Vehicle collision prediction and avoidance method, controller and vehicle

Abstract

The invention provides a vehicle collision prediction and avoidance method, a controller and a vehicle, and relates to the technical field of vehicle control. The method comprises the following steps: acquiring driving characteristic data of vehicles around a self-vehicle when the self-vehicle runs; when the surrounding vehicle has an emergency braking condition, according to the driving characteristic data and the traffic environment data of the surrounding vehicle, obtaining a prediction result of collision between the surrounding vehicle and the vehicle; and determining an evasion strategy based on the prediction result, and controlling the vehicle based on the evasion strategy. According to the method and the device, prospective prediction can be carried out on whether the surrounding vehicles and the self-vehicle have sudden collision or not to recognize the collision risk in advance, so that the corresponding avoidance strategy is determined to carry out avoidance control on the self-vehicle according to the obtained prediction result, the collision risk is effectively avoided, the collision accident rate is reduced, and the collision safety is improved. And life and property safety of drivers and passengers is protected.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and in particular to a vehicle collision prediction and avoidance method, controller and vehicle. Background Technology

[0002] With the continuous increase in car ownership, road traffic safety pressures are intensifying, and the incidence of vehicle collision accidents is constantly rising. During operation, vehicles face the risk of colliding with surrounding vehicles, such as colliding with the vehicle in front or being rear-ended.

[0003] In related technologies, it is impossible to effectively predict sudden collisions between a vehicle and surrounding vehicles, and it is also impossible to control the vehicle to make effective avoidance actions in a short period of time when the risk of collision increases sharply. This cannot effectively reduce the occurrence of collision accidents and endanger the lives and property of drivers and passengers. Summary of the Invention

[0004] This application provides a vehicle collision prediction and avoidance method, controller, and vehicle to solve the technical problem in related technologies that cannot effectively reduce the occurrence of collision accidents, endangering the life and property safety of drivers and passengers.

[0005] In a first aspect, embodiments of this application provide a vehicle collision prediction and avoidance method, including: While the vehicle is in motion, acquire driving characteristic data of vehicles around the vehicle; When surrounding vehicles brake suddenly, the vehicle is predicted to collide with the surrounding vehicles based on the driving characteristic data of the surrounding vehicles and traffic environment data. Based on the prediction results, an avoidance strategy is determined, and the vehicle is controlled based on the avoidance strategy.

[0006] In this embodiment, during the driving process of the vehicle, driving characteristic data of surrounding vehicles are acquired. When a surrounding vehicle brakes suddenly, it indicates that a sudden change in the driving state of a vehicle around the vehicle has occurred, and an accident may have already occurred or there is a potential collision risk. In order to reduce the occurrence of collision accidents, based on the driving characteristic data used to characterize the relative motion relationship between surrounding vehicles and the vehicle, the motion state of surrounding vehicles, and the traffic environment data used to provide scenario constraints, a forward-looking prediction is made on whether a sudden collision will occur between surrounding vehicles and the vehicle. This allows for early identification of collision risks, and based on the obtained prediction results, a corresponding avoidance strategy is determined to avoid collision risks, reduce the incidence of collision accidents, and protect the lives and property of drivers and passengers.

[0007] In one possible implementation, a forward acoustic array is provided at the front end of the vehicle, and a rearward acoustic array is provided at the rear end of the vehicle. The acquisition of driving characteristic data of vehicles surrounding the vehicle includes: The system controls the forward acoustic array to emit forward acoustic signals around the vehicle and receives the forward echo signals returned after the forward acoustic signals encounter surrounding vehicles. The rear acoustic array is controlled to emit rear acoustic signals around the vehicle, and the rear echo signals returned after the rear acoustic signals encounter surrounding vehicles are received. Based on the forward echo signal and the backward acoustic signal, the driving characteristic data of the surrounding vehicles are determined.

[0008] In this embodiment, by setting forward acoustic arrays and rear acoustic arrays at the front and rear of the vehicle respectively, the emitted acoustic signals can cover a wide area in front of and behind the vehicle, thereby reducing blind spots and enabling wide-range detection of the area around the vehicle. This allows for comprehensive and accurate detection of vehicles around the vehicle, and thus accurately obtains the relative motion relationship between the surrounding vehicles and the vehicle, as well as the driving status of the surrounding vehicles.

[0009] In one possible implementation, controlling the forward acoustic array to emit forward acoustic signals around the vehicle includes: Acquire noise signals around the vehicle; The characteristic frequency range of the forward acoustic signal is determined based on the frequency range of the noise signal and the preset frequency range. The forward acoustic array is controlled to emit forward acoustic signals within the characteristic frequency range around the vehicle.

[0010] Here, by adaptively adjusting the frequency range of the acoustic signal based on the noise signal around the vehicle, the impact of environmental noise on acoustic detection can be reduced, and the detection accuracy of vehicles around the vehicle can be improved, so that the driving characteristic data of the surrounding vehicles can be accurately obtained in the future.

[0011] In one possible implementation, before obtaining a prediction of a collision between the vehicle and surrounding vehicles based on the driving characteristic data and traffic environment data of the surrounding vehicles in the event of emergency braking by surrounding vehicles, the method further includes: For each vehicle in the surrounding area, determine whether there is deceleration in the vehicle's driving characteristic data within a preset time period. If deceleration exists, the average deceleration value is obtained based on the deceleration value, and it is determined whether the average deceleration value is greater than a preset braking threshold. If the average deceleration is greater than the preset braking threshold, it is determined that the vehicle has undergone emergency braking.

[0012] Specifically, for each vehicle in the surrounding area, emergency braking is determined based on multiple accelerations within a preset time period. This avoids the problem of misjudgment caused by relying on single-point deceleration, which is susceptible to interference from road bumps or other factors. By analyzing the relationship between the average deceleration and a preset braking threshold, it ensures that deceleration behaviors with large amplitude and strong continuity can be effectively filtered out. This allows for accurate differentiation between normal deceleration and emergency braking, accurately determining whether any vehicle in the surrounding area is undergoing emergency braking. This provides a valid basis for subsequent predictions of whether a collision between the vehicle and surrounding vehicles is possible.

[0013] In one possible implementation, obtaining the prediction result of a collision between the vehicle and surrounding vehicles based on the driving characteristic data of surrounding vehicles and traffic environment data includes: The driving characteristic data of the surrounding vehicles and the traffic environment data are input into the collision prediction model to obtain the prediction results of the collision prediction model for each of the surrounding vehicles colliding with the vehicle itself. The driving characteristic data includes the distance, relative speed, and relative orientation between the vehicle and surrounding vehicles, as well as the acceleration of surrounding vehicles. The traffic environment data includes the collision location, collision speed, traffic density, road surface condition, and weather. The collision location is the relative position of the vehicle and the surrounding vehicles that have already collided. The collision speed is the relative speed between the vehicle and the surrounding vehicles that have already collided before the collision.

[0014] In this embodiment, considering the chain reaction effect in traffic accidents, the collision location and time related to vehicles that have already collided with surrounding vehicles are used as traffic environment data. This improves the ability to predict collisions caused by a chain reaction of surrounding vehicles. Simultaneously, driving feature data, which characterizes the relative motion relationship between the vehicle and surrounding vehicles, as well as the motion state of surrounding vehicles, works synergistically with traffic environment data to provide scenario constraints. This provides complete input support for the collision prediction model, capturing the relative motion trend between the vehicle and surrounding vehicles through driving feature data and supplementing scenario constraints with traffic environment data, ultimately achieving accurate prediction of collisions between the vehicle and surrounding vehicles.

[0015] In one possible implementation, determining the avoidance strategy based on the prediction result includes: Obtain the obstacle avoidance space status of the adjacent lanes of the lane where the vehicle is located; the obstacle avoidance space status includes whether there is obstacle avoidance space and whether there is no obstacle avoidance space. Based on the prediction results and the avoidance space status, the avoidance strategy is determined.

[0016] Here, the collision probability, collision time and collision severity in the prediction results are considered simultaneously, as well as the avoidance space status of adjacent lanes. This achieves a precise coupling of risk prediction and avoidance feasibility, ensuring both the safety, stability and scenario adaptability of the final avoidance strategy, and guaranteeing the feasibility of the avoidance strategy in practice, thus avoiding ineffective decisions.

[0017] In one possible implementation, the prediction results include collision probability, collision time, and collision severity; Determining the avoidance strategy based on the prediction result and the avoidance space state includes: When there are multiple prediction results, the prediction result with the highest risk level is determined from all the prediction results; Based on the collision probability, collision time, and collision severity in the prediction result with the highest risk level, as well as preset probability thresholds, preset time thresholds, preset severity thresholds, and the avoidance space state, the optimal avoidance strategy is determined from the preset avoidance strategies, and the optimal avoidance strategy is used as the avoidance strategy.

[0018] In this embodiment, the optimal avoidance strategy is obtained based on the prediction result with the highest risk level, prioritizing driving safety in extremely dangerous scenarios. Furthermore, setting preset thresholds quantifies collision risk from multiple dimensions, avoiding the limitations of single-indicator decision-making. Considering the avoidance space state further ensures the feasibility of the final avoidance strategy. In other words, both collision risk level and avoidance space state are considered simultaneously to obtain the most suitable and feasible avoidance strategy.

[0019] In one possible implementation, the preset avoidance strategy includes braking control, lane changing, and vehicle speed adjustment; The step of determining the optimal avoidance strategy from preset avoidance strategies based on the collision probability, collision time, and collision severity in the prediction result with the highest risk level, as well as preset probability thresholds, preset time thresholds, preset severity thresholds, and the avoidance space state, includes: When the vehicle corresponding to the highest risk level prediction result is the vehicle in front of the vehicle, if the collision probability is greater than or equal to the preset probability threshold, or the collision time is less than or equal to the preset time threshold, then the optimal avoidance strategy is determined to be braking control. If the collision probability is less than the preset probability threshold, the collision time is greater than the preset time threshold, and the avoidance space status is no avoidance space, then the optimal avoidance strategy is determined to be speed adjustment or braking control. If the collision probability is less than a preset probability threshold, the collision time is greater than a preset time threshold, the collision severity is less than a preset severity threshold, and the avoidance space status is "there is avoidance space", then the optimal avoidance strategy is determined to be lane change or speed adjustment.

[0020] Secondly, embodiments of this application provide a vehicle collision prediction and avoidance device, including: The acquisition module is used to acquire driving characteristic data of vehicles around the vehicle while the vehicle is in motion.

[0021] The module is used to obtain a prediction result of a collision between the vehicle and surrounding vehicles based on the driving characteristic data of the surrounding vehicles and traffic environment data when surrounding vehicles brake suddenly.

[0022] The control module is used to determine an avoidance strategy based on the prediction results, and to control the vehicle based on the avoidance strategy.

[0023] Thirdly, embodiments of this application provide a controller, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the vehicle collision prediction and avoidance method as described in any of the first aspects.

[0024] Fourthly, embodiments of this application provide a vehicle including the controller described in the third aspect.

[0025] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the vehicle collision prediction and avoidance method as described in any of the first aspects.

[0026] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0027] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this specification. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art 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.

[0029] Figure 1 This is a schematic diagram of an application scenario provided by an embodiment of this application; Figure 2 This is a schematic flowchart of a vehicle collision prediction and avoidance method provided in an embodiment of this application; Figure 3 This is a flowchart illustrating a vehicle collision prediction and avoidance method provided in another embodiment of this application; Figure 4 This is a schematic diagram of the structure of a vehicle collision prediction and avoidance device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the controller provided in one embodiment of this application. Detailed Implementation

[0030] The present application will be described more clearly below with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the function of the present application, but do not limit the present application in any way. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application. These all fall within the protection scope of the present application.

[0031] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0032] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0033] In the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0034] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0035] Furthermore, the term "multiple" mentioned in the embodiments of this application should be interpreted as two or more.

[0036] When a vehicle is in motion, especially in road environments with high speeds and heavy traffic, it faces the risk of colliding with surrounding vehicles, such as colliding with the vehicle in front or being rear-ended.

[0037] Current technologies primarily provide warnings for known hazards, but cannot effectively predict sudden collisions between the vehicle and surrounding vehicles. Regarding collision avoidance, they rely mainly on the driver's observation of surrounding events, making it difficult to control the vehicle to take effective evasive action in a short period when the risk of a collision suddenly increases. Consequently, they cannot effectively reduce the occurrence of collisions, endangering the lives and property of passengers.

[0038] Based on the idea of ​​proactively predicting the collision risk between the vehicle and surrounding vehicles and taking timely and effective avoidance measures to reduce the collision accident rate, the embodiments of this application acquire driving characteristic data of surrounding vehicles during the vehicle's operation. When a surrounding vehicle brakes suddenly, it indicates that a sudden change in the driving state of a vehicle around the vehicle has occurred, suggesting that an accident may have already occurred or there may be a potential collision risk. To reduce the occurrence of collision accidents, based on the driving characteristic data used to characterize the relative motion relationship between the surrounding vehicles and the vehicle, the motion state of the surrounding vehicles, and the traffic environment data used to provide scenario constraints, a proactive prediction is made as to whether a sudden collision will occur between the surrounding vehicles and the vehicle. This allows for early identification of collision risks, and based on the obtained prediction results, a corresponding avoidance strategy is determined to control the vehicle to effectively avoid collision risks and reduce the collision accident rate.

[0039] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments will be described below in conjunction with the accompanying drawings.

[0040] First refer to Figure 1 , Figure 1 A schematic diagram illustrating an application scenario provided according to an embodiment of this application is shown. This application scenario includes a vehicle B1 and surrounding vehicles B2. In one scenario, the vehicle is equipped with an acoustic system for emitting acoustic signals to its surroundings. The acoustic system may include a forward acoustic array D1 and a rearward acoustic array D2, wherein the forward acoustic array is located at the front end of the vehicle, and the rearward acoustic array is located at the rear end of the vehicle.

[0041] During the vehicle's operation, the controller controls the acoustic system to emit acoustic signals around the vehicle. When the acoustic signals encounter surrounding vehicles, they are reflected back, returning corresponding echo signals. Based on the received echo signals, the controller determines the driving characteristic data of the surrounding vehicles. When a surrounding vehicle brakes suddenly, the controller determines the predicted collision result between the surrounding vehicle and the vehicle based on the aforementioned driving characteristic data and traffic environment parameters. Based on the predicted result, the controller determines an avoidance strategy to control the vehicle.

[0042] Optionally, in other scenarios, the vehicle can be equipped with radar, cameras, etc. During the vehicle's operation, the controller controls the radar or camera to obtain relevant information about the vehicles around the vehicle, such as the reflected signals from the radar after encountering the surrounding vehicles, or the continuous images of the surrounding vehicles captured by the camera. Then, the controller obtains the driving characteristic data of the surrounding vehicles based on the above relevant information.

[0043] Optionally, the controller included in the vehicle can be an on-board controller, an external controller, or a combination of both. It is a hardware device with data storage, processing, and analysis functions. Taking an on-board controller as an example, it can be an Electronic Control Unit (ECU), etc.

[0044] The following is combined with Figure 1 Application scenarios, refer to Figures 2-3 This application describes a vehicle collision prediction and avoidance method provided according to exemplary embodiments. It should be noted that the above application scenarios are shown only to facilitate understanding of the spirit and principles of this application, and the embodiments of this application are not limited in any way. Rather, the embodiments of this application can be applied to any applicable scenario.

[0045] refer to Figure 2 , Figure 2 This is a schematic flowchart of a vehicle collision prediction and avoidance method provided in an embodiment of this application. Figure 2 As shown, the method in the embodiments of this application may include: Step 201: While the vehicle is in motion, acquire the driving characteristic data of the vehicles around the vehicle.

[0046] Optionally, driving characteristic data may include the distance, relative speed, and relative orientation between the vehicle and surrounding vehicles, as well as the acceleration of surrounding vehicles, etc., to characterize the relative motion relationship between the vehicle and surrounding vehicles, and the motion state of surrounding vehicles. Therefore, it can provide effective data support for determining whether there is a risk of collision between the vehicle and surrounding vehicles.

[0047] For example, relevant information about vehicles around the vehicle can be obtained through cameras, radar such as lidar, acoustic systems, etc., and then driving characteristic data of the surrounding vehicles can be obtained based on the above-mentioned relevant information. This embodiment uses an acoustic system as an example for illustration.

[0048] The vehicle's acoustic system emits acoustic signals, which can be frequency-modulated continuous wave (FMCW) signals, around the vehicle. These acoustic signals are reflected when they encounter obstacles around the vehicle, such as other vehicles, returning corresponding echo signals.

[0049] Based on the frequency difference between the received echo signal and the acoustic signal corresponding to the time of reception of the echo signal, the driving characteristic data of the surrounding vehicles can be obtained. In a real-world scenario, there can be multiple vehicles surrounding the vehicle; correspondingly, based on the echo signal of each of the surrounding vehicles, the driving characteristic data of that vehicle can be obtained.

[0050] Step 202: When surrounding vehicles brake suddenly, based on the driving characteristic data of surrounding vehicles and traffic environment data, obtain the prediction result of a collision between the surrounding vehicles and the vehicle.

[0051] When a vehicle around your car brakes suddenly, indicating a sudden change in its driving behavior, it suggests that an accident may have occurred nearby or there is a potential collision risk. To reduce the likelihood of a collision with your vehicle, you can proactively predict whether a collision with surrounding vehicles will occur, and obtain corresponding prediction results.

[0052] Here, traffic environment data can include collision location, collision speed, traffic density, road conditions, and weather. The collision location is the relative position of the vehicle to the surrounding vehicles that have already collided, and can be represented by their relative distance and orientation, directly reflecting the positional relationship between the hazard and the vehicle. The collision speed is the relative speed between the vehicle and the surrounding vehicles before the collision, reflecting the severity of the collision.

[0053] Considering the chain reaction effect in traffic accidents, when there are vehicles around a vehicle that have already collided, it is very easy to trigger a secondary or tertiary chain collision. Therefore, using the collision location and collision speed as scenario constraints can improve the ability to predict collisions caused by chain collisions of surrounding vehicles.

[0054] Traffic density refers to the number of vehicles in the surrounding area. Under the same motion parameters, the higher the traffic density, the lower the feasibility of the vehicle's avoidance maneuver. Road surface condition includes the smoothness of the road surface on the current driving segment, while weather conditions such as rain and snow will affect the road surface adhesion. Traffic density, road surface condition, and weather are related to the actual driving environment of the vehicle and are used to correct the prediction results, making the prediction results more consistent with real driving scenarios.

[0055] It should be noted that if there are no vehicles in the surrounding area that have already collided, the collision location and collision speed in the traffic environment data can be equivalently replaced by the relative position and relative speed of the vehicle closest to the vehicle that has undergone emergency braking, in order to ensure the effectiveness and accuracy of the prediction results.

[0056] For example, the collision location and speed can be determined using the vehicle's cameras and radar sensors. For instance, images of surrounding vehicles can be acquired via cameras, and the images can be identified using a pre-defined target detection model. When a collision is detected, a visual stereo ranging method can be used to obtain the relative distance and orientation between the collided vehicles and the vehicle. Simultaneously, the driving characteristic data of the surrounding vehicles can be linked to determine the collided vehicles based on the collision location, and the relative speeds of the collided vehicles can be used as the collision speed. Of course, other methods can also be used to obtain the collision location and speed; no specific limitations are imposed here.

[0057] For road surface conditions such as road smoothness, the vehicle's onboard LiDAR can collect real-time data on the vertical distance between the road surface and the vehicle's chassis. Based on the amplitude and frequency of fluctuations in this vertical distance, the road smoothness level can be quantified, such as categorized as smooth, slightly bumpy, moderately bumpy, and heavily bumpy. Traffic density is determined directly based on the number of surrounding vehicles detected using echo signals. For weather, real-time weather data can be received from the cloud via vehicle-to-everything (V2X) communication, providing information on the weather conditions in the vehicle's location, such as sunny, rainy, or snowy. Of course, other methods can also be used to obtain weather information; no specific restrictions are placed here.

[0058] In some embodiments, a collision prediction model can be used to obtain the prediction results of collisions between the vehicle and surrounding vehicles. The input of the collision prediction model includes driving characteristic data of surrounding vehicles and traffic environment data, and the output is the prediction result of collisions between the vehicle and each of the surrounding vehicles.

[0059] The prediction results can include collision probability, collision time, and collision severity. Collision probability refers to the probability that the vehicle will collide with other vehicles in the surrounding area. Collision time refers to the remaining time from the current time until the collision occurs. Collision severity refers to the severity of injury to the vehicle and its occupants caused by the collision. In a rear-end collision scenario, this can be characterized by the relative speed difference between the vehicle and other vehicles in the surrounding area, specifically the difference between the rear vehicle's speed and the front vehicle's speed. The greater the relative speed difference, the higher the collision severity.

[0060] As mentioned above, the driving characteristic data of surrounding vehicles characterizes the relative motion relationship between the vehicle and surrounding vehicles, as well as the motion state of surrounding vehicles, while traffic environment data provides scenario constraints. The traffic environment data and the aforementioned driving characteristic data work together to form a complete input support for the collision prediction model. The driving characteristic data captures the relative motion trend between the vehicle and surrounding vehicles, while the traffic environment data supplements the scenario constraints, ultimately achieving accurate prediction of collisions between the vehicle and surrounding vehicles.

[0061] Step 203: Based on the prediction results, determine the avoidance strategy, and control the vehicle based on the avoidance strategy.

[0062] Based on the collision probability, collision time, and collision severity of each surrounding vehicle colliding with the vehicle, the prediction with the highest risk level is determined, such as the prediction with the highest collision probability being taken as the highest risk level prediction. Then, based on the collision probability, collision time, and collision severity of the highest risk level prediction, the avoidance space status of the vehicle's adjacent lanes, and whether the corresponding surrounding vehicle is in front of or behind the vehicle, an appropriate avoidance strategy is determined from preset collision strategies. For example, for the vehicle in front of the vehicle, the preset collision strategy may include braking control, lane changing, and speed adjustment; for the vehicle behind the vehicle, the preset collision strategy may include acceleration control, lane changing, and speed adjustment. Considering the avoidance space status of adjacent lanes is to determine whether the lane changing strategy can be safely executed. The avoidance space status can include whether there is avoidance space or not, and can be obtained based on the driving characteristic data of surrounding vehicles, such as the distance and relative speed between surrounding vehicles and the vehicle, and the road topology data of the road where the vehicle is located.

[0063] For example, if the predicted vehicle is the one in front of your vehicle, and the collision probability is high or the collision time is short, it indicates a high collision risk, and the avoidance strategy is determined to be braking control. If the avoidance space status is "no avoidance space," it means the lane change conditions are not met. If the collision probability is low but the collision time is long, speed adjustment or braking control can be chosen. If the avoidance space status is "avoidance space," it means the lane change conditions are met. If the collision probability is low, the collision time is long, and the collision severity is low, the avoidance strategy can be determined to be lane change or speed adjustment.

[0064] The vehicle collision prediction and avoidance method provided in this application acquires driving characteristic data of surrounding vehicles during the vehicle's operation. When a surrounding vehicle brakes suddenly, it indicates that a sudden change in the driving state of a vehicle around the vehicle has occurred, suggesting that an accident may have already happened or there may be a potential collision risk. To reduce the occurrence of collision accidents, the method uses the driving characteristic data that characterizes the relative motion relationship between the surrounding vehicles and the vehicle, the motion state of the surrounding vehicles, and the traffic environment data that provides scenario constraints to make a forward-looking prediction of whether a sudden collision will occur between the surrounding vehicles and the vehicle. This allows for early identification of collision risks, and based on the prediction results, a corresponding avoidance strategy is determined to control the vehicle to avoid collision risks, reduce the incidence of collision accidents, and protect the lives and property of drivers and passengers.

[0065] In addition, when transmitting acoustic signals around the vehicle, this embodiment of the application also considers covering the area around the vehicle as wide as possible in order to comprehensively and accurately detect the driving status of surrounding vehicles.

[0066] As mentioned above, the acoustic system installed in the vehicle includes a forward acoustic array installed at the front of the vehicle and a rearward acoustic array installed at the rear of the vehicle.

[0067] The forward acoustic array may include a forward speaker array and a forward microphone array. The forward speaker array periodically emits forward acoustic signals of a specific frequency to the area around the vehicle. The forward acoustic signals are reflected when they encounter vehicles around the vehicle, returning corresponding forward echo signals. The forward microphone array receives the forward echo signals.

[0068] Similarly, a rear-facing acoustic array may include a rear-facing speaker array and a rear-facing microphone array. The rear-facing speaker array periodically emits rear-facing acoustic signals of a specific frequency to the area around the vehicle. The rear-facing acoustic signals are reflected when they encounter vehicles around the vehicle, returning corresponding rear-facing echo signals. The rear-facing microphone array receives the rear-facing echo signals.

[0069] For example, the forward speaker array can consist of 16 high-frequency speaker units arranged in a 4×4 matrix, with a spacing of λ / 2 between each unit to suppress grating interference and ensure sound field directivity and beamforming effect. Here, λ is the center wavelength corresponding to the operating frequency range of each speaker. The forward microphone array can consist of 20 high-sensitivity microphone units arranged in a circle, with a radius of 0.3 meters. To ensure accurate echo signal capture, the sampling frequency of the microphone units can be set to 192kHz. The number and arrangement of the units constituting the forward speaker and microphone arrays can also be other quantities and forms, which can be adjusted according to the vehicle's shape and detection requirements; no specific limitations are imposed here. Furthermore, both the speaker and microphone units are waterproof and dustproof, adaptable to various climatic conditions, and unaffected by weather or light intensity, allowing for all-weather operation.

[0070] The forward speaker array and forward microphone array can be positioned inside the vehicle's front bumper, with their centers located on the vehicle's longitudinal axis to ensure symmetrical coverage of the forward sound field. The sound field coverage of the forward speaker array and the pickup coverage of the forward microphone array should cover as much of the area in front of the vehicle as possible; for example, the forward speaker array should provide sound field coverage of at least a 120° fan-shaped area in front of the vehicle. Based on these arrays, the driving status of surrounding vehicles can be comprehensively and accurately detected. To prevent interference between the forward speaker array and the forward microphone array, a physical sound barrier can be placed between them. Additionally, the forward microphone array also features echo cancellation.

[0071] Similarly, the rear speaker array can consist of 12 high-frequency speaker units arranged in a 3×4 matrix. The rear microphone array can consist of 16 high-sensitivity microphone units arranged in an elliptical pattern. Both the rear speaker and microphone arrays can be positioned inside the rear bumper, with their centers located on the vehicle's longitudinal axis to ensure symmetrical rear sound field coverage. Furthermore, the sound field coverage of the rear speaker array and the pickup coverage of the rear microphone array should cover as much of the area behind the vehicle as possible; for example, the rear microphone array should be able to achieve at least a 120° fan-shaped pickup coverage area behind the vehicle.

[0072] For details on the specific setup and functions of the rear speaker array and rear microphone array, please refer to the relevant descriptions of the front speaker array and front microphone array; they will not be repeated here.

[0073] This acoustic system, which includes a forward-facing acoustic array and a rearward-facing acoustic array, can emit acoustic signals that can cover a wide area around the vehicle, especially the area in front of and behind the vehicle.

[0074] Optionally, in this embodiment, the acoustic system can also be installed in different locations such as the top and / or side of the vehicle to adapt to various application scenarios.

[0075] In addition, this embodiment further refines how to determine when surrounding vehicles are braking suddenly, how to obtain predictions of collisions between surrounding vehicles and the vehicle itself, and how to determine avoidance strategies, so as to better avoid collision risks.

[0076] Figure 3 This is a flowchart illustrating a vehicle collision prediction and avoidance method provided in another embodiment of this application. Figure 3 As shown, the method in the embodiments of this application may include: Step 301: While the vehicle is in motion, emit acoustic signals to the surrounding area and receive the corresponding echo signals.

[0077] In some embodiments, a forward acoustic array can be controlled to emit forward acoustic signals around the vehicle and receive forward echo signals returned after the forward acoustic signals encounter surrounding vehicles; and a rearward acoustic array can be controlled to emit rearward acoustic signals around the vehicle and receive rearward echo signals returned after the rearward acoustic signals encounter surrounding vehicles.

[0078] A forward-facing acoustic array can achieve sound field coverage and sound pickup coverage in a fan-shaped area at a preset angle in front of the vehicle, and a rear-facing acoustic array can achieve sound field coverage and sound pickup coverage in a fan-shaped area at a preset angle behind the vehicle, such as a preset angle of 120°.

[0079] Both the forward and backward acoustic signals mentioned above can be FMCW signals. An FMCW signal is a continuous wave signal whose frequency changes continuously with time according to a specific pattern. In this embodiment, the frequency range of the acoustic signal can be 20kHz to 40kHz. This frequency range belongs to the ultrasonic band and has high directivity, anti-interference, and propagation characteristics. For example, when transmitting a forward acoustic signal, the forward acoustic signal serves as an FMCW signal with a starting frequency of 20kHz, a bandwidth of 20kHz, and a frequency modulation period of 10ms. The backward acoustic signal is similar to the forward acoustic signal and will not be described further here.

[0080] Optionally, the acoustic signal can be an FMCW signal, or it can be a pulse signal, a phase-coded signal, a linear frequency modulated signal, etc.

[0081] In this embodiment, by setting forward acoustic arrays at the front and rear of the vehicle respectively, the emitted acoustic signals can cover a wide area in front of and behind the vehicle, thereby reducing blind spots and enabling wide-range detection of the area around the vehicle. This allows for comprehensive and accurate detection of vehicles around the vehicle, and thus accurately obtains the relative motion relationship between the surrounding vehicles and the vehicle, as well as the driving status of the surrounding vehicles.

[0082] In some embodiments, when controlling the forward acoustic array to emit forward acoustic signals around the vehicle, noise signals around the vehicle can be acquired, and the characteristic frequency range of the forward acoustic signal can be determined based on the frequency range of the noise signal and a preset frequency range. Then, the forward acoustic array is controlled to emit forward acoustic signals within the characteristic frequency range around the vehicle.

[0083] Here, noise signals around the vehicle are also considered to reduce their impact on acoustic and echo signals. A preset frequency range, such as 20kHz to 40kHz, represents the frequency range of the emitted acoustic signal under noise-free conditions. When the frequency range of the detected noise signal around the vehicle overlaps with the preset frequency range, a characteristic frequency range is determined based on the non-overlapping frequency ranges to minimize noise interference. For example, if the frequency range of the noise signal around the vehicle is 15kHz to 25kHz due to various factors, the characteristic frequency range of the acoustic signal is determined based on the 25kHz to 40kHz range; for instance, the characteristic frequency range could be 30kHz to 40kHz.

[0084] In this embodiment, the frequency range of the acoustic signal is adaptively adjusted based on the noise signal around the vehicle, which can reduce the impact of environmental noise on acoustic detection and improve the detection accuracy of vehicles around the vehicle, so that the driving characteristic data of the surrounding vehicles can be accurately obtained in the future.

[0085] Step 302: Determine the driving characteristic data of vehicles around the vehicle based on the echo signal.

[0086] Specifically, the driving characteristic data of surrounding vehicles are determined based on the forward echo signal and the rear acoustic signal. The driving characteristic data includes the distance, relative speed, and relative orientation between the surrounding vehicles and the vehicle, as well as the acceleration of the surrounding vehicles, etc., which are used to characterize the relative motion relationship between the surrounding vehicles and the vehicle, and the motion state of the surrounding vehicles.

[0087] In some embodiments, when determining the driving characteristic data of vehicles surrounding the vehicle, for each of the surrounding vehicles, an acoustic signal corresponding to the reception time of the echo signal of that vehicle can be obtained. The echo signal and the acoustic signal corresponding to the reception time can be mixed and filtered to obtain a difference frequency signal. The distance and relative speed between the vehicle and the vehicle can be determined based on the peak frequency of the difference frequency signal.

[0088] Then, based on the time difference or phase difference of the echo signal received by the vehicle, the relative orientation between the vehicle and the vehicle is obtained, and based on the relative speed between the vehicle and the vehicle, the acceleration of the vehicle is determined.

[0089] Optionally, for each of the surrounding vehicles, the acoustic signal corresponding to the time the echo signal was received is mixed with the echo signal and low-pass filtered to obtain an effective difference frequency signal. Then, a Fast Fourier Transform (FFT) is performed on the difference frequency signal, and the peak frequency is extracted based on the transform result. The ratio between the product of the speed of sound, the peak frequency, and the sweep period, and twice the bandwidth, is then used as the distance between the vehicle and the vehicle in question.

[0090] A distance-Doppler two-dimensional FFT is performed on the difference frequency signal over multiple periods. Based on the processing results, the Doppler frequency shift is separated. The ratio between the product of the Doppler frequency shift and the speed of sound, and twice the center frequency of the acoustic signal, is then used as the relative speed between the vehicle and its own vehicle. The vehicle's acceleration can be obtained based on the rate of change of relative speed, such as the rate of change of relative speed over multiple consecutive frames.

[0091] For relative orientation, the acoustic system includes multiple microphone units. Based on the time difference or phase difference of the echo signals received by the multiple microphone units, beamforming algorithms such as delay-summation algorithms are used to form multiple beams pointing in a specific direction. Then, based on the direction corresponding to the beam with the strongest intensity, the relative orientation between the vehicle and the other vehicle is obtained.

[0092] Step 303: For each vehicle in the surrounding area, determine whether there is deceleration in the vehicle's driving characteristic data within a preset time period. If there is deceleration, obtain the average deceleration value based on the deceleration and determine whether the average deceleration value is greater than a preset braking threshold.

[0093] As mentioned above, when a vehicle around your car brakes suddenly, indicating a sudden change in its driving status, it suggests that an accident may have occurred nearby or there is a potential collision risk. Therefore, it is necessary to proactively predict whether a collision will occur between your vehicle and the surrounding vehicles to reduce the likelihood of a collision. Thus, it is essential to identify whether surrounding vehicles are braking suddenly.

[0094] For each vehicle in the surrounding area, if, within a preset time period, among the multiple accelerations corresponding to that vehicle, there is a deceleration and the average value of the deceleration is greater than a preset braking threshold, it indicates that the vehicle's deceleration behavior within the preset time period is characterized by a large amplitude and strong continuity, belonging to an emergency braking situation of sudden and drastic change in driving state. In other words, it is determined that the surrounding vehicles are experiencing emergency braking. Here, acceleration includes positive and negative values. If the acceleration is negative, it indicates the presence of deceleration, and the magnitude of the deceleration is equal to the absolute value of the acceleration.

[0095] If, among the multiple accelerations corresponding to a vehicle, there is a deceleration and the average value of the deceleration is less than or equal to a preset braking threshold, the vehicle is considered to be undergoing normal deceleration rather than emergency braking. If, among the multiple accelerations corresponding to a vehicle, there is no deceleration, it indicates that the vehicle has not undergone emergency braking. In other words, it is determined that there is no emergency braking situation among surrounding vehicles.

[0096] In this embodiment, for each vehicle in the surrounding area, emergency braking is determined based on multiple accelerations within a preset time period. This avoids the problem of misjudgment caused by direct judgment based on a single point deceleration, which is easily affected by road bumps or other factors. By considering the relationship between the average deceleration and a preset braking threshold, it ensures that deceleration behaviors with large amplitude and strong persistence can be effectively filtered out. This accurately distinguishes between normal deceleration and emergency braking, and accurately determines whether any vehicle in the surrounding area is undergoing emergency braking. This provides a valid basis for subsequent prediction of whether a collision between the vehicle and surrounding vehicles will occur.

[0097] Step 304: If the average deceleration is greater than the preset braking threshold, input the driving characteristic data of the surrounding vehicles and the traffic environment data into the collision prediction model to obtain the prediction results of the collision between each of the surrounding vehicles and the vehicle itself.

[0098] The driving characteristic data of surrounding vehicles refers to driving characteristic data within a certain time period. The collision prediction model above takes into account the driving characteristic data of surrounding vehicles within a certain time period and traffic environment data, and outputs the prediction results of collisions between each of the surrounding vehicles and the vehicle itself. The prediction results include collision probability, collision time and collision severity.

[0099] As mentioned above, traffic environment data includes collision location, collision speed, traffic density, road surface condition, and weather, which are used to provide scenario constraints in collision prediction. The collision location is the relative position of the vehicle to the surrounding vehicles that have already collided, and the collision speed is the relative speed between the vehicle and the surrounding vehicles before the collision. Traffic density refers to the number of surrounding vehicles; under the same motion parameters, the higher the traffic density, the lower the feasibility of the vehicle's avoidance maneuver. Road surface condition includes the smoothness of the road surface on the current driving segment, while weather conditions such as rain and snow affect the road surface adhesion.

[0100] It is important to note that if there are no vehicles in the surrounding area that have already collided, the collision location and collision speed in the traffic environment data can be equivalently replaced by the relative position and relative speed of the vehicle closest to the vehicle that has undergone emergency braking. This ensures the validity of the input to the collision prediction model, and thus guarantees the accuracy and validity of the prediction results output by the collision prediction model.

[0101] Here, collision prediction based on vehicles that have already collided with the vehicle's surroundings takes into account the chain reaction effect in traffic accidents. When there are vehicles that have already collided with the vehicle's surroundings, secondary and tertiary chain collisions are highly likely to occur. Therefore, in this embodiment, the collision prediction uses relevant data of vehicles that have already collided with surroundings as scenario constraints to improve the ability to predict collisions caused by chain collisions with surrounding vehicles. Traffic density, road conditions, and weather, which are related to the vehicle's actual driving environment, are used to correct the prediction results output by the collision prediction model, making the prediction results more consistent with real driving scenarios.

[0102] It should be noted that, in this embodiment, considering that the probability of a collision between the vehicle and the surrounding vehicle is higher when the vehicle is braking suddenly, the prediction results of a collision between the vehicle and the surrounding vehicle when the vehicle is braking suddenly can be taken from the prediction results of a collision between each of the surrounding vehicles and the vehicle output by the collision prediction model, and used to determine the avoidance strategy in the future.

[0103] In this embodiment, considering the chain reaction effect in traffic accidents, the collision location and collision time related to vehicles that have already collided with surrounding vehicles are used as traffic environment data. This improves the ability to predict collisions caused by a chain reaction of collisions involving surrounding vehicles. Simultaneously, driving feature data, used to characterize the relative motion relationship between the vehicle and surrounding vehicles, as well as the motion state of surrounding vehicles, works synergistically with traffic environment data to provide scenario constraints. This provides complete input support for the collision prediction model, capturing the relative motion trend between the vehicle and surrounding vehicles through driving feature data and supplementing scenario constraints with traffic environment data, ultimately achieving accurate prediction of collisions between the vehicle and surrounding vehicles.

[0104] Here, the specific implementation method and principle of step 304 can be found in [reference]. Figure 2 The relevant descriptions in the embodiments will not be repeated here. For a detailed description and explanation of the driving characteristic data, traffic environment data, and prediction results, please refer to [reference needed]. Figure 2 The relevant descriptions in the embodiments.

[0105] In some embodiments, the collision prediction model may include a deep learning predictor and / or a shock wave simulation model. The inputs to the deep learning predictor and the shock wave simulation model are driving characteristic data of surrounding vehicles within a certain time period, as well as traffic environment data, and the outputs are the prediction results of collisions between the vehicle and each of the surrounding vehicles.

[0106] Optionally, before using the deep learning predictor for prediction, the deep learning predictor is first trained. The training samples are the driving characteristic data of surrounding vehicles and traffic environment data of different vehicles within a historical time period. Each training sample is labeled with the collision probability, collision time, and collision severity between the surrounding vehicle and the corresponding vehicle. Based on the training samples and corresponding labels, the initial deep learning predictor is jointly trained through multiple tasks until a preset stopping condition is met, resulting in a trained deep learning predictor. The initial deep learning model can be a temporal network, etc.

[0107] Here, reaching the preset stopping condition can be reaching a preset number of training iterations, or satisfying that the loss value is less than a preset loss threshold. The loss value can be a weighted sum of probability loss, time loss, and severity loss.

[0108] For example, the shock wave simulation model is an analytical model built based on vehicle kinematics and traffic flow theory. Before using the shock wave simulation model for prediction, the initial shock wave simulation model is first corrected and adjusted using the driving characteristic data of surrounding vehicles and traffic environment data of different vehicles within a historical time period, as well as the collision probability, collision time and collision degree of surrounding vehicles and corresponding vehicles. For example, the correction coefficients in the initial shock wave simulation model are adjusted to reduce errors, resulting in the adjusted shock wave simulation model.

[0109] When the collision prediction model includes both a deep learning predictor and a shockwave simulation model, the driving characteristic data of surrounding vehicles and traffic environment data are input into the deep learning predictor and the shockwave simulation model, respectively, to obtain the prediction results of each of the surrounding vehicles colliding with the vehicle. The collision probability, collision time, and collision severity from the deep learning predictor's output are averaged with the collision probability, collision time, and collision severity from the shockwave simulation model's output to obtain the final prediction result of each of the surrounding vehicles colliding with the vehicle. An avoidance strategy is then determined based on the final prediction result.

[0110] Here, the deep learning predictor and the shock wave simulation model are two prediction models with different core logics. The deep learning predictor is trained based on massive amounts of historical driving feature data and traffic environment data, and can output high-precision prediction results for complex traffic scenarios around the vehicle. The shock wave simulation model, on the other hand, is built based on physical laws and traffic flow theory, and has high reliability and interpretability. Therefore, in this embodiment, the final prediction result is obtained based on the prediction results of the deep learning predictor and the prediction results of the shock wave simulation model, balancing accurate prediction of complex traffic scenarios with interpretability of physical laws. This achieves synergistic optimization of high-precision prediction and forward-looking extrapolation, making the final collision probability, collision time, and collision severity more reliable and accurate.

[0111] Step 305: Obtain the avoidance space status of the adjacent lanes of the vehicle's lane, determine the avoidance strategy based on the prediction results and the avoidance space status, and control the vehicle based on the avoidance strategy.

[0112] The avoidance space status includes both having avoidance space and not having avoidance space. Preset avoidance strategies can include braking control, acceleration control, lane changing, and speed adjustment. Braking control is emergency braking, with the core objective of rapidly reducing vehicle speed and shortening braking distance. Acceleration control is emergency acceleration, with the objective of rapidly increasing vehicle speed and increasing the relative distance from the hazard. Speed ​​adjustment aims to smoothly regulate vehicle speed and maintain a safe following distance.

[0113] Here, the collision probability, collision time and collision severity in the prediction results are considered simultaneously, as well as the avoidance space status of adjacent lanes. This achieves a precise coupling of risk prediction and avoidance feasibility, ensuring both the safety, stability and scenario adaptability of the final avoidance strategy, and guaranteeing the feasibility of the avoidance strategy in practice, thus avoiding ineffective decisions.

[0114] In some embodiments, when determining an avoidance strategy, if there are multiple prediction results, the prediction result with the highest risk level can be determined from each prediction result. Based on the collision probability, collision time, and collision degree in the prediction result with the highest risk level, as well as the preset probability threshold, preset time threshold, preset degree threshold, and avoidance space state, the optimal avoidance strategy can be determined from the preset avoidance strategies and used as the avoidance strategy.

[0115] Optionally, there may be multiple vehicles braking suddenly among the vehicles surrounding the vehicle. Accordingly, each of the aforementioned vehicles braking suddenly corresponds to a predicted collision with the vehicle. When there are multiple predicted outcomes, the optimal avoidance strategy can be determined from the preset avoidance strategies based on the prediction outcome with the highest risk level, so as to prioritize the driving safety of the vehicle in extremely dangerous scenarios.

[0116] Based on the collision probability, collision time, and collision severity from the highest-risk prediction, along with the corresponding preset thresholds, the collision risk level corresponding to the prediction can be determined. Considering the avoidance space state ensures the feasibility of the avoidance strategy. Thus, by simultaneously considering both the collision risk level and the avoidance space state, the feasibility and optimality of the determined avoidance strategy can be guaranteed.

[0117] In this embodiment, the optimal avoidance strategy is obtained based on the prediction result with the highest risk level, prioritizing driving safety in extremely dangerous scenarios. Furthermore, setting preset thresholds quantifies collision risk from multiple dimensions, avoiding the limitations of single-indicator decision-making. Considering the avoidance space state further ensures the feasibility of the final avoidance strategy. In other words, both collision risk level and avoidance space state are considered simultaneously to obtain the most suitable and feasible avoidance strategy.

[0118] Optionally, when the predicted vehicle is the vehicle in front of the vehicle, the preset avoidance strategy includes braking control, lane changing, and speed adjustment; when the predicted vehicle is the vehicle behind the vehicle, the preset avoidance strategy includes acceleration control, lane changing, and speed adjustment. As mentioned above, braking control is an emergency braking strategy, with the core objective of rapidly reducing vehicle speed and shortening braking distance. Similarly, acceleration control is an emergency acceleration strategy, with the objective of rapidly increasing vehicle speed and increasing the relative distance from the hazard. Braking control and acceleration control are suitable for scenarios with high collision risk. Speed ​​adjustment, on the other hand, aims to smoothly regulate vehicle speed and maintain a safe following distance, balancing safety and driving stability, and is suitable for scenarios with low collision risk.

[0119] In some embodiments, when the vehicle corresponding to the highest risk level prediction is the vehicle in front of the user, A1. If the collision probability is greater than or equal to the preset probability threshold, or the collision time is less than or equal to the preset time threshold, then the optimal avoidance strategy is determined to be braking control.

[0120] A2. If the collision probability is less than the preset probability threshold, the collision time is greater than the preset time threshold, and the avoidance space status is no avoidance space, then the optimal avoidance strategy is determined to be speed adjustment or braking control.

[0121] A3. If the collision probability is less than the preset probability threshold, the collision time is greater than the preset time threshold, the collision severity is less than the preset severity threshold, and the avoidance space status is "there is avoidance space", then the optimal avoidance strategy is determined to be lane change or speed adjustment.

[0122] A4. If the collision probability is less than the preset probability threshold, the collision time is greater than the preset time threshold, the collision severity is greater than or equal to the preset severity threshold, and the avoidance space status is "there is avoidance space", then the optimal avoidance strategy is determined to be lane change or braking control.

[0123] In case A1, the collision risk level is considered high. In this case, strong braking is required to quickly reduce the vehicle speed, shorten the collision time window, and prioritize avoiding the collision risk.

[0124] In scenario A2, if the collision severity is less than a preset threshold, the collision risk is considered low and there is no opportunity to change lanes. In this case, simply reduce the vehicle's speed slightly and increase the safe following distance to balance safety and driving stability. If the collision severity is greater than the preset threshold, and there is no opportunity to change lanes, quickly reduce the vehicle's speed to prioritize avoiding the collision risk.

[0125] In scenario A3, where the collision risk level is considered low and there is a lane-changing opportunity, you can directly control your vehicle to change to a safe lane to escape the risk area, or you can control your vehicle to slow down slightly to increase the safe following distance.

[0126] In the case of A4, given the high degree of collision and the opportunity to change lanes, you can directly control your vehicle to change to a safe lane to escape the risk area. Alternatively, to prioritize avoiding collision risks, you can quickly reduce your vehicle speed.

[0127] Optionally, the preset probability threshold, preset time threshold, and preset severity threshold mentioned above can be set according to actual conditions and specific needs. Each of the preset thresholds may also include multiple sub-thresholds. For example, the preset probability threshold may include multiple sub-thresholds. Thus, based on the collision probability, collision time, and collision severity in the prediction results, as well as the avoidance space status and each sub-threshold, a suitable and feasible avoidance strategy can be determined. The specific implementation method can be referred to steps A1 to A4 above, which will not be elaborated here.

[0128] In some embodiments, when the vehicle corresponding to the highest risk level prediction is the vehicle behind the vehicle, if the collision probability is greater than or equal to a preset probability threshold, or the collision time is less than or equal to a preset time threshold, the collision risk level is considered high. In this case, a high-intensity acceleration is required to quickly increase the vehicle's speed and prioritize avoiding the collision risk; that is, the optimal avoidance strategy is determined to be acceleration control. Other situations and corresponding avoidance strategies can be referred to the relevant descriptions of steps A1 to A4 in the above embodiments, and will not be repeated here.

[0129] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0130] Figure 4 This is a schematic diagram of the structure of a vehicle collision prediction and avoidance device provided in an embodiment of this application. Figure 4 As shown, the vehicle collision prediction and avoidance device provided in this embodiment may include: an acquisition module 401, an acquisition module 402, and a control module 403.

[0131] The acquisition module 401 is used to acquire driving characteristic data of vehicles around the vehicle while the vehicle is driving.

[0132] The module 402 is used to obtain a prediction result of a collision between the vehicle and the surrounding vehicles based on the driving characteristic data of the surrounding vehicles and traffic environment data when the surrounding vehicles brake suddenly.

[0133] The control module 403 is used to determine an avoidance strategy based on the prediction result, and to control the vehicle based on the avoidance strategy.

[0134] In one possible implementation, a forward acoustic array is provided at the front end of the vehicle, and a rear acoustic array is provided at the rear end of the vehicle; the acquisition module 401 is also used for: The system controls the forward acoustic array to emit forward acoustic signals around the vehicle and receives the forward echo signals returned after the forward acoustic signals encounter surrounding vehicles. The rear acoustic array is controlled to emit rear acoustic signals around the vehicle, and the rear echo signals returned after the rear acoustic signals encounter surrounding vehicles are received. Based on the forward echo signal and the backward acoustic signal, the driving characteristic data of the surrounding vehicles are determined.

[0135] In one possible implementation, the acquisition module 401 is also used for: Acquire noise signals around the vehicle; The characteristic frequency range of the forward acoustic signal is determined based on the frequency range of the noise signal and the preset frequency range. The forward acoustic array is controlled to emit forward acoustic signals within the characteristic frequency range around the vehicle.

[0136] In one possible implementation, module 402 is also used for: For each vehicle in the surrounding area, determine whether there is deceleration in the vehicle's driving characteristic data within a preset time period. If deceleration exists, the average deceleration value is obtained based on the deceleration value, and it is determined whether the average deceleration value is greater than a preset braking threshold. If the average deceleration is greater than the preset braking threshold, it is determined that the vehicle has undergone emergency braking.

[0137] In one possible implementation, module 402 is also used for: The driving characteristic data of the surrounding vehicles and the traffic environment data are input into the collision prediction model to obtain the prediction results of the collision prediction model for each of the surrounding vehicles colliding with the vehicle itself. The driving characteristic data includes the distance, relative speed, and relative orientation between the vehicle and surrounding vehicles, as well as the acceleration of surrounding vehicles. The traffic environment data includes the collision location, collision speed, traffic density, road surface condition, and weather. The collision location is the relative position of the vehicle and the surrounding vehicles that have already collided. The collision speed is the relative speed between the vehicle and the surrounding vehicles that have already collided before the collision.

[0138] In one possible implementation, the control module 403 is also used for: Obtain the obstacle avoidance space status of the adjacent lanes of the lane where the vehicle is located; the obstacle avoidance space status includes whether there is obstacle avoidance space and whether there is no obstacle avoidance space. Based on the prediction results and the avoidance space status, the avoidance strategy is determined.

[0139] In one possible implementation, the prediction result includes collision probability, collision time, and collision severity; the control module 403 is further configured to: When there are multiple prediction results, the prediction result with the highest risk level is determined from all the prediction results; Based on the collision probability, collision time, and collision severity in the prediction result with the highest risk level, as well as preset probability thresholds, preset time thresholds, preset severity thresholds, and the avoidance space state, the optimal avoidance strategy is determined from the preset avoidance strategies, and the optimal avoidance strategy is used as the avoidance strategy.

[0140] In one possible implementation, the preset avoidance strategy includes braking control, lane changing, and speed adjustment; the control module 403 is further configured to: When the vehicle corresponding to the highest risk level prediction result is the vehicle in front of the vehicle, if the collision probability is greater than or equal to the preset probability threshold, or the collision time is less than or equal to the preset time threshold, then the optimal avoidance strategy is determined to be braking control. If the collision probability is less than the preset probability threshold, the collision time is greater than the preset time threshold, and the avoidance space status is no avoidance space, then the optimal avoidance strategy is determined to be speed adjustment or braking control. If the collision probability is less than a preset probability threshold, the collision time is greater than a preset time threshold, the collision severity is less than a preset severity threshold, and the avoidance space status is "there is avoidance space", then the optimal avoidance strategy is determined to be lane change or speed adjustment.

[0141] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0142] Figure 5 This is a schematic diagram of the controller provided in one embodiment of this application. Figure 5 As shown, the controller 500 of this embodiment includes a processor 510 and a memory 520, wherein the memory 520 stores a computer program 521 that can run on the processor 510. When the processor 510 executes the computer program 521, it implements the steps in any of the above method embodiments, for example... Figure 2 Steps 201 to 203 are shown. Alternatively, when processor 510 executes computer program 521, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 4 The functions of modules 401 to 403 are shown.

[0143] For example, computer program 521 may be divided into one or more modules / units, one or more of which are stored in memory 520 and executed by processor 510 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 521 in controller 500.

[0144] Those skilled in the art will understand that Figure 5 This is merely an example of a controller and does not constitute a limitation on the controller. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0145] The processor 510 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0146] The memory 520 can be an internal storage unit of the controller, such as the controller's hard drive or memory, or an external storage device of the controller, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. The memory 520 can also include both internal and external storage units. The memory 520 is used to store computer programs and other programs and data required by the controller. The memory 520 can also be used to temporarily store data that has been output or will be output.

[0147] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0148] One embodiment of this application also provides a vehicle including the controller as described above.

[0149] An embodiment of this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described vehicle collision prediction and avoidance method.

[0150] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0151] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0152] In the embodiments provided in this application, it should be understood that the disclosed devices / controllers and methods can be implemented in other ways. For example, the device / controller embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0153] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0154] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0155] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0156] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A vehicle collision prediction and avoidance method, characterized in that, include: While the vehicle is in motion, acquire driving characteristic data of vehicles around the vehicle; When surrounding vehicles brake suddenly, the vehicle is predicted to collide with the surrounding vehicles based on the driving characteristic data of the surrounding vehicles and traffic environment data. Based on the prediction results, an avoidance strategy is determined, and the vehicle is controlled based on the avoidance strategy.

2. The vehicle collision prediction and avoidance method according to claim 1, characterized in that, The vehicle is equipped with a forward acoustic array at the front and a rear acoustic array at the rear. The acquisition of driving characteristic data of vehicles surrounding the vehicle includes: The system controls the forward acoustic array to emit forward acoustic signals around the vehicle and receives the forward echo signals returned after the forward acoustic signals encounter surrounding vehicles. The rear acoustic array is controlled to emit rear acoustic signals around the vehicle, and the rear echo signals returned after the rear acoustic signals encounter surrounding vehicles are received. Based on the forward echo signal and the backward acoustic signal, the driving characteristic data of the surrounding vehicles are determined.

3. The vehicle collision prediction and avoidance method according to claim 2, characterized in that, The control of the forward acoustic array to emit forward acoustic signals around the vehicle includes: Acquire noise signals around the vehicle; The characteristic frequency range of the forward acoustic signal is determined based on the frequency range of the noise signal and the preset frequency range. The forward acoustic array is controlled to emit forward acoustic signals within the characteristic frequency range around the vehicle.

4. The vehicle collision prediction and avoidance method according to any one of claims 1 to 3, characterized in that, Before obtaining a prediction of a collision between the vehicle and surrounding vehicles based on the driving characteristic data and traffic environment data of the surrounding vehicles in the event of emergency braking by surrounding vehicles, the method further includes: For each vehicle in the surrounding area, determine whether there is deceleration in the vehicle's driving characteristic data within a preset time period. If deceleration exists, the average deceleration value is obtained based on the deceleration value, and it is determined whether the average deceleration value is greater than a preset braking threshold. If the average deceleration is greater than the preset braking threshold, it is determined that the vehicle has undergone emergency braking.

5. The vehicle collision prediction and avoidance method according to any one of claims 1 to 3, characterized in that, The step of obtaining a prediction result of a collision between the vehicle and surrounding vehicles based on the driving characteristic data of surrounding vehicles and traffic environment data includes: The driving characteristic data of the surrounding vehicles and the traffic environment data are input into the collision prediction model to obtain the prediction results of the collision prediction model for each of the surrounding vehicles colliding with the vehicle itself. The driving characteristic data includes the distance, relative speed, and relative orientation between the vehicle and surrounding vehicles, as well as the acceleration of surrounding vehicles. The traffic environment data includes the collision location, collision speed, traffic density, road surface condition, and weather. The collision location is the relative position of the vehicle and the surrounding vehicles that have already collided. The collision speed is the relative speed between the vehicle and the surrounding vehicles that have already collided before the collision.

6. The vehicle collision prediction and avoidance method according to any one of claims 1 to 3, characterized in that, The step of determining an avoidance strategy based on the prediction results includes: Obtain the obstacle avoidance space status of the adjacent lanes of the lane where the vehicle is located; the obstacle avoidance space status includes whether there is obstacle avoidance space and whether there is no obstacle avoidance space. Based on the prediction results and the avoidance space status, the avoidance strategy is determined.

7. The vehicle collision prediction and avoidance method according to claim 6, characterized in that, The prediction results include collision probability, collision time, and collision severity. Determining the avoidance strategy based on the prediction result and the avoidance space state includes: When there are multiple prediction results, the prediction result with the highest risk level is determined from all the prediction results; Based on the collision probability, collision time, and collision severity in the prediction result with the highest risk level, as well as preset probability thresholds, preset time thresholds, preset severity thresholds, and the avoidance space state, the optimal avoidance strategy is determined from the preset avoidance strategies, and the optimal avoidance strategy is used as the avoidance strategy.

8. The vehicle collision prediction and avoidance method according to claim 7, characterized in that, The preset avoidance strategies include braking control, lane changing, and vehicle speed adjustment; The step of determining the optimal avoidance strategy from preset avoidance strategies based on the collision probability, collision time, and collision severity in the prediction result with the highest risk level, as well as preset probability thresholds, preset time thresholds, preset severity thresholds, and the avoidance space state, includes: When the vehicle corresponding to the highest risk level prediction result is the vehicle in front of the vehicle, if the collision probability is greater than or equal to the preset probability threshold, or the collision time is less than or equal to the preset time threshold, then the optimal avoidance strategy is determined to be braking control. If the collision probability is less than the preset probability threshold, the collision time is greater than the preset time threshold, and the avoidance space status is no avoidance space, then the optimal avoidance strategy is determined to be speed adjustment or braking control. If the collision probability is less than a preset probability threshold, the collision time is greater than a preset time threshold, the collision severity is less than a preset severity threshold, and the avoidance space status is "there is avoidance space", then the optimal avoidance strategy is determined to be lane change or speed adjustment.

9. A controller comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the vehicle collision prediction and avoidance method as described in any one of claims 1 to 8.

10. A vehicle, characterized in that, Includes the controller as described in claim 9.

Images