high-collision-risk targets; Based on the collision verification results of each high-collision-risk target, braking decision information for the current moment is generated.

[0007] Optionally, a collision risk assessment is performed on each candidate target in the candidate target set to obtain a collision risk score for each candidate target, including: Based on the vehicle's current speed, determine the target speed range where the current speed is located from multiple speed ranges; Based on the target vehicle speed range, determine the target TTC threshold, the target distance threshold, and the target weight combination; Based on the target TTC threshold and the collision time TTC of each candidate target in the candidate target set, determine the TTC score of each candidate target in the candidate target set; Based on the target distance threshold and the distances between each candidate target in the candidate target set and the vehicle, a distance score is determined for each candidate target in the candidate target set. Predict the collision probability of each candidate target in the candidate target set to obtain a collision probability score for each candidate target in the candidate target set; For each candidate target in the candidate target set, the TTC score, distance score, and collision probability score of the candidate target are fused according to the TTC weight, distance weight, and collision probability weight in the target weight combination to obtain the collision risk score of the candidate target.

[0008] Optionally, collision verification is performed on each high-collision-risk target in the set of high-collision-risk targets, including: Based on the TTC of the high-collision-risk target, multiple future moments are determined, including: a first future moment, a second future moment, and a third future moment; the first future moment is the moment corresponding to the product of the TTC of the high-collision-risk target and a first preset ratio, where the first preset ratio is greater than 0 and less than 1; the second future moment is the moment corresponding to the product of the TTC of the high-collision-risk target and a second preset ratio, where the second preset ratio is greater than the first preset ratio and less than 1; the third future moment is the moment corresponding to the TTC of the high-collision-risk target; the time intervals between the first future moment, the second future moment, the third future moment, and the current moment increase sequentially; Determine the OBB of the vehicle and the high-collision-risk target at the current time, the first future time, the second future time, and the third future time, respectively; Based on whether there is overlap between the vehicle and the high-collision-risk target's OBB at the current time, the first future time, the second future time, and the third future time, it is determined whether the vehicle and the high-collision-risk target collide at the current time, the first future time, the second future time, and the third future time.

[0009] Optionally, based on the collision verification results of each high-collision-risk target, braking decision information for the current moment is generated, including: Based on the collision verification results of each high-collision-risk target, the braking phase of each high-collision-risk target is determined; From the braking phases of the various high-collision-risk targets, the highest priority target braking phase is determined; among them, the priority of the Automatic Emergency Braking (AEB) phase, Automatic Warning Braking (AWB) phase, and Automatic Braking Pre-charge phase decreases in that order. If the target braking phase is the AEB phase, braking decision information for performing automatic emergency braking operation at the current moment is generated; When the target braking stage is the AWB stage, braking decision information is generated to perform automatic alarm braking operation at the current moment; If the target braking phase is the automatic braking pre-charge phase, braking decision information is generated to perform the automatic braking pre-charge operation at the current moment.

[0010] Optionally, the high-collision-risk targets include a first high-collision-risk target, a second high-collision-risk target, and a third high-collision-risk target; based on the collision verification results of the high-collision-risk targets, the braking phase of each high-collision-risk target is determined, including: If the collision verification result of the first high-collision-risk target is that the vehicle and the first high-collision-risk target collide at each of the current time and at least one future time, the braking phase of the first high-collision-risk target is determined to be the automatic emergency braking (AEB) phase. If the collision verification result of the second high collision risk target is: the vehicle and the second high collision risk target collide at both the second future time and the third future time, and no collision occurs at the current time and the first future time, then the braking phase of the second high collision risk target is determined to be the Automatic Warning Braking (AWB) phase. If the collision verification result of the third highest collision risk target is: the vehicle and the third highest collision risk target collide at the third future time, and no collision occurs at the current time, the first future time, and the second future time, then the braking phase of the third highest collision risk target is determined to be the automatic braking pre-charge phase.

[0011] Optionally, after generating the braking decision information for the current moment, the method further includes: The first high-collision-risk target is locked as the old target; Based on the driving environment data at the next moment, determine the new target for the next moment in the Automatic Emergency Braking (AEB) phase; Based on the first distance between the new target and the vehicle at the next moment, and the second distance between the old target and the vehicle at the next moment, determine the distance difference between the first distance and the second distance; If the distance difference is not less than the lower limit threshold of the distance difference and not greater than the upper limit threshold of the distance difference, the first high-collision-risk target is kept as the target for performing automatic emergency braking at the next moment. If the distance difference is less than the lower limit threshold of the distance difference, the new target is determined as the target for performing automatic emergency braking operation at the next moment.

[0012] Optionally, the collision time (TTC) of each candidate target in the candidate target set is determined according to the following steps: The initial TTC of the candidate target is determined based on the current vehicle speed, the current speed of each candidate target in the candidate target set, the relative distance between the vehicle and the candidate target, and the relative acceleration. Based on the current vehicle speed, determine the time for the vehicle to come to a complete stop; If the initial TTC of the candidate target is less than or equal to the complete braking time of the vehicle, the initial TTC is determined to be valid, and the initial TTC is determined as the final TTC of the candidate target. If the initial TTC of the candidate target is greater than the vehicle's complete braking time, the following steps are performed based on the candidate target's complete braking time: If the initial TTC of the candidate target is greater than the complete braking time of the candidate target, after the candidate target comes to a complete stop, the TTC of the candidate target is re-determined, and the re-determined TTC is determined as the final TTC of the candidate target. If the initial TTC of the candidate target is less than or equal to the complete braking time of the candidate target, after the vehicle comes to a complete stop, the TTC of the candidate target is re-determined, and the re-determined TTC is determined as the final TTC of the candidate target. The method further includes: If, after both the vehicle and the candidate target have come to a complete stop, there is still no collision between the vehicle and the candidate target, the final TTC of the candidate target is set to the no-collision risk indication value.

[0013] Optionally, target detection is performed on the driving environment data at the current moment to determine a set of candidate targets, including: Target detection is performed on the driving environment data at the current moment to obtain multiple initial targets, and the corresponding vehicle-mounted sensor for each initial target is determined. The vehicle-mounted sensors include: a forward-looking camera, a forward-facing radar, a line radar, and an angle radar. If the initial target corresponds to at least two vehicle-mounted sensors, the initial target is identified as a candidate target and added to the candidate target set; In the case where the initial target corresponds to a single vehicle-mounted sensor, it is determined whether the initial target meets the exemption conditions. The exemption conditions include at least one of the following: the type of the initial target is a preset target type, the ratio of the lateral velocity to the longitudinal velocity of the initial target is greater than a ratio threshold, the motion angle of the initial target is within the angle range, the longitudinal distance between the initial target and the vehicle is less than a distance threshold, and the lateral velocity of the initial target is greater than a velocity threshold. If the initial target satisfies the exemption conditions, the initial target is identified as a candidate target and added to the candidate target set; If the initial target does not meet the exemption conditions, the initial target is removed.

[0014] A second aspect of this application provides a braking decision information generation system, the system comprising: The determination module is used to perform target detection on the driving environment data at the current moment and determine the set of candidate targets; The evaluation module is used to evaluate the collision risk of each candidate target in the candidate target set, obtain the collision risk score of each candidate target, and determine the set of high collision risk targets based on the collision risk scores of each candidate target. The verification module is used to perform collision verification on each high-collision-risk target in the set of high-collision-risk targets; The generation module is used to generate braking decision information for the current moment based on the collision verification results of each high-collision-risk target.

[0015] A third aspect of this application provides an electronic device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the braking decision information generation method as described in the first aspect of this application.

[0016] A fourth aspect of this application provides a readable storage medium storing a program or instructions that, when executed by a processor, implement the steps of the braking decision information generation method as described in the first aspect of this application.

[0017] The beneficial effects of this application are: This application provides a method for generating braking decision information. The method includes: performing target detection on driving environment data at the current moment to determine a set of candidate targets; performing collision risk assessment on each candidate target in the set of candidate targets to obtain a collision risk score for each candidate target; determining a set of high collision risk targets based on the collision risk scores of each candidate target; performing collision verification on each high collision risk target in the set of high collision risk targets; and generating braking decision information at the current moment based on the collision verification results of each high collision risk target.

[0018] This application detects targets in driving environment data, first filtering high-collision-risk targets from candidate targets based on collision risk scores, then performing collision verification on these high-collision-risk targets, and generating braking decision information based on the collision verification results. This improves the accuracy of hazardous target identification, reduces misjudgments caused by perception fluctuations and multi-target interference, and ultimately improves the accuracy of braking decisions. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the steps of a braking decision information generation method provided in an embodiment of this application; Figure 2 This is a schematic diagram of a collision risk scoring process provided in an embodiment of this application; Figure 3 This is a geometrical diagram of collision confirmation provided in an embodiment of this application; Figure 4 This is a schematic diagram of a braking decision generation process provided in an embodiment of this application; Figure 5 This is a flowchart of a progressive TTC calculation logic provided in an embodiment of this application; Figure 6 This is a schematic diagram of a braking decision information generation system provided in an embodiment of this application; Figure 7 This is a schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0021] Exemplary embodiments of this application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of this application and to fully convey the scope of this application to those skilled in the art.

[0022] In a first aspect, this application provides a method for generating braking decision information, such as... Figure 1 As shown, the method includes: S101: Target detection is performed on the driving environment data at the current moment to determine the set of candidate targets.

[0023] In this step, the driving environment data corresponding to the current moment is obtained. In practical applications, the driving environment data may include environmental perception data collected by multiple sources of sensors such as forward-looking cameras, millimeter-wave radar, and lidar.

[0024] Furthermore, target detection is performed on the driving environment data to identify targets present in the vehicle's current driving environment. Based on the target detection results, a candidate target set is determined. In some cases, the acquired multi-source environmental perception data can be fused to obtain the current driving environment data, and target detection can be performed on the fused driving environment data to obtain a candidate target set. In some cases, the candidate target set may include one or more candidate targets.

[0025] S102, perform collision risk assessment on each candidate target in the candidate target set to obtain a collision risk score for each candidate target, and determine a high collision risk target set based on the collision risk scores of each candidate target.

[0026] In this step, after determining the candidate target set, a collision risk assessment can be performed on each candidate target in the set to determine the collision risk score for each candidate target. The collision risk score characterizes the degree of risk of a collision between the corresponding candidate target and the vehicle.

[0027] Furthermore, based on the collision risk scores corresponding to each candidate target, high collision risk targets can be determined from the candidate target set in one of the following ways: candidate targets with collision risk scores greater than or equal to a preset risk score threshold are determined as high collision risk targets; or, the collision risk scores are sorted from high to low, and the top N candidate targets are determined as high collision risk targets, where N is a preset number; the determined candidate targets are used as high collision risk targets to obtain a set of high collision risk targets.

[0028] In one implementation, risk scores for candidate targets can be calculated based on multiple risk dimensions. These risk dimensions may include at least one of the following: Time To Collision (TTC), target distance, path overlap between the vehicle and the target, Longitudinal Distance Necessary for Evasive Collision Avoidance (LDNEC), and collision probability. Then, based on the dynamic weights corresponding to each risk dimension, the risk scores are fused to obtain a comprehensive risk score for the candidate target.

[0029] S103, perform collision verification on each high collision risk target in the set of high collision risk targets.

[0030] In this step, after determining the set of high-collision-risk targets, collision verification can be performed on each high-collision-risk target in the set to determine the collision verification result corresponding to each high-collision-risk target.

[0031] By performing further collision verification after the collision risk assessment, high-collision-risk targets identified in the assessment can be further confirmed. For example, multiple time points can be set, such as the current time, the time corresponding to a certain distance traveled by the vehicle, the target motion prediction time, the 0.5 times TTC time, and the TTC time itself. This allows for the construction of the vehicle's corresponding OBB (Oriented Bounding Box) and the target's corresponding OBB, and collision detection is performed using the separating axis theorem. When the vehicle's OBB and the target's OBB overlap at any predicted time point, the corresponding high-collision-risk target is determined to have a collision risk.

[0032] S104, Based on the collision verification results of each high-collision-risk target, generate braking decision information for the current moment.

[0033] In this step, after obtaining the collision verification results for each high-collision-risk target, braking decision information for the current moment can be generated based on these results. In practical applications, a progressive TTC calculation method can be adopted, sequentially considering the relative motion relationship between the vehicle and the high-collision-risk target, the vehicle's own braking constraints, and the rear-end collision scenario after the high-collision-risk target stops early, to obtain a more accurate collision time. During the collision time calculation, a multi-level confidence degradation margin calculation mechanism can also be used to calculate the braking safety margin based on different data sources, and the confidence level of each calculation result can be marked. When the high-confidence calculation link is unavailable, it can automatically switch to the low-confidence calculation link to ensure that effective braking decisions can still be output even in the event of sensor obstruction or data anomalies. Based on the collision verification results, TTC calculation results, and safety margin results, arbitration is performed on the current braking stage to determine the corresponding braking decision information.

[0034] This application detects targets in driving environment data, first filtering high-collision-risk targets from candidate targets based on collision risk scores, then performing collision verification on these high-collision-risk targets, and generating braking decision information based on the collision verification results. This improves the accuracy of hazardous target identification, reduces misjudgments caused by perception fluctuations and multi-target interference, and ultimately improves the accuracy of braking decisions.

[0035] In one embodiment, a collision risk assessment is performed on each candidate target in the candidate target set to obtain a collision risk score for each candidate target, including: Based on the vehicle's current speed, determine the target speed range where the current speed is located from multiple speed ranges; Based on the target vehicle speed range, determine the target TTC threshold, the target distance threshold, and the target weight combination; Based on the target TTC threshold and the collision time TTC of each candidate target in the candidate target set, determine the TTC score of each candidate target in the candidate target set; Based on the target distance threshold and the distances between each candidate target in the candidate target set and the vehicle, a distance score is determined for each candidate target in the candidate target set. Predict the collision probability of each candidate target in the candidate target set to obtain a collision probability score for each candidate target in the candidate target set; For each candidate target in the candidate target set, the TTC score, distance score, and collision probability score of the candidate target are fused according to the TTC weight, distance weight, and collision probability weight in the target weight combination to obtain the collision risk score of the candidate target.

[0036] In this embodiment, reference Figure 2 The diagram illustrating the collision risk scoring process shows that multiple speed ranges can be pre-defined, and corresponding TTC thresholds, distance thresholds, and weight combinations can be configured for each speed range. For example, speed ranges can be divided into low-speed ranges (0~30km / h), medium-speed ranges (30~80km / h), and high-speed ranges (above 80km / h). Different speed ranges correspond to different TTC thresholds, distance thresholds, and weight combinations, with lower TTC and distance thresholds corresponding to lower speed ranges.

[0037] After obtaining the vehicle's current speed, the target speed range corresponding to the current speed can be determined, and the target TTC threshold, target distance threshold, and target weight combination corresponding to the target speed range can be further obtained.

[0038] For each candidate target in the candidate target set, the collision time (TTC) corresponding to that candidate target can be obtained, and combined with the target TTC threshold, the TTC score corresponding to the candidate target can be determined. Specifically, the TTC score is determined based on the ratio of the candidate target's TTC to the target TTC threshold; the smaller the TTC, the higher the TTC score; for example, TTC score = 1. (TTC / Target TTC Threshold), the TTC score is 0 when TTC is greater than or equal to the target TTC threshold, and 1 when TTC is 0. In this embodiment, the TTC score is used to characterize the collision risk level obtained from the collision time dimension.

[0039] The distance between the candidate target and the vehicle is obtained, and combined with a target distance threshold, a distance score is determined for the candidate target. Specifically, the distance score is determined based on the ratio of the distance between the candidate target and the vehicle to the target distance threshold; the smaller the distance, the higher the distance score. For example, distance score = 1. (Distance / Target Distance Threshold), the distance score is 0 when the distance is greater than or equal to the target distance threshold, and 1 when the distance is 0. In this embodiment, the distance score is used to characterize the collision risk level obtained from the distance dimension between the target and the vehicle.

[0040] The probability of a collision between a candidate target and a vehicle is predicted, and a collision probability score is obtained based on the prediction results. Specifically, at least one of the relative speed, relative distance, and lateral offset between the candidate target and the vehicle is input into a pre-trained collision probability prediction model to obtain the probability of a collision, which is then used as the collision probability score. Alternatively, the collision probability score can be obtained by looking up the relative speed, relative distance, and lateral offset between the candidate target and the vehicle using a lookup table model. In this embodiment, the collision probability score is used to characterize the likelihood of a collision between the candidate target and the vehicle.

[0041] Furthermore, after obtaining the TTC score, distance score, and collision probability score corresponding to the candidate target, these scores can be fused according to the TTC weight, distance weight, and collision probability weight in the target weight combination to obtain the collision risk score corresponding to the candidate target. Specifically, the collision risk score = TTC weight × TTC score + distance weight × distance score + collision probability weight × collision probability score, where the sum of the TTC weight, distance weight, and collision probability weight is 1. For example, the vehicle's operating state can be divided into low-speed, medium-speed, and high-speed ranges based on the vehicle's current speed. In the low-speed range, due to the lower vehicle speed, static obstacles have a greater impact on vehicle driving safety, so the weight corresponding to the distance dimension can be increased. For example, the weight combination corresponding to the low-speed range can be: TTC weight 0.15, distance weight 0.35, overlap weight 0.20, LDNEC weight 0.25, and collision probability weight 0.05. Within the medium-speed range, both the distance between the vehicle and the target, as well as the collision time, affect collision risk. Therefore, a balanced configuration of the distance and TTC dimensions can be achieved. For example, a weighting combination for the medium-speed range could be: TTC weight 0.25, distance weight 0.25, overlap weight 0.20, LDNEC weight 0.25, and collision probability weight 0.05. Within the high-speed range, due to the higher vehicle speed, dynamic targets may enter the collision zone in a shorter time, thus requiring a higher weighting for the TTC dimension. For example, a weighting combination for the high-speed range could be: TTC weight 0.35, distance weight 0.15, overlap weight 0.20, LDNEC weight 0.25, and collision probability weight 0.05. In this application, the above examples are only a portion of the examples shown; other weighting combinations are not limited here.

[0042] This embodiment determines the corresponding TTC threshold, distance threshold, and weight combination based on the vehicle's current speed, and fuses them with the TTC score, distance score, and collision probability score of the candidate target. This enables the collision risk assessment to adapt to the risk characteristics at different vehicle speeds and improves the accuracy of the collision risk score for the candidate target.

[0043] In one embodiment, collision verification is performed on each high-collision-risk target in the set of high-collision-risk targets, including: Based on the TTC of the high-collision-risk target, multiple future moments are determined, including: a first future moment, a second future moment, and a third future moment; the first future moment is the moment corresponding to the product of the TTC of the high-collision-risk target and a first preset ratio, where the first preset ratio is greater than 0 and less than 1; the second future moment is the moment corresponding to the product of the TTC of the high-collision-risk target and a second preset ratio, where the second preset ratio is greater than the first preset ratio and less than 1; the third future moment is the moment corresponding to the TTC of the high-collision-risk target; the time intervals between the first future moment, the second future moment, the third future moment, and the current moment increase sequentially; Determine the OBB of the vehicle and the high-collision-risk target at the current time, the first future time, the second future time, and the third future time, respectively; Based on whether there is overlap between the vehicle and the high-collision-risk target's OBB at the current time, the first future time, the second future time, and the third future time, it is determined whether the vehicle and the high-collision-risk target collide at the current time, the first future time, the second future time, and the third future time.

[0044] In this embodiment, refer to Figure 3 The collision confirmation geometry diagram shown illustrates that after determining the set of high-collision-risk targets, collision verification can be performed on each high-collision-risk target in the set, thereby further confirming the high-collision-risk targets obtained based on the collision risk score.

[0045] Specifically, multiple future moments can be determined based on the collision time (TTC) of the high-collision-risk target. These multiple future moments include a first future moment, a second future moment, and a third future moment. The first future moment is the moment corresponding to the product of the TTC of the high-collision-risk target and a first preset ratio; the second future moment is the moment corresponding to the product of the TTC of the high-collision-risk target and the second preset ratio; and the third future moment is the moment corresponding to the TTC of the high-collision-risk target.

[0046] In this system, the first preset ratio is greater than 0 and less than 1, and the second preset ratio is greater than the first preset ratio and less than 1. Therefore, the time intervals between the first future time, the second future time, and the third future time and the current time increase sequentially. For example, the first preset ratio can be 0.5, and the second preset ratio can be 0.8. In this case, collision verification can be performed at the current time, the first future time corresponding to 0.5 TTC, the second future time corresponding to 0.8 TTC, and the third future time corresponding to TTC, respectively, to determine whether there is spatial overlap between the vehicle and the high-collision-risk target at different time points.

[0047] Furthermore, the OBBs (On-Board Areas) of the vehicle and the high-collision-risk target at the current moment, the first future moment, the second future moment, and the third future moment can be determined separately. The OBB characterizes the spatial area occupied by the vehicle or the high-collision-risk target at the corresponding moment.

[0048] For the current moment, the OBB (On-Board Balance) for the vehicle and the high-collision-risk target can be determined based on their current position, attitude, and size information. For each future moment, the position and attitude of the vehicle and the high-collision-risk target at the corresponding future moment can be predicted based on their motion state, and the OBB for that future moment can be further determined.

[0049] For example, when determining the OBB of a vehicle at various future moments, the position and attitude of the vehicle at the first, second, and third future moments can be predicted based on the vehicle's current speed, direction of travel, and steering status. Similarly, when determining the OBB of a high-collision-risk target at various future moments, the position and attitude of the high-collision-risk target at various future moments can be predicted based on the target's current speed, direction of travel, and historical motion status.

[0050] After obtaining the OBBs of the vehicle and the high-collision-risk target at the current moment and at multiple future moments, it is possible to determine whether there is any overlap between the OBB of the vehicle and the OBB of the high-collision-risk target at each moment.

[0051] Specifically, if the two OBBs corresponding to the current moment overlap, it can be determined that the vehicle collided with the high-collision-risk target at the current moment; if the two OBBs corresponding to the current moment do not overlap, it can be determined that the vehicle did not collide with the high-collision-risk target at the current moment.

[0052] If the two OBBs at a first future time point overlap, it can be determined that the vehicle collided with the high-collision-risk target at the first future time point; if the two OBBs at a second future time point overlap, it can be determined that the vehicle collided with the high-collision-risk target at the second future time point; if the two OBBs at a third future time point overlap, it can be determined that the vehicle collided with the high-collision-risk target at the third future time point. If the two OBBs at any given time point do not overlap, it can be determined that the vehicle did not collide with the high-collision-risk target at that time point.

[0053] Based on this embodiment, collision verification results corresponding to the current time, the first future time, the second future time, and the third future time can be obtained respectively, and the collision relationship between the vehicle and the high collision risk target can be determined according to the collision verification results corresponding to each time.

[0054] In this embodiment, by setting multiple progressive future times based on the TTC of high collision risk targets, and performing OBB overlap verification on the vehicle and high collision risk targets at the current time and at multiple future times respectively, the spatial positional relationship between the vehicle and high collision risk targets within the predicted time range corresponding to the TTC can be verified. This can more accurately identify the collisions that may occur between the vehicle and high collision risk targets during future driving, thereby reducing the collision missed detection or misjudgment caused by judgment at a single time point and improving the accuracy of collision confirmation results.

[0055] In one embodiment, braking decision information for the current moment is generated based on the collision verification results of each high-collision-risk target, including: Based on the collision verification results of each high-collision-risk target, the braking phase of each high-collision-risk target is determined; From the braking phases of each high-collision-risk target, the highest priority target braking phase is determined. In this embodiment, the braking phase includes the AEB (Automatic Emergency Braking) phase, the AWB (Automatic Warning Braking) phase, and the automatic braking pre-charge phase. The priority of the AEB phase, AWB phase, and automatic braking pre-charge phase decreases sequentially. If the target braking phase is the AEB phase, braking decision information for performing automatic emergency braking operation at the current moment is generated; When the target braking stage is the AWB stage, braking decision information is generated to perform automatic alarm braking operation at the current moment; If the target braking phase is the automatic braking pre-charge phase, braking decision information is generated to perform the automatic braking pre-charge operation at the current moment.

[0056] In this embodiment, reference Figure 4 The schematic diagram of the braking decision generation process shown can determine the braking stage corresponding to each high collision risk target based on the collision verification results of each high collision risk target, and determine the target braking stage that needs to be executed at the current moment based on the priority relationship between different braking stages.

[0057] Specifically, the priorities corresponding to different braking stages can be preset. In this embodiment, the priorities of the Automatic Emergency Braking (AEB) stage, Automatic Warning Braking (AWB) stage, and Automatic Braking Pre-charge stage decrease sequentially. When different high-collision-risk targets correspond to different braking stages, the braking stage with the highest priority is selected as the target braking stage.

[0058] For example, when at least one of multiple high-collision-risk targets is identified as being in the AEB (Automatic Emergency Braking) stage, the AEB stage can be designated as the target braking stage, and braking decision information can be generated to instruct the vehicle to perform automatic emergency braking at the current moment. Even if other high-collision-risk targets are in the AWB (Automatic Emergency Braking) stage or the automatic braking pre-charge stage, the braking operation corresponding to the AEB stage can be prioritized.

[0059] When there are no targets in the AEB stage among multiple high-collision-risk targets, but at least one high-collision-risk target is in the AWB stage, the AWB stage can be identified as the target braking stage, and braking decision information can be generated to instruct the vehicle to perform automatic warning braking at the current moment.

[0060] When there are no targets in the AEB or AWB phase among multiple high-collision-risk targets, but at least one high-collision-risk target is in the automatic braking pre-charge phase, the automatic braking pre-charge phase can be identified as the target braking phase, and braking decision information can be generated to instruct the vehicle to perform automatic braking pre-charge operation at the current moment.

[0061] In this embodiment, when multiple high-collision-risk targets exist simultaneously, the braking demands corresponding to each high-collision-risk target can be uniformly arbitrated, avoiding the output of conflicting braking commands for different targets. Simultaneously, by prioritizing the braking phase with higher risk, the vehicle's braking response can be matched to the most urgent collision risk among the multiple targets, thereby improving the adaptability of braking decisions in multi-target scenarios.

[0062] In one embodiment, the high-collision-risk targets include a first high-collision-risk target, a second high-collision-risk target, and a third high-collision-risk target; based on the collision verification results of the high-collision-risk targets, determining the braking phase of each high-collision-risk target includes: If the collision verification result of the first high-collision-risk target is that the vehicle and the first high-collision-risk target collide at each of the current time and at least one future time, the braking phase of the first high-collision-risk target is determined to be the automatic emergency braking (AEB) phase. If the collision verification result of the second high collision risk target is: the vehicle and the second high collision risk target collide at both the second future time and the third future time, and no collision occurs at the current time and the first future time, then the braking phase of the second high collision risk target is determined to be the Automatic Warning Braking (AWB) phase. If the collision verification result of the third highest collision risk target is: the vehicle and the third highest collision risk target collide at the third future time, and no collision occurs at the current time, the first future time, and the second future time, then the braking phase of the third highest collision risk target is determined to be the automatic braking pre-charge phase.

[0063] Based on the above embodiments, in this embodiment, we continue to refer to... Figure 4 Based on the collision verification results of each high-collision-risk target at the current moment and at multiple future moments, the braking phase corresponding to each high-collision-risk target can be determined separately.

[0064] For the highest collision risk target, if the collision verification results indicate that a collision occurs between the vehicle and the target at the current moment and at least one future moment, it indicates that the collision risk corresponding to the highest collision risk target is highly urgent and requires immediate braking intervention. Therefore, the braking phase for the highest collision risk target can be determined as the Automatic Emergency Braking (AEB) phase.

[0065] For the second highest collision risk target, the collision verification results indicate that no collision occurred between the vehicle and the target at the current moment or the first future moment. However, collisions occurred at the second and third future moments, suggesting that no immediate collision has yet occurred. However, as the prediction time progresses, collisions between the vehicle and the target will continue to occur. Therefore, the braking phase for the second highest collision risk target can be determined as the Automatic Warning and Braking (AWB) phase, allowing for early braking intervention.

[0066] For the third highest collision risk target, the collision verification results show that no collision occurred between the vehicle and the target at the current time, the first future time, and the second future time, but only at the third future time. This indicates that the collision risk corresponding to the third highest collision risk target is relatively low, and there is still a certain time margin before a collision may occur. Therefore, the braking phase of the third highest collision risk target can be determined as the automatic braking pre-charge phase to establish a braking preparation state in advance.

[0067] Using the above method, the collision urgency of different high-collision-risk targets can be classified according to the collision state distribution of high-collision-risk targets at multiple times, and the corresponding AEB stage, AWB stage, or automatic braking pre-charge stage can be determined respectively. This embodiment can make more detailed distinctions between braking stages by utilizing the trend of collision risk changing over time, so that the degree of braking intervention matches the actual collision urgency, thereby reducing the problems of premature braking intervention or insufficient braking response, and improving the accuracy and rationality of braking decisions.

[0068] In one embodiment, after generating the braking decision information at the current moment, the method further includes: The first high-collision-risk target is locked as the old target; Based on the driving environment data at the next moment, determine the new target for the next moment in the Automatic Emergency Braking (AEB) phase; Based on the first distance between the new target and the vehicle at the next moment, and the second distance between the old target and the vehicle at the next moment, determine the distance difference between the first distance and the second distance; If the distance difference is not less than the lower limit threshold of the distance difference and not greater than the upper limit threshold of the distance difference, the first high-collision-risk target is kept as the target for performing automatic emergency braking at the next moment. If the distance difference is less than the lower limit threshold of the distance difference, the new target is determined as the target for performing automatic emergency braking operation at the next moment.

[0069] In this embodiment, after determining that the first high-collision-risk target is in the automatic emergency braking (AEB) stage and generating the first braking decision information for the first high-collision-risk target in the above embodiment, the target currently performing the AEB braking operation can be further locked to avoid frequent switching of the AEB braking target due to changes in the target detection results at continuous times.

[0070] Specifically, the highest collision risk target at the current moment can be identified as the old target, and the target identification information corresponding to the old target can be saved.

[0071] After entering the next moment, the target detection, collision risk assessment and braking phase judgment can be re-performed based on the driving environment data corresponding to the next moment. The target in the automatic emergency braking (AEB) phase can be identified from the multiple detected targets and identified as the new target.

[0072] Furthermore, the first distance between the new target and the vehicle at the next moment and the second distance between the old target and the vehicle at the next moment can be obtained respectively, and the distance difference can be determined based on the first distance and the second distance.

[0073] The distance difference can be used to characterize the change in the urgency of the new target relative to the old target. Specifically, the distance difference can be determined according to the following relationship: first distance minus second distance. After determining the distance difference, the relationship between the distance difference and a preset distance difference threshold can be used to determine whether it is necessary to switch the AEB execution target.

[0074] Specifically, if the distance difference is greater than or equal to the lower threshold and less than or equal to the upper threshold, it indicates that the distance change between the new and old targets is small, and the difference in collision risk between them is insufficient to trigger target switching. In this case, the target with the highest collision risk can be kept as the target for performing automatic emergency braking in the next moment.

[0075] If the distance difference is less than the lower limit threshold, it indicates that the new target is closer to the vehicle than the old target, and the collision risk of the new target is significantly higher than that of the old target. Therefore, the new target can be designated as the target for automatic emergency braking in the next moment, improving the timeliness of response to higher-risk targets.

[0076] In one embodiment, the collision time TTC of each candidate target in the candidate target set is determined according to the following steps: The initial TTC of the candidate target is determined based on the current vehicle speed, the current speed of each candidate target in the candidate target set, the relative distance between the vehicle and the candidate target, and the relative acceleration. Based on the current vehicle speed, determine the time for the vehicle to come to a complete stop; If the initial TTC of the candidate target is less than or equal to the complete braking time of the vehicle, the initial TTC is determined to be valid, and the initial TTC is determined as the final TTC of the candidate target. If the initial TTC of the candidate target is greater than the vehicle's complete braking time, the following steps are performed based on the candidate target's complete braking time: If the initial TTC of the candidate target is greater than the complete braking time of the candidate target, after the candidate target comes to a complete stop, the TTC of the candidate target is re-determined, and the re-determined TTC is determined as the final TTC of the candidate target. If the initial TTC of the candidate target is less than or equal to the complete braking time of the candidate target, after the vehicle comes to a complete stop, the TTC of the candidate target is re-determined, and the re-determined TTC is determined as the final TTC of the candidate target. The method further includes: If, after both the vehicle and the candidate target have come to a complete stop, there is still no collision between the vehicle and the candidate target, the final TTC of the candidate target is set to the no-collision risk indication value.

[0077] In this embodiment, the process of determining the collision time (TTC) for each candidate target in the candidate target set is described in detail, as follows: Figure 5 The flowchart shown is a progressive TTC calculation logic flowchart.

[0078] The system acquires the vehicle's current speed, the current speed of each candidate target in the candidate target set, the relative distance between the vehicle and the candidate targets, and the relative acceleration between the vehicle and the candidate targets. Based on these motion state parameters, it determines the initial Time To Collision (TTC) for each candidate target. The initial TTC characterizes the estimated time required for a collision between the vehicle and the candidate target, assuming both maintain their current states of motion. By calculating the initial TTC, a preliminary assessment can be made as to whether a collision risk exists between the vehicle and the candidate target.

[0079] Furthermore, the corresponding complete braking time of the vehicle is determined based on the vehicle's current speed. The complete braking time characterizes the time required for the vehicle to come to a complete stop from its current speed.

[0080] If the initial TTC of the candidate target is less than or equal to the vehicle's complete braking time, it indicates that the vehicle cannot come to a complete stop before the expected collision occurs, meaning the vehicle may still collide with the candidate target. Therefore, the current initial TTC can be determined to be valid, and this initial TTC is identified as the final TTC corresponding to the candidate target.

[0081] If the initial TTC of the candidate target is greater than the vehicle's complete braking time, it indicates that the vehicle can bring itself to a complete stop before the expected collision occurs. However, the vehicle's stopped state may alter the relative motion between the vehicle and the candidate target, necessitating a reassessment of the collision risk based on the candidate target's motion state. Specifically, the candidate target's complete braking time can be determined based on its current speed. The candidate target's complete braking time characterizes the time required for the candidate target to decelerate from its current speed to a complete stop.

[0082] Furthermore, if the initial TTC of the candidate target is greater than the candidate target's complete braking time, it indicates that the candidate target can stop before the expected collision occurs. In this case, the TTC between the vehicle and the candidate target can be re-determined after the candidate target has come to a complete stop, and the re-determined TTC can be used as the final TTC corresponding to the candidate target.

[0083] For example, when both the vehicle and the candidate target are able to stop in advance, their relative motion states will change. The original initial TTC cannot accurately reflect the collision risk after stopping, so the TTC needs to be recalculated based on the state of the candidate target after stopping.

[0084] If the initial TTC of the candidate target is less than or equal to the candidate target's complete braking time, it indicates that the candidate target cannot stop before the expected collision occurs, while the vehicle has already stopped in advance. In this case, after the vehicle has come to a complete stop, the TTC between the vehicle and the candidate target can be re-determined based on the vehicle's stopped state and the candidate target's continued movement state, and the re-determined TTC can be used as the final TTC for the candidate target.

[0085] Furthermore, if no collision occurs between the vehicle and the candidate target after both have come to a complete stop, it indicates that there is no actual collision risk under the current motion state. In this case, the final TTC corresponding to the candidate target can be set to the no-collision-risk indicator value, indicating that the candidate target does not need to trigger subsequent braking decisions.

[0086] In this embodiment, when determining the TTC of a candidate target, not only is the current relative motion state between the vehicle and the candidate target considered, but the initial TTC is also judged or re-determined by combining the vehicle's complete braking time and the candidate target's complete braking time, so that the final TTC can reflect the collision time after the braking state of the vehicle and the candidate target changes.

[0087] In one embodiment, target detection is performed on the driving environment data at the current moment to determine a set of candidate targets, including: Target detection is performed on the driving environment data at the current moment to obtain multiple initial targets, and the corresponding vehicle-mounted sensor for each initial target is determined. The vehicle-mounted sensors include: a forward-looking camera, a forward-facing radar, a line radar, and an angle radar. If the initial target corresponds to at least two vehicle-mounted sensors, the initial target is identified as a candidate target and added to the candidate target set; In the case where the initial target corresponds to a single vehicle-mounted sensor, it is determined whether the initial target meets the exemption conditions. The exemption conditions include at least one of the following: the type of the initial target is a preset target type, the ratio of the lateral velocity to the longitudinal velocity of the initial target is greater than a ratio threshold, the motion angle of the initial target is within the angle range, the longitudinal distance between the initial target and the vehicle is less than a distance threshold, and the lateral velocity of the initial target is greater than a velocity threshold. If the initial target satisfies the exemption conditions, the initial target is identified as a candidate target and added to the candidate target set; If the initial target does not meet the exemption conditions, the initial target is removed.

[0088] In this embodiment, target detection is performed on the current driving environment data to obtain multiple initial targets, and the corresponding vehicle-mounted sensors for each initial target are determined. These vehicle-mounted sensors may include a forward-facing camera, forward-facing radar, line radar, and corner radar.

[0089] For each initial target, the initial targets can be filtered based on the number of vehicle-mounted sensors that detect them. If an initial target corresponds to at least two vehicle-mounted sensors, it indicates that the initial target can be detected by multiple vehicle-mounted sensors, and this initial target can be identified as a candidate target and added to the candidate target set.

[0090] When the initial target corresponds to only a single vehicle-mounted sensor, it can be further determined whether the initial target meets the exemption conditions. The exemption conditions may include at least one of the following: the type of the initial target is a preset target type, the ratio of the lateral velocity to the longitudinal velocity of the initial target is greater than a ratio threshold, the motion angle of the initial target is within the angle range, the longitudinal distance between the initial target and the vehicle is less than a distance threshold, and the lateral velocity of the initial target is greater than a velocity threshold.

[0091] If the initial target meets the exemption conditions, it can be retained, identified as a candidate target, and added to the candidate target set. In this embodiment, for initial targets detected by only a single vehicle-mounted sensor, they are not directly eliminated, but rather their type, motion state, and positional relationship with the vehicle are further considered to determine whether they need to be retained.

[0092] If the initial target does not meet the exemption conditions, the initial target can be removed so that it does not enter the candidate target set.

[0093] In this embodiment, targets detected by multiple vehicle sensors are preferentially retained, while targets detected by a single vehicle sensor but meeting the exemption conditions are also retained. This reduces the possibility of targets with potential collision risks being directly eliminated while filtering targets detected by a single sensor, thereby further reducing the missed detection rate and improving the accuracy of collision risk assessment.

[0094] Based on the same inventive concept, a second aspect of the embodiments of this application provides a braking decision information generation system, such as... Figure 6 As shown, the system includes: The determination module 201 is used to perform target detection on the driving environment data at the current moment and determine the candidate target set; The evaluation module 202 is used to evaluate the collision risk of each candidate target in the candidate target set, obtain the collision risk score of each candidate target, and determine the high collision risk target set based on the collision risk scores of each candidate target. The verification module 203 is used to perform collision verification on each high collision risk target in the set of high collision risk targets; The generation module 204 is used to generate braking decision information at the current moment based on the collision verification results of each high-collision-risk target.

[0095] Optionally, a collision risk assessment is performed on each candidate target in the candidate target set to obtain a collision risk score for each candidate target. The assessment module 202 includes: The first determining submodule is used to determine the target speed range where the current vehicle speed is located from multiple speed ranges based on the current vehicle speed. The second determining submodule is used to determine the target TTC threshold, the target distance threshold, and the target weight combination based on the target vehicle speed range. The third determining submodule is used to determine the TTC score of each candidate target in the candidate target set based on the target TTC threshold and the collision time TTC of each candidate target in the candidate target set. The fourth determining submodule is used to determine the distance score of each candidate target in the candidate target set based on the target distance threshold and the distance between each candidate target in the candidate target set and the vehicle. The first prediction submodule is used to predict the collision probability of each candidate target in the candidate target set, so as to obtain the collision probability score of each candidate target in the candidate target set. The fusion submodule is used to fuse the TTC score, distance score and collision probability score of each candidate target in the candidate target set according to the TTC weight, distance weight and collision probability weight in the target weight combination, so as to obtain the collision risk score of the candidate target.

[0096] Optionally, collision verification is performed on each high-collision-risk target in the set of high-collision-risk targets, and the verification module 203 includes: The fifth determining submodule is used to determine multiple future times based on the TTC of the high-collision-risk target. These multiple future times include: a first future time, a second future time, and a third future time. The first future time is the time corresponding to the product of the TTC of the high-collision-risk target and a first preset ratio, where the first preset ratio is greater than 0 and less than 1. The second future time is the time corresponding to the product of the TTC of the high-collision-risk target and a second preset ratio, where the second preset ratio is greater than the first preset ratio and less than 1. The third future time is the time corresponding to the TTC of the high-collision-risk target. The time intervals between the first future time, the second future time, the third future time, and the current time increase sequentially. The sixth determination submodule is used to determine the OBB of the vehicle and the high-collision-risk target at the current time, the first future time, the second future time, and the third future time, respectively; The seventh determination submodule is used to determine whether the vehicle and the high-collision-risk target will collide at the current time, the first future time, the second future time, and the third future time, respectively, based on whether there is overlap between the OBB of the vehicle and the high-collision-risk target at the current time, the first future time, the second future time, and the third future time.

[0097] Optionally, based on the collision verification results of each high-collision-risk target, braking decision information for the current moment is generated. The generation module 204 includes: The eighth determining submodule is used to determine the braking phase of each high-collision-risk target based on the collision verification results of each high-collision-risk target; The ninth determining submodule is used to determine the highest priority target braking phase from the braking phases of each high collision risk target; wherein the priority of the automatic emergency braking (AEB) phase, the automatic warning braking (AWB) phase, and the automatic braking pre-charge phase decreases in that order. The first generation submodule is used to generate braking decision information for performing automatic emergency braking operation at the current moment when the target braking phase is the AEB phase. The second generation submodule is used to generate braking decision information for performing automatic alarm braking operation at the current moment when the target braking stage is the AWB stage. The third generation submodule is used to generate braking decision information for performing automatic braking pre-charging operation at the current moment when the target braking phase is the automatic braking pre-charging phase.

[0098] Optionally, each high-collision-risk target includes a first high-collision-risk target, a second high-collision-risk target, and a third high-collision-risk target; based on the collision verification results of each high-collision-risk target, the braking phase of each high-collision-risk target is determined, and the ninth determining submodule includes: The first determining subunit is used to determine the braking phase of the first high-collision-risk target as the automatic emergency braking (AEB) phase when the collision verification result of the first high-collision-risk target is: the vehicle and the first high-collision-risk target collide at each of the current time and at least one future time. The second determining subunit is used to determine the braking phase of the second high-collision-risk target as the Automatic Warning Braking (AWB) phase when the collision verification result of the second high-collision-risk target is: the vehicle and the second high-collision-risk target collide at both the second future time and the third future time, and no collision occurs at the current time and the first future time. The third determining subunit is used to determine the braking phase of the third high collision risk target as the automatic braking pre-charge phase if the collision verification result of the third high collision risk target is: the vehicle and the third high collision risk target collide at the third future time, and no collision occurs at the current time, the first future time, and the second future time.

[0099] Optionally, the system further includes: The locking submodule is used to lock the first high-collision-risk target as an old target; The tenth determination submodule is used to determine the new target in the automatic emergency braking (AEB) phase in the next moment based on the driving environment data of the next moment. The eleventh determination submodule is used to determine the distance difference between the first distance and the second distance based on the first distance between the new target and the vehicle at the next moment, and the second distance between the old target and the vehicle at the next moment; The holding submodule is used to maintain the first high-collision-risk target as the target for performing automatic emergency braking operation at the next moment, provided that the distance difference is not less than the lower limit threshold of the distance difference and not greater than the upper limit threshold of the distance difference. The twelfth determination submodule is used to determine the new target as the target for performing automatic emergency braking operation at the next moment when the distance difference is less than the lower limit threshold of the distance difference.

[0100] Optionally, the system further includes: The thirteenth determination submodule is used to determine the initial TTC of the candidate target based on the current vehicle speed, the current speed of each candidate target in the candidate target set, the relative distance between the vehicle and the candidate target, and the relative acceleration. The fourteenth determining submodule is used to determine the complete braking time of the vehicle based on the current vehicle speed; The fifteenth determination submodule is used to determine that the initial TTC is valid when the initial TTC of the candidate target is less than or equal to the complete braking time of the vehicle, and to determine the initial TTC as the final TTC of the candidate target. The sixteenth determination submodule is used to redetermine the TTC of the candidate target after the candidate target has come to a complete stop, when the initial TTC of the candidate target is greater than the complete stopping time of the candidate target, and to determine the redetermined TTC as the final TTC of the candidate target. The seventeenth determination submodule is used to redetermine the TTC of the candidate target after the vehicle comes to a complete stop when the initial TTC of the candidate target is less than or equal to the complete braking time of the candidate target, and to determine the redetermined TTC as the final TTC of the candidate target. The configuration submodule is used to set the final TTC of the candidate target to a no-collision risk indicator value when the vehicle and the candidate target have come to a complete stop and the vehicle and the candidate target have not yet collided.

[0101] Optionally, target detection is performed on the driving environment data at the current moment to determine a set of candidate targets. The determining module 201 includes: The target detection submodule is used to detect targets in the current driving environment data, obtain multiple initial targets, and determine the corresponding vehicle sensor for each initial target. The vehicle sensors include: a forward-looking camera, a forward-facing radar, a line radar, and an angle radar. The first addition submodule is used to determine the initial target as a candidate target and add it to the candidate target set when the initial target corresponds to at least two vehicle-mounted sensors; The eighteenth determination submodule is used to determine whether the initial target meets the exemption conditions when the initial target corresponds to a single vehicle sensor. The exemption conditions include at least one of the following: the type of the initial target is a preset target type, the ratio of the lateral velocity to the longitudinal velocity of the initial target is greater than a ratio threshold, the motion angle of the initial target is within the angle range, the longitudinal distance between the initial target and the vehicle is less than a distance threshold, and the lateral velocity of the initial target is greater than a velocity threshold. The second addition submodule is used to determine the initial target as a candidate target and add it to the candidate target set if the initial target meets the exemption conditions; The elimination submodule is used to eliminate the initial target if the initial target does not meet the exemption conditions.

[0102] Based on the same inventive concept, a third aspect of the embodiments of this application provides a method as follows: Figure 7 The electronic device 100 shown includes a processor 120, a memory 110, and a program or instructions stored in the memory 110 and executable on the processor 120. When the program or instructions are executed by the processor 120, they implement the steps of the braking decision information generation method as described in the first aspect of this application.

[0103] Based on the same inventive concept, in a fourth aspect of this application, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the braking decision information generation method as described in the first aspect of this application.

[0104] Each embodiment in this specification focuses on the differences from other embodiments. For the same or similar parts between the embodiments, please refer to each other.

[0105] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, embodiments of this application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of this application can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0106] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0107] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0108] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0109] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.

[0110] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0111] The above provides a detailed description of the braking decision information generation method, system, electronic device, and storage medium. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for generating braking decision information, characterized in that, The method includes: Target detection is performed on the current driving environment data to determine a set of candidate targets; Each candidate target in the candidate target set is assessed for collision risk to obtain a collision risk score for each candidate target. Based on the collision risk scores of each candidate target, a set of high-collision-risk targets is determined. Collision verification is performed on each high-collision-risk target in the set of high-collision-risk targets; Based on the collision verification results of each high-collision-risk target, braking decision information for the current moment is generated.

2. The braking decision information generation method according to claim 1, characterized in that, Each candidate target in the candidate target set is subjected to a collision risk assessment to obtain a collision risk score for each candidate target, including: Based on the vehicle's current speed, determine the target speed range where the current speed is located from multiple speed ranges; Based on the target vehicle speed range, determine the target TTC threshold, the target distance threshold, and the target weight combination; Based on the target TTC threshold and the collision time TTC of each candidate target in the candidate target set, determine the TTC score of each candidate target in the candidate target set; Based on the target distance threshold and the distances between each candidate target in the candidate target set and the vehicle, a distance score is determined for each candidate target in the candidate target set. Predict the collision probability of each candidate target in the candidate target set to obtain a collision probability score for each candidate target in the candidate target set; For each candidate target in the candidate target set, the TTC score, distance score, and collision probability score of the candidate target are fused according to the TTC weight, distance weight, and collision probability weight in the target weight combination to obtain the collision risk score of the candidate target.

3. The braking decision information generation method according to claim 1, characterized in that, Collision verification is performed on each high-collision-risk target in the set of high-collision-risk targets, including: Based on the TTC of the high-collision-risk target, multiple future moments are determined, including: a first future moment, a second future moment, and a third future moment; the first future moment is the moment corresponding to the product of the TTC of the high-collision-risk target and a first preset ratio, where the first preset ratio is greater than 0 and less than 1; the second future moment is the moment corresponding to the product of the TTC of the high-collision-risk target and a second preset ratio, where the second preset ratio is greater than the first preset ratio and less than 1; the third future moment is the moment corresponding to the TTC of the high-collision-risk target; the time intervals between the first future moment, the second future moment, the third future moment, and the current moment increase sequentially; Determine the OBB of the vehicle and the high-collision-risk target at the current time, the first future time, the second future time, and the third future time, respectively; Based on whether there is overlap between the vehicle and the high-collision-risk target's OBB at the current time, the first future time, the second future time, and the third future time, it is determined whether the vehicle and the high-collision-risk target collide at the current time, the first future time, the second future time, and the third future time.

4. The braking decision information generation method according to claim 3, characterized in that, Based on the collision verification results of each high-collision-risk target, braking decision information for the current moment is generated, including: Based on the collision verification results of each high-collision-risk target, the braking phase of each high-collision-risk target is determined; From the braking phases of the various high-collision-risk targets, the highest priority target braking phase is determined; among them, the priority of the Automatic Emergency Braking (AEB) phase, Automatic Warning Braking (AWB) phase, and Automatic Braking Pre-charge phase decreases in that order. If the target braking phase is the AEB phase, braking decision information for performing automatic emergency braking operation at the current moment is generated; When the target braking stage is the AWB stage, braking decision information is generated to perform automatic alarm braking operation at the current moment; If the target braking phase is the automatic braking pre-charge phase, braking decision information is generated to perform the automatic braking pre-charge operation at the current moment.

5. The braking decision information generation method according to claim 4, characterized in that, The high-collision-risk targets include a first high-collision-risk target, a second high-collision-risk target, and a third high-collision-risk target; Based on the collision verification results of each high-collision-risk target, the braking phase of each high-collision-risk target is determined, including: If the collision verification result of the first high-collision-risk target is that the vehicle and the first high-collision-risk target collide at each of the current time and at least one future time, the braking phase of the first high-collision-risk target is determined to be the automatic emergency braking (AEB) phase. If the collision verification result of the second high collision risk target is: the vehicle and the second high collision risk target collide at both the second future time and the third future time, and no collision occurs at the current time and the first future time, then the braking phase of the second high collision risk target is determined to be the Automatic Warning Braking (AWB) phase. If the collision verification result of the third highest collision risk target is: the vehicle and the third highest collision risk target collide at the third future time, and no collision occurs at the current time, the first future time, and the second future time, then the braking phase of the third highest collision risk target is determined to be the automatic braking pre-charge phase.

6. The braking decision information generation method according to claim 5, characterized in that, After generating the braking decision information for the current moment, it also includes: The first high-collision-risk target is locked as the old target; Based on the driving environment data at the next moment, determine the new target for the next moment in the Automatic Emergency Braking (AEB) phase; Based on the first distance between the new target and the vehicle at the next moment, and the second distance between the old target and the vehicle at the next moment, determine the distance difference between the first distance and the second distance; If the distance difference is not less than the lower limit threshold of the distance difference and not greater than the upper limit threshold of the distance difference, the first high-collision-risk target is kept as the target for performing automatic emergency braking at the next moment. If the distance difference is less than the lower limit threshold of the distance difference, the new target is determined as the target for performing automatic emergency braking operation at the next moment.

7. The braking decision information generation method according to claim 2, characterized in that, The collision time (TTC) of each candidate target in the candidate target set is determined according to the following steps: The initial TTC of the candidate target is determined based on the current vehicle speed, the current speed of each candidate target in the candidate target set, the relative distance between the vehicle and the candidate target, and the relative acceleration. Based on the current vehicle speed, determine the time for the vehicle to come to a complete stop; If the initial TTC of the candidate target is less than or equal to the complete braking time of the vehicle, the initial TTC is determined to be valid, and the initial TTC is determined as the final TTC of the candidate target. If the initial TTC of the candidate target is greater than the vehicle's complete braking time, the following steps are performed based on the candidate target's complete braking time: If the initial TTC of the candidate target is greater than the complete braking time of the candidate target, after the candidate target comes to a complete stop, the TTC of the candidate target is re-determined, and the re-determined TTC is determined as the final TTC of the candidate target. If the initial TTC of the candidate target is less than or equal to the complete braking time of the candidate target, after the vehicle comes to a complete stop, the TTC of the candidate target is re-determined, and the re-determined TTC is determined as the final TTC of the candidate target. The method further includes: If, after both the vehicle and the candidate target have come to a complete stop, there is still no collision between the vehicle and the candidate target, the final TTC of the candidate target is set to the no-collision risk indication value.

8. The braking decision information generation method according to any one of claims 1-7, characterized in that, Target detection is performed on the current driving environment data to determine a set of candidate targets, including: Target detection is performed on the driving environment data at the current moment to obtain multiple initial targets, and the corresponding vehicle-mounted sensor for each initial target is determined. The vehicle-mounted sensors include: a forward-looking camera, a forward-facing radar, a line radar, and an angle radar. If the initial target corresponds to at least two vehicle-mounted sensors, the initial target is identified as a candidate target and added to the candidate target set; In the case where the initial target corresponds to a single vehicle-mounted sensor, it is determined whether the initial target meets the exemption conditions. The exemption conditions include at least one of the following: the type of the initial target is a preset target type, the ratio of the lateral velocity to the longitudinal velocity of the initial target is greater than a ratio threshold, the motion angle of the initial target is within the angle range, the longitudinal distance between the initial target and the vehicle is less than a distance threshold, and the lateral velocity of the initial target is greater than a velocity threshold. If the initial target satisfies the exemption conditions, the initial target is identified as a candidate target and added to the candidate target set; If the initial target does not meet the exemption conditions, the initial target is removed.

9. A braking decision information generation system, characterized in that, The system includes: The determination module is used to perform target detection on the driving environment data at the current moment and determine the set of candidate targets; The evaluation module is used to evaluate the collision risk of each candidate target in the candidate target set, obtain the collision risk score of each candidate target, and determine the set of high collision risk targets based on the collision risk scores of each candidate target. The verification module is used to perform collision verification on each high-collision-risk target in the set of high-collision-risk targets; The generation module is used to generate braking decision information for the current moment based on the collision verification results of each high-collision-risk target.

10. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein when the program or instructions are executed by the processor, they implement the steps of the braking decision information generation method as described in any one of claims 1-8.

11. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the braking decision information generation method as described in any one of claims 1-8.

Images