A vehicle platooning control method, device, electronic equipment, and storage medium

Abstract

This application provides a vehicle platooning control method, device, electronic equipment, and storage medium, including: a lead vehicle sending a platooning request signal; subsequent vehicles, upon detecting the platooning request signal, sending back a platooning handshake signal; the lead vehicle completing platooning upon detecting the platooning handshake signal; subsequent vehicles acquiring real-time images of the lead vehicle's taillights via a forward-looking camera, determining a high-precision real-time relative distance based on monocular vision principles and a multi-feature fusion-based distance measurement error dynamic correction model; dynamically adjusting the following distance based on the lead vehicle's driving intention and the high-precision real-time relative distance for platooning and following control; and responding to a preset exit trigger signal to control the vehicle to exit the platoon or disband the entire platoon. This systematically solves key problems of existing V2X-dependent platooning technologies, such as communication interruption in extreme weather, target confusion in multi-vehicle scenarios, inaccurate distance measurement due to slope / lighting, and control lag leading to jerking and rear-end collision risks, thus improving the accuracy of vehicle platooning control.

Description

Technical Field

[0001] This application relates to the field of vehicle cooperative control technology, and in particular to a vehicle platooning control method, device, electronic device and storage medium. Background Technology

[0002] With the rapid development of intelligent connected vehicle technology, the application scenarios of vehicle platooning control technology are becoming increasingly widespread. It has penetrated into multiple fields such as new car assembly line platooning, long-distance commercial vehicle transportation, park commuting, and special operations (such as emergency rescue, mining area transfer, and port container trucks). Especially in scenarios where autonomous driving functions are prone to failure, such as extreme weather (heavy rain, heavy snow, dense fog, strong backlight) and complex road conditions, vehicle platooning technology can ensure the continuity and safety of vehicle transportation through multi-vehicle collaboration and human assistance, solve the problem of transportation interruption after the failure of autonomous driving of a single vehicle, reduce the intensity of personnel operation, and improve transportation efficiency. However, current mainstream vehicle platooning technologies still rely on V2X radio frequency communication, radar / liDAR fusion perception, and fixed distance control. These technologies have numerous shortcomings in practical applications, making them difficult to adapt to complex scenarios, especially transportation needs under extreme weather conditions. Most existing platooning technologies depend on V2X radio frequency communication for command interaction and status synchronization between vehicles. In extreme weather (heavy fog, heavy rain, blizzards), radio frequency signals are easily interfered with and attenuated, leading to communication delays, interruptions, or signaling conflicts. In scenarios with no communication coverage or communication congestion, such as dense traffic flow, closed mining areas, and remote road sections, the platooning system will completely fail, unable to establish a platoon, follow the vehicle, or maintain formation, making it difficult to address emergency transportation needs after autonomous driving fails in extreme weather. Furthermore, V2X communication requires vehicles to be equipped with dedicated communication modules, resulting in high deployment costs and limiting the large-scale application of platooning technology. Meanwhile, traditional platooning distance measurement often relies on single-lens monocular vision or radar, which is susceptible to significant errors due to factors such as road slope, vehicle pitch, lighting changes, and extreme weather conditions. Furthermore, it lacks an effective dynamic correction mechanism. Fixed-distance control modes cannot adapt to the acceleration, deceleration, or turning intentions of the vehicle in front, nor can they compensate for decreased perception accuracy and delayed braking response in extreme weather. This leads to jerky following, insufficient safety redundancy, and a higher risk of accidents due to distance measurement errors in extreme weather. Existing platooning control largely depends on post-event feedback control, meaning that acceleration and deceleration commands are only executed after changes in the status of the vehicle in front are detected, resulting in significant control lag. It lacks the ability to anticipate the driving intentions of the vehicle in front and predict the relative motion trends of the two vehicles. In extreme weather, with slippery roads and extended braking distances, control lag can lead to following too closely and delayed deceleration, increasing the risk of rear-end collisions. It also reduces the smoothness of platooning, impacting the driving experience and transportation efficiency. Therefore, improving the accuracy of vehicle platooning control has become a significant technical challenge. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide a vehicle platooning control method, device, electronic device and storage medium, which realizes platooning handshake, multi-dimensional visual feature adaptive fusion to lock the preceding vehicle, dynamic weighted ranging correction based on monocular vision and multi-source complementary observations, and intention-driven predictive closed-loop control by using the original vehicle lights as the communication medium. It systematically solves the key problems of existing V2X-dependent platooning technology, such as communication interruption in extreme weather, target confusion in multi-vehicle scenes, ranging inaccuracy caused by slope / lighting, and control lag leading to jerking and rear-end collision risks, thereby improving the accuracy of vehicle platooning control.

[0004] This application provides a vehicle platooning control method, the vehicle platooning control method comprising: The lead vehicle sends a formation request signal, and the following vehicles send back a formation handshake signal after detecting the formation request signal. The lead vehicle completes the vehicle formation after detecting the formation handshake signal. The following vehicle uses a forward-facing camera to capture images of the taillights of the vehicle in front in real time. Based on the principle of monocular vision and combined with a multi-feature fusion ranging error dynamic correction model, a high-precision real-time relative distance is determined. Based on the driving intention of the vehicle in front and the high-precision real-time relative distance, the following distance is dynamically adjusted for platooning and following control; wherein, the driving intention of the vehicle in front is obtained by recognizing the timing characteristics of the vehicle in front's lights. In response to a preset exit trigger signal, control the vehicle to leave the formation or disband the entire formation.

[0005] In one possible implementation, after the lead vehicle detects the formation handshake signal and completes vehicle formation, the vehicle formation control method further includes: After formation is completed, adaptive weight allocation and feature-level fusion are performed using static spatial features, dynamic temporal change features, and light intensity distribution features to generate a unique vehicle light feature ID to lock onto the vehicle in front.

[0006] In one possible implementation, the ranging error dynamic correction model based on monocular vision principles and combined with multi-feature fusion, which determines the high-precision real-time relative distance, includes: The image of the front vehicle's taillights is processed based on the principle of monocular vision to determine the basic relative distance; The dynamic correction model for ranging error based on multi-feature fusion corrects the real-time relative distance between vehicles and determines a high-precision real-time relative distance.

[0007] In one possible implementation, the step of processing the image of the front vehicle's taillights based on the monocular vision principle to determine the real-time relative distance between the front and rear vehicles includes: Image preprocessing and ROI constraint detection are performed on the original image; First, a single taillight is coarsely screened by area and shape ratio. Then, the left and right taillights are matched in pairs based on the symmetry, height, size consistency and motion synchronization of the left and right taillights, and the coordinates of the center point of the left and right taillights are output. The pixel spacing of the imaging is determined based on the coordinates of the center points of the left and right taillights; Based on a similar triangle model, the basic relative distance is calculated using the actual physical distance between the left and right taillights of the preceding vehicle, the focal length of the camera, and the pixel spacing of the imaging.

[0008] In one possible implementation, the multi-feature fusion-based ranging error dynamic correction model corrects the real-time relative distance between the front and rear vehicles to determine a high-precision real-time relative distance, including: The basic observation value of the taillight spacing is determined based on the aforementioned basic relative distance; Based on the inverse square relationship between the physical area of ​​a single taillight and the area of ​​an imaging pixel, the observed value of the taillight area is determined. The brightness gradient correction observation value is determined by compensating the basic observation value of the taillight spacing based on the brightness gradient correction coefficient. The basic observation value of the taillight spacing is compensated based on the brake light status correction coefficient to determine the brake light status correction observation value. The high-precision real-time relative distance is determined by dynamically allocating the weights of each group of observations according to the real-time operating conditions through an adaptive weighted fusion method, and then summing the weights after normalization.

[0009] In one possible implementation, the following distance is dynamically adjusted using the following formula: ; in, The adjusted following distance, This refers to the high-precision real-time relative distance. Correction factor for driving intention. This is the environmental adaptive coefficient.

[0010] In one possible implementation, the vehicle platooning control method further includes the following steps: dynamically adjusting the following distance based on the preceding vehicle's driving intention and the high-precision real-time relative distance for platooning control. When the light features are briefly lost or obscured, the formation is maintained based on the last effective distance, speed information and IMU information. After the light features are restored, the following distance is smoothly corrected.

[0011] This application embodiment also provides a vehicle platooning control device, the vehicle platooning control device comprising: The formation establishment module is used for the lead vehicle to send a formation request signal, and the following vehicles to send back a formation handshake signal after detecting the formation request signal. The lead vehicle completes the vehicle formation after detecting the formation handshake signal. The distance calculation module is used by the following vehicle to collect the taillight image of the vehicle in front in real time through the forward-looking camera. Based on the monocular vision principle and combined with the multi-feature fusion distance measurement error dynamic correction model, it determines the high-precision real-time relative distance. The formation control module is used to dynamically adjust the following distance based on the driving intention of the preceding vehicle and the high-precision real-time relative distance for formation and following control; wherein, the driving intention of the preceding vehicle is obtained by identifying the timing characteristics of the preceding vehicle's lights. The formation exit module is used to control the vehicle to exit the formation or disband the entire formation in response to a preset exit trigger signal.

[0012] This application also provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the vehicle platooning control method described above are performed.

[0013] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of the vehicle platooning control method described above.

[0014] This application provides a vehicle platooning control method, device, electronic device, and storage medium. The vehicle platooning control method includes: a lead vehicle sending a platooning request signal; subsequent vehicles sending a platooning handshake signal after detecting the platooning request signal; and the lead vehicle completing platooning after detecting the platooning handshake signal. The subsequent vehicles use a forward-looking camera to capture real-time images of the taillights of the preceding vehicle, and determine a high-precision real-time relative distance based on the principle of monocular vision and a multi-feature fusion-based distance measurement error dynamic correction model. Based on the preceding vehicle's driving intention and the high-precision real-time relative distance, the following distance is dynamically adjusted for platooning and following control. The preceding vehicle's driving intention is identified based on the timing characteristics of its headlights. In response to a preset exit trigger signal, the vehicle exits the platoon or disbands the entire platoon. By using the vehicle's original lights as a communication medium to achieve formation handshake, adaptive fusion of multi-dimensional visual features to lock onto the preceding vehicle, dynamic weighted ranging correction based on monocular vision and multi-source complementary observations, and intention-driven predictive closed-loop control, the system systematically solves key problems of existing V2X-dependent formation technology, such as communication interruption in extreme weather, target confusion in multi-vehicle scenarios, ranging inaccuracy caused by slope / lighting, and control lag leading to jerking and rear-end collision risks, thereby improving the accuracy of vehicle formation control.

[0015] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart of a vehicle platooning control method provided in this application embodiment; Figure 2 This is one of the structural schematic diagrams of a vehicle platooning control device provided in the embodiments of this application; Figure 3 This is a second schematic diagram of the structure of a vehicle platooning control device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0019] First, the applicable application scenarios of this application will be introduced. This application can be applied to the field of automotive cooperative control technology.

[0020] Research has revealed that current mainstream vehicle platooning technologies, still centered on V2X radio frequency communication, radar / liDAR fusion perception, and fixed-distance control, suffer from numerous shortcomings in practical applications. They struggle to adapt to complex scenarios, especially transportation demands under extreme weather conditions. Existing platooning technologies largely rely on V2X radio frequency communication for command interaction and status synchronization between vehicles. In extreme weather (heavy fog, heavy rain, blizzards), radio frequency signals are susceptible to interference and attenuation, leading to communication delays, interruptions, or signaling conflicts. In scenarios with no communication coverage or congestion, such as dense traffic flow, closed mining areas, and remote road sections, platooning systems completely fail, unable to establish platoons, follow vehicles, or maintain formation, making it difficult to address emergency transportation needs after autonomous driving malfunctions in extreme weather. Furthermore, V2X communication requires vehicles to be equipped with dedicated communication modules, resulting in high deployment costs and limiting the large-scale application of platooning technology. Meanwhile, traditional platooning distance measurement often relies on single-lens monocular vision or radar, which is susceptible to significant errors due to factors such as road slope, vehicle pitch, lighting changes, and extreme weather conditions. Furthermore, it lacks an effective dynamic correction mechanism. Fixed-distance control modes cannot adapt to the acceleration, deceleration, or turning intentions of the vehicle in front, nor can they compensate for decreased perception accuracy and delayed braking response in extreme weather. This leads to jerky following, insufficient safety redundancy, and a higher risk of accidents due to distance measurement errors in extreme weather. Existing platooning control largely depends on post-event feedback control, meaning that acceleration and deceleration commands are only executed after changes in the status of the vehicle in front are detected, resulting in significant control lag. It lacks the ability to anticipate the driving intentions of the vehicle in front and predict the relative motion trends of the two vehicles. In extreme weather, with slippery roads and extended braking distances, control lag can lead to following too closely and delayed deceleration, increasing the risk of rear-end collisions. It also reduces the smoothness of platooning, impacting the driving experience and transportation efficiency. Therefore, improving the accuracy of vehicle platooning control has become a significant technical challenge.

[0021] Based on this, this application provides a vehicle platooning control method. By using the vehicle's original lights as a communication medium to achieve platooning handshake, adaptive fusion of multi-dimensional visual features to lock onto the preceding vehicle, dynamic weighted ranging correction based on monocular vision and multi-source complementary observations, and intention-driven predictive closed-loop control, this method systematically solves key problems of existing V2X-dependent platooning technologies, such as communication interruption in extreme weather, target confusion in multi-vehicle scenarios, ranging inaccuracies caused by slope / lighting, and control lag leading to jerking and rear-end collision risks. This method improves the accuracy of vehicle platooning control.

[0022] Please see Figure 1 , Figure 1 A flowchart illustrating a vehicle platooning control method provided in an embodiment of this application. Figure 1 As shown in the figure, the vehicle platooning control method provided in this application embodiment includes: S101: The lead vehicle sends a formation request signal. After detecting the formation request signal, the following vehicles send back a formation handshake signal. After the lead vehicle detects the formation handshake signal, the vehicle formation is completed.

[0023] In this step, the lead vehicle sends a formation request signal, and the following vehicles send back a formation handshake signal after detecting the formation request signal. The lead vehicle completes the vehicle formation after detecting the formation handshake signal.

[0024] Here, after the user triggers the formation command through the lead vehicle's infotainment interface, the lead vehicle sends a formation request signal outward with a preset rhythm of flashing lights or audio prompts, and waits for the following vehicles to respond. The operator moves the following vehicle to a position in the same direction as the lead vehicle and at a distance of about 1 meter, and enters formation mode through the following vehicle's infotainment interface; the following vehicle detects the lead vehicle's lights / audio signals in real time, and after recognizing the formation request, it sends a formation handshake signal to the lead vehicle with a preset rhythm of flashing lights or audio prompts. After the lead vehicle detects this handshake signal through sensors such as a rearview camera, the vehicles in front and behind complete the formation.

[0025] In one possible implementation, after the lead vehicle detects the formation handshake signal and completes vehicle formation, the vehicle formation control method further includes: after completing formation, performing adaptive weight allocation and feature-level fusion based on static spatial features, dynamic temporal change features, and light intensity distribution features to generate a unique vehicle light feature ID to lock onto the lead vehicle.

[0026] Here, the following vehicle uses a forward-facing camera to extract the static spatial features, dynamic temporal change features, and light intensity distribution features of the preceding vehicle in real time. The following vehicle performs adaptive weight allocation and feature-level fusion on the three types of features to generate a unique vehicle light feature ID, thus completing the identification of the preceding vehicle.

[0027] Among them, static spatial characteristics include the taillight installation position, shape outline and physical dimensions; dynamic temporal characteristics include the light flashing frequency, on / off sequence and combination mode; and light intensity characteristics include the brightness of the parking lights, the brightness gradient of the brake lights and the uniformity of light emission.

[0028] It should be noted that the system can dynamically adjust the weights of three types of features based on real-time environmental conditions (lighting conditions, traffic flow, and occlusion): increasing the weight of static spatial features in good lighting conditions; increasing the weight of light intensity features in nighttime or backlit scenes; and increasing the weight of dynamic temporal features in congested or multi-vehicle similar scenarios, thereby maintaining stable target recognition and tracking under complex conditions. If it is necessary to further expand the platoon, the user can trigger the platooning button again on the lead vehicle's operating interface, allowing the last vehicle in the already platooned queue to send platooning requests to accept subsequent vehicles. During platooning, the lead vehicle's interface displays platooning status information in real time, including the number of vehicles in platooning and their platooning sequence number. After all vehicles have completed platooning, the user confirms the completion of platooning on the lead vehicle's operating interface, and the vehicle system enters platooning control mode.

[0029] S102: The following vehicle uses a forward-facing camera to capture real-time images of the taillights of the vehicle in front. Based on the principle of monocular vision and combined with a multi-feature fusion ranging error dynamic correction model, a high-precision real-time relative distance is determined.

[0030] In this step, the following vehicle uses a forward-facing camera to capture real-time images of the taillights of the vehicle in front. Based on the principle of monocular vision and combined with a multi-feature fusion ranging error dynamic correction model, a high-precision real-time relative distance is determined.

[0031] In one possible implementation, the ranging error dynamic correction model based on monocular vision principles and combined with multi-feature fusion, which determines the high-precision real-time relative distance, includes: A: The image of the front vehicle's taillights is processed based on the principle of monocular vision to determine the basic relative distance.

[0032] Here, the image of the taillights of the vehicle in front is processed according to the principle of monocular vision to determine the basic relative distance.

[0033] In one possible implementation, the step of processing the image of the front vehicle's taillights based on the monocular vision principle to determine the real-time relative distance between the front and rear vehicles includes: a: Perform image preprocessing and ROI constraint detection on the original image.

[0034] Here, the rear-view camera performs real-time distortion correction on the original image, eliminating pixel coordinate offsets caused by radial and tangential lens distortion, and improving the accuracy of taillight position and size calculation. To reduce computational load and minimize interference, the rear-view camera applies vertical ROI constraints to the image, performing taillight detection and feature calculation only on the road surface and vehicle area in the vertical direction of the image (30% to 100%), while removing irrelevant areas such as the sky, streetlights, and roadside billboards.

[0035] b: First, perform coarse screening of individual taillights by area screening and shape ratio screening. Then, match the left and right taillights in pairs based on the symmetry, height, size consistency and motion synchronization of the left and right taillights, and output the coordinates of the center points of the left and right taillights. The imaging pixel spacing is determined based on the coordinates of the center points of the left and right taillights.

[0036] Here, area filtering retains regions with pixel areas ranging from 20 to 5000, covering the taillight imaging scale at following distances of 10 to 100 meters. Shape ratio filtering limits the aspect ratio of the bounding rectangle of the connected region to between 0.3 and 3.0, eliminating irregular and elongated areas. Left and right taillight pairing is performed based on the symmetry, height consistency, size consistency, and motion synchronization of the left and right taillights of the same vehicle. The matching rules are as follows: the vertical coordinate difference of the center point does not exceed 5 pixels, ensuring consistent horizontal height, area difference ≤15%, aspect ratio difference ≤20%, and similar size and shape. After matching, the coordinates of the center points of the left and right taillights of the same preceding vehicle are output, and their imaging pixel spacing is calculated.

[0037] c: Based on a similar triangle model, the basic relative distance is calculated using the actual physical distance between the left and right taillights of the vehicle in front, the focal length of the camera, and the pixel spacing of the imaging.

[0038] Here, the basic relative distance is determined using the following formula:

[0039] in, Based on relative distance, This represents the actual physical distance between the taillights; The focal length of the rear-view camera; The pixel spacing for the left and right taillight images.

[0040] B: The dynamic correction model for ranging error based on multi-feature fusion corrects the real-time relative distance between the front and rear vehicles, and determines a high-precision real-time relative distance.

[0041] To eliminate system ranging bias caused by road slope, vehicle pitch, and changes in lighting, this application constructs multiple sets of complementary observations and achieves dynamic correction of the distance between the front and rear vehicles through adaptive weighted fusion.

[0042] In one possible implementation, the multi-feature fusion-based ranging error dynamic correction model corrects the real-time relative distance between the front and rear vehicles to determine a high-precision real-time relative distance, including: (1): The basic observation value of the taillight spacing is determined based on the basic relative distance.

[0043] (2): Based on the inverse square relationship between the physical area of ​​a single taillight and the area of ​​the imaging pixels, the observed value of the taillight area is determined.

[0044] Here, based on the inverse square relationship between the physical area of ​​a single taillight and the area of ​​its imaging pixels, the value is insensitive to vehicle pitch / slope and represents the core stable observation value under harsh operating conditions: ,in This refers to the actual physical area of ​​the taillights. The pixel area of ​​the taillight. This represents the observed area of ​​the taillights.

[0045] (3): The brightness gradient correction observation value is determined by compensating the basic observation value of the taillight spacing based on the brightness gradient correction coefficient.

[0046] Here, for errors caused by changes in illumination, a brightness gradient correction coefficient is used. For basic observations Provide compensation: ,in This represents the taillight brightness gradient for the current frame. The reference gradient under standard illumination. The observed values ​​are corrected for the brightness gradient.

[0047] (4): Based on the brake light status correction coefficient, the basic observation value of the taillight spacing is compensated to determine the brake light status correction observation value.

[0048] Here, regarding the dimensional deviation caused by changes in the taillight's illumination state, a state correction coefficient is used. compensate: Brake light status The actual vehicle calibration value is 0.92-0.98. Correct the observed values ​​for the status of the brake lights.

[0049] (5): The high-precision real-time relative distance is determined by dynamically allocating the weights of each group of observations according to the real-time working conditions through an adaptive weighted fusion method, and then summing them after normalization.

[0050] Here, the high-precision real-time relative distance is determined using the following formula. :

[0051] in, , , as well as To correct the normalized fusion weights corresponding to the observations.

[0052] S103: Based on the driving intention of the preceding vehicle and the high-precision real-time relative distance, dynamically adjust the following distance to perform platooning and following control; wherein, the driving intention of the preceding vehicle is identified based on the timing characteristics of the preceding vehicle's lights.

[0053] In this step, the following distance is dynamically adjusted based on the driving intention of the vehicle in front and the high-precision real-time relative distance to perform platooning and following control.

[0054] In one possible implementation, the following distance is dynamically adjusted using the following formula: ; in, The adjusted following distance, This refers to the high-precision real-time relative distance. Correction factor for driving intention. This is the environmental adaptive coefficient.

[0055] It should be noted that after identifying the driving intention of the vehicle in front based on the timing characteristics of the vehicle's lights in real time, a corresponding correction coefficient is applied to respond to the vehicle's acceleration, deceleration, lane change, turning, and other actions, thereby achieving dynamic adaptation of the distance to the vehicle. The coefficient value is dynamically adjusted based on the type of intent of the vehicle in front; the more urgent the intent, the larger the coefficient. This is an environmental adaptive coefficient used to compensate for the impact of harsh environments on perception accuracy and braking response. It compensates for the delay in perception and braking response based on road condition complexity, lighting, weather, and other conditions. The more complex the environment, the larger the coefficient, ensuring safe following under complex operating conditions.

[0056] To avoid false detections in a single frame and distance jumps caused by noise, this application employs a two-stage filtering approach to ensure the stability of the output results: ① Outlier removal: A threshold for the rate of change of distance between adjacent frames is set (maximum relative speed ≤ 120 km / h, i.e., maximum change in a single frame ≤ 1.1 m). Outliers exceeding the threshold are directly removed and replaced with valid values ​​from the previous frame. ② Sliding window smoothing: A sliding window is used to remove the maximum and minimum values ​​within the window. The weighted average of the remaining 3 valid values ​​is then calculated to output the final smoothed real-time relative distance, balancing real-time performance and stability.

[0057] In one possible implementation, the vehicle platooning control method further includes the following steps: dynamically adjusting the following distance based on the preceding vehicle's driving intention and the high-precision real-time relative distance for platooning control. When the light features are briefly lost or obscured, the formation is maintained based on the last effective distance, speed information and IMU information. After the light features are restored, the following distance is smoothly corrected.

[0058] In this application, predictive distance adjustment is achieved based on the temporal change rate of the imaging characteristics of the taillights of the preceding vehicle, outputting acceleration and deceleration commands in advance to avoid the lag of traditional control. By continuously collecting the pixel spacing of the taillights and calculating its change rate, the relative motion trend is determined: if the pixel spacing increases rapidly, it is determined that the two vehicles are rapidly approaching, and the system decelerates in advance; if the pixel spacing decreases, it is determined that the preceding vehicle is gradually moving away, and the system controls the vehicle to follow smoothly.

[0059] S104: In response to a preset exit trigger signal, control this vehicle to exit the formation or disband the entire formation.

[0060] This step supports active disengagement of turn signal triggering by the following vehicle, passive disengagement of hazard flasher triggering due to malfunction, and overall dissolution of combination light triggering by the leading vehicle.

[0061] This application provides a vehicle platooning control method, which includes: a lead vehicle sending a platooning request signal; subsequent vehicles sending a platooning handshake signal after detecting the platooning request signal; the lead vehicle completing platooning after detecting the platooning handshake signal; subsequent vehicles acquiring real-time images of the taillights of the preceding vehicle using a forward-looking camera; determining a high-precision real-time relative distance based on the principle of monocular vision and a multi-feature fusion-based distance measurement error dynamic correction model; dynamically adjusting the following distance based on the preceding vehicle's driving intention and the high-precision real-time relative distance for platooning and following control; wherein the preceding vehicle's driving intention is identified based on the timing characteristics of the preceding vehicle's lights; and responding to a preset exit trigger signal, controlling the vehicle to exit the platoon or disband the entire platoon. By using the vehicle's original lights as a communication medium to achieve formation handshake, adaptive fusion of multi-dimensional visual features to lock onto the preceding vehicle, dynamic weighted ranging correction based on monocular vision and multi-source complementary observations, and intention-driven predictive closed-loop control, the system systematically solves key problems of existing V2X-dependent formation technology, such as communication interruption in extreme weather, target confusion in multi-vehicle scenarios, ranging inaccuracy caused by slope / lighting, and control lag leading to jerking and rear-end collision risks, thereby improving the accuracy of vehicle formation control.

[0062] Please see Figure 2 , Figure 3 , Figure 2 This is one of the structural schematic diagrams of a vehicle platooning control device provided in the embodiments of this application; Figure 3 This is a second schematic diagram of a vehicle platooning control device provided in an embodiment of this application. Figure 2 As shown, the vehicle platooning control device 200 includes: The formation establishment module 210 is used for the lead vehicle to send a formation request signal, the following vehicles to send back a formation handshake signal after detecting the formation request signal, and the lead vehicle to complete the vehicle formation after detecting the formation handshake signal. The distance calculation module 220 is used by the following vehicle to collect the taillight image of the preceding vehicle in real time through the forward-looking camera. Based on the monocular vision principle and combined with the multi-feature fusion distance measurement error dynamic correction model, it determines the high-precision real-time relative distance. The formation control module 230 is used to dynamically adjust the following distance based on the driving intention of the preceding vehicle and the high-precision real-time relative distance for formation following control; wherein, the driving intention of the preceding vehicle is obtained by recognizing the timing characteristics of the preceding vehicle's lights. Formation exit module 240 is used to control the vehicle to exit the formation or disband the entire formation in response to a preset exit trigger signal.

[0063] Furthermore, the formation creation module 210 is also used for: After formation is completed, adaptive weight allocation and feature-level fusion are performed using static spatial features, dynamic temporal change features, and light intensity distribution features to generate a unique vehicle light feature ID to lock onto the vehicle in front.

[0064] Furthermore, the distance calculation module 220 is used to determine the high-precision real-time relative distance using the ranging error dynamic correction model based on monocular vision principles and combined with multi-feature fusion. The image of the front vehicle's taillights is processed based on the principle of monocular vision to determine the basic relative distance; The dynamic correction model for ranging error based on multi-feature fusion corrects the real-time relative distance between vehicles and determines a high-precision real-time relative distance.

[0065] Furthermore, the distance calculation module 220 is used to process the image of the front vehicle's taillights based on the monocular vision principle to determine the real-time relative distance between the front and rear vehicles: Image preprocessing and ROI constraint detection are performed on the original image; First, a single taillight is coarsely screened by area and shape ratio. Then, the left and right taillights are matched in pairs based on the symmetry, height, size consistency and motion synchronization of the left and right taillights, and the coordinates of the center point of the left and right taillights are output. The pixel spacing of the imaging is determined based on the coordinates of the center points of the left and right taillights; Based on a similar triangle model, the basic relative distance is calculated using the actual physical distance between the left and right taillights of the preceding vehicle, the focal length of the camera, and the pixel spacing of the imaging.

[0066] Furthermore, the distance calculation module 220 is used to correct the real-time relative distance between the front and rear vehicles using the multi-feature fusion-based distance measurement error dynamic correction model, thereby determining a high-precision real-time relative distance: The basic observation value of the taillight spacing is determined based on the aforementioned basic relative distance; Based on the inverse square relationship between the physical area of ​​a single taillight and the area of ​​an imaging pixel, the observed value of the taillight area is determined. The brightness gradient correction observation value is determined by compensating the basic observation value of the taillight spacing based on the brightness gradient correction coefficient. The basic observation value of the taillight spacing is compensated based on the brake light status correction coefficient to determine the brake light status correction observation value. The high-precision real-time relative distance is determined by dynamically allocating the weights of each group of observations according to the real-time operating conditions through an adaptive weighted fusion method, and then summing the weights after normalization.

[0067] Furthermore, the formation control module 230 dynamically adjusts the following distance using the following formula: ; in, The adjusted following distance, This refers to the high-precision real-time relative distance. Correction factor for driving intention. This is the environmental adaptive coefficient.

[0068] Furthermore, such as Figure 3 As shown, the vehicle formation control device 200 also includes an IMU inertial processing module 250, which is used for: When the light features are briefly lost or obscured, the formation is maintained based on the last effective distance, speed information and IMU information. After the light features are restored, the following distance is smoothly corrected.

[0069] This application provides a vehicle platooning control device, comprising: a platooning establishment module, used for a lead vehicle to send a platooning request signal, and subsequent vehicles to send back a platooning handshake signal after detecting the platooning request signal, and the lead vehicle to complete the vehicle platooning after detecting the platooning handshake signal; a distance calculation module, used for subsequent vehicles to collect images of the taillights of the preceding vehicle in real time through a forward-looking camera, and to determine a high-precision real-time relative distance based on the principle of monocular vision and combined with a multi-feature fusion distance measurement error dynamic correction model; a platooning control module, used for dynamically adjusting the following distance based on the driving intention of the preceding vehicle and the high-precision real-time relative distance for platooning and following control; wherein the driving intention of the preceding vehicle is identified based on the timing characteristics of the preceding vehicle's lights; and a platooning exit module, used to control the vehicle to exit the platoon or disband the entire platoon in response to a preset exit trigger signal. By using the vehicle's original lights as a communication medium to achieve formation handshake, adaptive fusion of multi-dimensional visual features to lock onto the preceding vehicle, dynamic weighted ranging correction based on monocular vision and multi-source complementary observations, and intention-driven predictive closed-loop control, the system systematically solves key problems of existing V2X-dependent formation technology, such as communication interruption in extreme weather, target confusion in multi-vehicle scenarios, ranging inaccuracy caused by slope / lighting, and control lag leading to jerking and rear-end collision risks, thereby improving the accuracy of vehicle formation control.

[0070] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device 400 includes a processor 410, a memory 420, and a bus 430.

[0071] The memory 420 stores machine-readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, they can perform the operations described above. Figure 1 The steps of the vehicle platooning control method in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0072] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 1 The steps of the vehicle platooning control method in the illustrated method embodiment can be found in the method embodiment for specific implementation, and will not be repeated here.

[0073] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0074] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0075] 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.

[0076] In addition, 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.

[0077] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0078] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, 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 covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A vehicle platooning control method, characterized in that, The vehicle platooning control method includes: The lead vehicle sends a formation request signal, and the following vehicles send back a formation handshake signal after detecting the formation request signal. The lead vehicle completes the vehicle formation after detecting the formation handshake signal. The following vehicle uses a forward-facing camera to capture images of the taillights of the vehicle in front in real time. Based on the principle of monocular vision and combined with a multi-feature fusion ranging error dynamic correction model, a high-precision real-time relative distance is determined. Based on the driving intention of the vehicle in front and the high-precision real-time relative distance, the following distance is dynamically adjusted for platooning and following control; wherein, the driving intention of the vehicle in front is obtained by recognizing the timing characteristics of the vehicle in front's lights. In response to a preset exit trigger signal, control the vehicle to leave the formation or disband the entire formation.

2. The vehicle platooning control method according to claim 1, characterized in that, After the lead vehicle detects the formation handshake signal and completes vehicle formation, the vehicle formation control method further includes: After formation is completed, adaptive weight allocation and feature-level fusion are performed using static spatial features, dynamic temporal change features, and light intensity distribution features to generate a unique vehicle light feature ID to lock onto the vehicle in front.

3. The vehicle platooning control method according to claim 1, characterized in that, The dynamic ranging error correction model based on monocular vision principles and combined with multi-feature fusion determines high-precision real-time relative distances, including: The image of the front vehicle's taillights is processed based on the principle of monocular vision to determine the basic relative distance; The dynamic correction model for ranging error based on multi-feature fusion corrects the real-time relative distance between vehicles and determines a high-precision real-time relative distance.

4. The vehicle platooning control method according to claim 3, characterized in that, The process of processing the image of the front vehicle's taillights based on the principle of monocular vision to determine the real-time relative distance between the front and rear vehicles includes: Image preprocessing and ROI constraint detection are performed on the original image; First, a single taillight is coarsely screened by area and shape ratio. Then, the left and right taillights are matched in pairs based on the symmetry, height, size consistency and motion synchronization of the left and right taillights, and the coordinates of the center point of the left and right taillights are output. The pixel spacing of the imaging is determined based on the coordinates of the center points of the left and right taillights; Based on a similar triangle model, the basic relative distance is calculated using the actual physical distance between the left and right taillights of the preceding vehicle, the focal length of the camera, and the pixel spacing of the imaging.

5. The vehicle platooning control method according to claim 3, characterized in that, The multi-feature fusion-based dynamic distance measurement error correction model corrects the real-time relative distance between the front and rear vehicles to determine a high-precision real-time relative distance, including: The basic observation value of the taillight spacing is determined based on the aforementioned basic relative distance; Based on the inverse square relationship between the physical area of ​​a single taillight and the area of ​​an imaging pixel, the observed value of the taillight area is determined. The brightness gradient correction observation value is determined by compensating the basic observation value of the taillight spacing based on the brightness gradient correction coefficient. The basic observation value of the taillight spacing is compensated based on the brake light status correction coefficient to determine the brake light status correction observation value. The high-precision real-time relative distance is determined by dynamically allocating the weights of each group of observations according to the real-time operating conditions through an adaptive weighted fusion method, and then summing the weights after normalization.

6. The vehicle platooning control method according to claim 1, characterized in that, The following formula can be used to dynamically adjust the following distance: ； in, The adjusted following distance, This refers to the high-precision real-time relative distance. Correction factor for driving intention. This is the environmental adaptive coefficient.

7. The vehicle platooning control method according to claim 1, characterized in that, In the process of dynamically adjusting the following distance based on the driving intention of the preceding vehicle and the high-precision real-time relative distance for platooning control, the vehicle platooning control method further includes: When the light features are briefly lost or obscured, the formation is maintained based on the last effective distance, speed information and IMU information. After the light features are restored, the following distance is smoothly corrected.

8. A vehicle platooning control device, characterized in that, The vehicle platooning control device includes: The formation establishment module is used for the lead vehicle to send a formation request signal, and the following vehicles to send back a formation handshake signal after detecting the formation request signal. The lead vehicle completes the vehicle formation after detecting the formation handshake signal. The distance calculation module is used by the following vehicle to collect the taillight image of the vehicle in front in real time through the forward-looking camera. Based on the monocular vision principle and combined with the multi-feature fusion distance measurement error dynamic correction model, it determines the high-precision real-time relative distance. The formation control module is used to dynamically adjust the following distance based on the driving intention of the preceding vehicle and the high-precision real-time relative distance for formation and following control; wherein, the driving intention of the preceding vehicle is obtained by identifying the timing characteristics of the preceding vehicle's lights. The formation exit module is used to control the vehicle to exit the formation or disband the entire formation in response to a preset exit trigger signal.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. The machine-readable instructions are executed by the processor to perform the steps of the vehicle platooning control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the vehicle platooning control method as described in any one of claims 1 to 7.

Images