A pedestrian and vehicle early warning and automatic braking control system adapted for new energy vehicles

Abstract

The application discloses a pedestrian and vehicle early warning and automatic braking control system suitable for a new energy vehicle, and relates to the technical field of active safety braking control of vehicles.The system comprises an image acquisition module, a target recognition and distance measurement module, a three-stage early warning judgment module, a hierarchical braking execution module and a multi-modal interaction module.The pedestrian and vehicle early warning and automatic braking control system suitable for the new energy vehicle sets three early warning thresholds, combines dynamic vehicle speed correction and inner wheel difference key detection mechanism, significantly improves the identification accuracy of collision risks in large vehicle blind areas, effectively avoids collision accidents caused by visual blind areas, adopts a hierarchical differentiated braking strategy, matches the dual architecture of electric braking and hydraulic braking of the new energy vehicle, realizes gentle deceleration and brake pre-charge pressure through kinetic energy recovery in the yellow early warning stage, executes cooperative braking in the red early warning stage, and takes into account braking response speed and driving comfort, and meanwhile, brake energy can be recycled to improve the vehicle endurance performance.

Description

Technical Field

[0001] This invention relates to the field of active safety braking control technology for vehicles, specifically to a pedestrian and vehicle warning and automatic braking control system adapted to new energy vehicles. Background Technology

[0002] With the increasing popularity of new energy commercial vehicles, the risk of side blind spot collisions in large new energy trucks and buses is becoming increasingly prominent. Due to the limitations of their vehicle structure, large vehicles have large blind spots on the left and right sides and in the inner wheel difference area when turning. Drivers cannot detect pedestrians and non-motorized vehicles in these blind spots in time, which can easily lead to serious accidents such as side collisions and running over pedestrians while turning. Existing blind spot warning systems are mostly based on single audible and visual alerts, which can only convey warning information to the driver and cannot actively intervene in braking control. Moreover, the warning thresholds are mostly fixed values ​​and are not dynamically adjusted according to vehicle speed and steering conditions, resulting in significant deficiencies in the timeliness and adaptability of the warnings.

[0003] Meanwhile, existing vehicle automatic braking systems are generally designed for forward collision scenarios, prioritizing maximizing deceleration as the control objective, without considering the detection distance limitations in side blind spot scenarios and the differences in the physical characteristics of the cargo. For large vehicles transporting rigid, heavy-duty cargo such as steel coils and heavy structural steel, the violent inertial impact generated by emergency braking can easily cause cargo to slip, tilt forward, or even penetrate the cab, leading to secondary accidents that are more serious than the collision itself. Existing braking control strategies neither dynamically adjust the braking deceleration curve according to the type of cargo nor optimize the intervention timing based on the distance characteristics of side visual detection, making it difficult to simultaneously address side blind spot collision avoidance and cargo transportation safety. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a pedestrian and vehicle warning and automatic braking control system adapted to new energy vehicles, solving the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution: a pedestrian and vehicle warning and automatic braking control system adapted to new energy vehicles, including an image acquisition module, a target recognition and ranging module, a three-level warning judgment module, a graded braking execution module and a multimodal interaction module, characterized in that: the image acquisition module is integrated into the electronic rearview mirrors on the left and right sides of the vehicle, and is used to collect pedestrian and vehicle image data from the side and rear in real time. The target recognition and ranging module is used to classify and identify targets and calculate the lateral distance and relative motion parameters between the target and the vehicle body; The three-level early warning determination module has built-in red, yellow and green three-level lateral distance thresholds, which are used to dynamically adjust the thresholds in combination with the vehicle's real-time driving parameters, and to determine the early warning level and braking intervention timing in combination with the collision time and cargo type. The graded braking execution module is communicatively connected to the electric braking system and hydraulic braking system of new energy vehicles, and has a built-in cargo-adaptive braking unit for outputting differentiated braking deceleration curves according to the warning level and cargo type. The multimodal interaction module is used to synchronously project and display the early warning information on the screen and output multi-channel early warning signals.

[0006] Furthermore, the image acquisition module includes two sets of binocular camera modules integrated into the housings of the left and right electronic rearview mirrors, respectively. Each set of binocular camera modules is arranged along the front-rear direction of the vehicle body. The field of view of the two sets of binocular cameras together cover the side blind spot, the rear side blind spot and the inner wheel difference area when turning, with an effective lateral distance measurement range of 0.5~10m.

[0007] Furthermore, the real-time vehicle driving parameters obtained by the three-level warning determination module include vehicle speed, steering wheel angle, and vehicle body yaw rate; the three-level lateral distance threshold increases with vehicle speed, decreases the warning trigger distance on the inside of the turn with the increase of steering wheel angle, and simultaneously increases the detection priority of the inside target.

[0008] Furthermore, the target recognition and ranging module calculates the initial lateral distance between the target and the vehicle body based on the principle of binocular vision parallax, and performs kinematic correction on the initial distance by combining the motion parameters output by the vehicle wheel speed sensor and steering angle sensor, eliminating static interference objects and falsely detected targets, and outputting the target's relative motion speed and trajectory direction.

[0009] Furthermore, the basic differentiated braking control commands of the graded braking execution module are as follows: the green level corresponds to the passage release state, and no braking intervention is triggered; the yellow level corresponds to the detection reminder state, triggers the hydraulic braking system to pre-charge, and at the same time calls the electric braking system of the new energy vehicle to perform mild kinetic energy recovery deceleration; the red level corresponds to the emergency danger state, triggers the electric braking system and the hydraulic braking system to work together, and performs automatic braking according to the corresponding deceleration curve.

[0010] Furthermore, the tactile warning signal of the multimodal interaction module is output through the vibration motor built into the driver's seat and steering wheel. The yellow warning corresponds to low-frequency intermittent vibration, and the red warning corresponds to high-frequency continuous vibration. The audible and visual warning signal is output through the in-vehicle buzzer and the instrument panel warning light.

[0011] Furthermore, the three-level early warning judgment module has a built-in inner wheel difference calculation unit, which is used to calculate the inner wheel difference trajectory area in real time based on the steering wheel angle, vehicle wheelbase and track parameters, and set this area as a high-priority detection area, and the overall early warning threshold in the area is reduced by 20%.

[0012] Furthermore, the cargo-adaptive braking unit has a pre-stored braking strategy library and collision time threshold for different cargo types, and will obtain the type and load parameters of the cargo on the vehicle; when the red warning triggers emergency braking, for rigid heavy-load cargo, the braking intervention time is advanced, and a constant deceleration uniform speed braking mode is adopted to limit the maximum deceleration and the rate of change of deceleration; during the braking process, it is linked with the vehicle controller (VCU) and battery management system (BMS), and prioritizes the use of electric braking for kinetic energy recovery. When the braking force is insufficient, hydraulic braking is superimposed to supplement it; when the lateral distance of the target is less than the extremely close safety threshold, the maximum braking force is executed to break through the cargo constraint.

[0013] Furthermore, the following implementation steps are included: S1. System initialization, complete the binocular camera calibration, vehicle CAN bus signal docking and cargo parameter input; S2. The dual-lens cameras of the left and right electronic rearview mirrors collect real-time side image data of the vehicle and simultaneously obtain real-time driving parameters of the vehicle. S3. Perform target detection and classification on the image data, identify pedestrians, motor vehicles and non-motor vehicles, and calculate the lateral distance and relative motion parameters between the target and the vehicle body based on binocular vision; S4. Correct the target lateral distance by combining the vehicle's real-time driving parameters, and dynamically calculate the red, yellow and green three-level lateral distance thresholds under the current working conditions. S5. Compare the corrected target lateral distance with the thresholds at each level, and combine the collision time prediction results with the cargo type to determine the warning level and braking intervention timing corresponding to the target. S6. Output the corresponding warning signal to the multimodal interaction module according to the warning level, and simultaneously project and display the target location and warning level on the screen; S7. Match the braking deceleration curve of the corresponding cargo type, and output the corresponding control command to the graded braking execution module in combination with the warning level to perform differentiated braking intervention. S8. Continuously monitor the target status. Once the target leaves the danger zone, gradually release the warning and braking intervention to restore normal driving status.

[0014] This invention provides a pedestrian and vehicle warning and automatic braking control system adapted to new energy vehicles, which has the following beneficial effects: 1. This pedestrian and vehicle warning and automatic braking control system adapted for new energy vehicles achieves accurate detection of targets in the side blind spot through the integration of binocular vision in the electronic rearview mirror. It sets three warning thresholds with the lateral distance between the target and the vehicle body as the core, and the threshold range matches the effective detection capability of the side camera. Combined with dynamic vehicle speed correction and inner wheel difference key detection mechanism, it can adapt to different vehicle speeds and turning conditions, significantly improve the recognition accuracy of collision risk in the blind spot of large vehicles, effectively avoid collision accidents caused by blind spots, and adopt a graded differentiated braking strategy, matching the dual architecture of electric braking and hydraulic braking of new energy vehicles. In the yellow warning stage, kinetic energy recovery is used to achieve smooth deceleration and brake pre-charging. In the red warning stage, coordinated braking is performed to balance braking response speed and driving comfort. At the same time, braking energy can be recovered to improve the vehicle's range performance.

[0015] 2. This pedestrian and vehicle warning and automatic braking control system adapted to new energy vehicles, through the built-in cargo-adaptive braking unit, adopts an adaptation scheme that combines early intervention timing with deceleration control, taking into account the characteristics of scenarios with limited lateral detection distance. For rigid heavy-duty cargo such as steel coils, it relaxes the collision time threshold to initiate braking earlier, while adopting a constant deceleration uniform braking mode to limit sudden deceleration. Within the limited detection distance, it achieves smooth deceleration, effectively avoiding secondary accidents such as cargo slippage and instability caused by inertial impact, and achieving dual protection of collision protection and cargo transportation safety.

[0016] 3. This pedestrian and vehicle warning and automatic braking control system adapted for new energy vehicles adopts a multi-modal linkage warning mechanism. The warning information is simultaneously covered by the electronic rearview mirror projection, the central control screen display, the sound and light prompts, and the seat tactile vibration. Different warning levels correspond to different feedback intensities, which can effectively reduce the probability of drivers missing warnings and improve the warning perception effect. In addition, the target recognition and ranging module combines kinematic parameters to correct the ranging results, which can effectively eliminate static interference targets and reduce the false alarm rate. At the same time, the system is based on the existing vehicle electronic rearview mirror and CAN bus architecture, with low hardware modification costs, strong adaptability, and easy to promote and apply in batches in the field of new energy commercial vehicles. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the steps of a pedestrian and vehicle warning and automatic braking control system adapted to new energy vehicles according to the present invention. Detailed Implementation

[0018] The embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0019] like Figure 1As shown, the present invention provides a technical solution: a pedestrian and vehicle warning and automatic braking control system adapted to new energy vehicles, specifically including the following implementation steps: S1. System Initialization and Parameter Calibration After the vehicle is powered on, the system completes self-check and initial configuration: It calibrates the internal and external parameters of the binocular camera modules in the left and right electronic rearview mirrors, establishes a mapping relationship between image pixel coordinates and the actual spatial coordinates of the vehicle body, calibrates installation deviations, and ensures that the lateral distance measurement accuracy error is ≤0.1m; it connects to the vehicle controller VCU14, steering angle sensor, wheel speed sensor, hydraulic braking system controller 10, motor controller, and battery management system BMS through the vehicle CAN bus module to complete communication protocol matching, and can acquire parameters such as vehicle speed, steering wheel angle, yaw rate, braking system status, and motor torque in real time; it loads basic parameters from the braking strategy library, including the maximum permissible deceleration, deceleration rate of change constraints, and collision time thresholds corresponding to different cargo types; it completes cargo parameter input, and the driver selects the current cargo type through the on-board terminal. The system synchronously reads the total weight data from the suspension load sensor to complete the initial matching of the cargo adaptation strategy.

[0020] S2. Image acquisition and driving parameter acquisition are synchronized. During vehicle operation, the system performs parallel data acquisition: the two sets of binocular camera modules in the left and right electronic rearview mirrors continuously acquire video images of the vehicle's sides and rear at a frame rate of 30fps. The left module covers the left blind spot and the inner wheel difference area for left turns, while the right module covers the right blind spot and the inner wheel difference area for right turns. The effective lateral detection distance range is 0.5~10m, covering the entire front and rear length of the vehicle. The system synchronously acquires real-time vehicle driving parameters at a frequency of 100Hz through the CAN bus module, including the current vehicle speed, steering wheel angle, vehicle yaw rate, and four-wheel wheel speed pulse signals.

[0021] S3. Target Recognition and Initial Lateral Ranging The target recognition and ranging module incorporates a lightweight deep learning target detection algorithm to perform real-time inference processing on each frame of image: it identifies three types of targets in the image: pedestrians, motor vehicles, and non-motor vehicles, outputs the pixel bounding box and category label of the target, and filters out invalid targets that are too small or located at the edge of the image; based on the principle of binocular vision parallax, it calculates the parallax value by matching the same targets in the left and right eye images, and calculates the initial lateral distance between the target and the side of the vehicle by combining the camera calibration parameters; by analyzing the target position changes in three adjacent frames, it calculates the lateral and longitudinal relative velocities of the target relative to the vehicle, and predicts the direction of the target's trajectory.

[0022] S4. Distance Correction and Dynamic Threshold Calculation This step is divided into two parts: target lateral distance correction and dynamic three-level threshold generation. Target distance correction: Combining vehicle wheel speed and steering angle data, the initial distance measurement results are compensated for errors through the vehicle kinematic model. When the vehicle turns, the distance offset caused by the vehicle body rotation is compensated. At the same time, through motion trajectory matching, static interference targets such as roadside guardrails and road signs are eliminated, and only valid targets that are in motion or whose trajectory is facing the vehicle body are retained. Finally, the corrected target lateral distance is output.

[0023] Dynamic Level 3 Threshold Calculation: The Level 3 Warning Judgment Module dynamically adjusts the Level 3 lateral distance threshold based on the current vehicle speed $$v$$ Low speed range of 0~30km / h: Yellow warning threshold 2.5m, red braking threshold 1.0m; Medium speed range of 30~60km / h: Yellow warning threshold 4.0m, red braking threshold 1.8m; For highway sections with speeds above 60 km / h: the yellow warning threshold is 5.5 m, and the red braking threshold is 2.5 m.

[0024] Steering condition ( Under these conditions, the warning threshold on the inside of the turn decreases as the turning angle increases. For every 10° increase in the turning angle, the yellow and red thresholds on the inside decrease by 10% simultaneously, with a minimum not lower than 80% of the straight-ahead threshold. Simultaneously, the inner wheel difference danger zone for the turn is calculated by the inner wheel difference calculation unit, based on the formula: Calculate the turning radius and the difference in width between the inner and outer wheels, where... This refers to the vehicle's wheelbase. This refers to the width of the vehicle body; The turning radius; The inner wheel difference width is used to draw a fan-shaped danger zone on the inside of the turn in real time based on the inner wheel difference width. The target detection priority in this area is increased to ensure that close targets can be captured in time when turning at low speed.

[0025] S5, Comprehensive Judgment of Warning Level The Level 3 warning determination module compares the corrected target lateral distance with the dynamic Level 3 threshold, and combines the collision time (TTC) and cargo type to comprehensively determine the warning level. Green level: The target's lateral distance is greater than the yellow threshold, and the collision time (TTC) is greater than the warning TTC threshold for the corresponding cargo. It is judged as a safe state and the vehicle is allowed to pass. Yellow level: If the red threshold is less than the target lateral distance and less than the yellow threshold, or the warning TTC threshold is greater than or equal to the TTC threshold and the braking TTC threshold, it is judged as a warning state and a corresponding detection reminder is issued; Red level: The target lateral distance is less than or equal to the red threshold, or the TTC is less than or equal to the braking TTC threshold. This is considered an emergency danger state, and an emergency automatic stop is triggered.

[0026] The corresponding TTC thresholds for different cargo types are as follows: for general cargo, the warning TTC is 3s and the braking TTC is 1.5s; for rigid heavy-load steel coils, the warning TTC is 4s and the braking TTC is 2.5s. By relaxing the TTC threshold for rigid heavy-load steel coils, braking can be initiated earlier without exceeding the lateral detection distance, allowing sufficient time for gradual braking. When the lateral distance of the target is less than 0.8m, which is very close to the safety threshold, it is directly classified as red, and the maximum braking force is applied beyond the cargo constraints. When multiple targets exist on the same side, the system takes the target with the highest risk level as the output basis and prioritizes the highest-level warning.

[0027] S6, Multimodal Early Warning Output and Projection Display After determining the warning level, the multimodal interaction module simultaneously executes multi-channel warning feedback: In terms of screen display, the target position and real-time lateral distance are marked with corresponding colored borders on the left and right electronic rearview mirror displays. During a yellow warning, the target frame flashes; during a red warning, a red warning bar appears at the edge of the screen. Simultaneously, the panoramic blind spot interface on the in-vehicle central control screen displays the distribution of targets around the vehicle. In terms of sound and light warning, a yellow warning triggers an intermittent buzzer sound, while a red warning triggers a continuous high-frequency warning sound, and the side warning lights on the dashboard illuminate simultaneously. In terms of tactile warning, during a yellow warning, the vibration motor on the corresponding side of the driver's seat triggers low-frequency intermittent vibration; during a red warning, the vibration motors on both the seat and steering wheel trigger continuous high-frequency vibration, enhancing the driver's perception through touch.

[0028] S7. Cargo Adaptation and Classification Braking Intervention Execution The graded braking execution module outputs control commands based on the warning level and cargo type, adapting to the dual braking architecture of new energy vehicles, which combines electric and hydraulic braking systems. Under the green level, no braking command is output, and the vehicle maintains normal driving status; At the yellow level, a pre-charge command is sent to the hydraulic braking system controller to establish basic pressure in the brake lines and shorten the response time of subsequent emergency braking; at the same time, a slight deceleration command is sent to the vehicle controller (VCU) to call the drive motor to perform kinetic energy recovery, generating a deceleration of 1~2 m / s², reducing the vehicle speed without affecting the stability of the cargo, and reserving more reaction time. At the red level, the cargo-adaptive braking unit first verifies the lateral distance to the target. If it is within the extremely close safety threshold, it directly activates the maximum braking force; otherwise, it executes the specific braking strategy for the corresponding cargo. Taking rigid heavy-duty cargo such as steel coils as an example, it executes a constant deceleration uniform speed braking mode, prioritizing the control of the drive motor to output a constant braking torque, combined with hydraulic brake fine-tuning, to keep the vehicle at a constant deceleration of 3m / s², with a deceleration change rate ≤1m / s³ throughout the braking process, without sudden impact. Combined with early braking intervention, it achieves smooth deceleration within a limited detection distance, preventing the steel coil from slipping or tilting forward due to inertia. During braking, the battery management system (BMS) monitors the battery charging power in real time to prevent the recovery power from exceeding the battery's capacity limit, while continuously tracking the target position. If the target actively avoids and moves away from the danger zone, it immediately and smoothly exits the braking.

[0029] S8. Warning Cancellation and Status Reset The system continuously tracks the target's position and movement. When the target leaves the danger zone, i.e., the lateral distance is greater than the yellow threshold and TTC is greater than the warning threshold, the system executes a step-by-step deactivation logic: first, the hydraulic brake is disengaged, and after a brief transition of electric braking kinetic energy recovery, the normal driving state is restored to avoid sudden deceleration affecting cargo stability; then, the audible, visual, and vibration warnings are turned off, the electronic rearview mirror display returns to the normal rearview view, and the central control screen exits the warning interface; finally, the system returns to normal detection mode and continues to collect and identify the next round of targets.

[0030] In summary, this pedestrian and vehicle warning and automatic braking control system adapted for new energy vehicles achieves accurate detection of targets in side blind spots through the integration of binocular vision in the electronic rearview mirror. It sets three warning thresholds based on the lateral distance between the target and the vehicle body, with the threshold range matching the effective detection capability of the side cameras. Combined with dynamic speed correction and inner wheel difference key detection mechanisms, it can adapt to different vehicle speeds and turning conditions, significantly improving the accuracy of identifying collision risks in blind spots for large vehicles and effectively avoiding collisions caused by blind spots. Furthermore, it adopts a graded differentiated braking strategy, matching the dual architecture of electric and hydraulic braking in new energy vehicles. During the yellow warning stage, kinetic energy recovery achieves smooth deceleration and brake pre-charging; during the red warning stage, coordinated braking is executed, balancing braking response speed and driving comfort, while also recovering braking energy to improve vehicle range.

[0031] With its built-in cargo-adaptive braking unit, this system addresses the challenges of limited lateral detection distances by employing an early intervention timing and deceleration control scheme. For rigid, heavy-duty cargo such as steel coils, it relaxes the collision time threshold to initiate braking earlier, while using a constant deceleration braking mode to limit sudden deceleration changes. This allows for smooth deceleration within the limited detection distance, effectively preventing cargo slippage and instability caused by inertial impacts, thus providing dual protection for collision safety and cargo transportation safety.

[0032] Employing a multimodal linkage early warning mechanism, the warning information is simultaneously covered by the electronic rearview mirror projection, central control screen display, audio and visual prompts, and seat tactile vibration. Different warning levels correspond to differentiated feedback intensities, which can effectively reduce the probability of drivers missing warnings and improve the warning perception effect. Furthermore, the target recognition and ranging module combines kinematic parameters to correct the ranging results, which can effectively eliminate static interference targets and reduce the false alarm rate. At the same time, the system is based on the existing vehicle electronic rearview mirror and CAN bus architecture, which has low hardware modification costs, strong adaptability, and is easy to promote and apply in batches in the field of new energy commercial vehicles.

[0033] The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A pedestrian and vehicle warning and automatic braking control system adapted to new energy vehicles, comprising an image acquisition module, a target recognition and ranging module, a three-level warning determination module, a graded braking execution module, and a multimodal interaction module, characterized in that: The image acquisition module is integrated into the electronic rearview mirrors on the left and right sides of the vehicle, and is used to collect real-time image data of pedestrians and vehicles on the side and rear. The target recognition and ranging module is used to classify and identify targets and calculate the lateral distance and relative motion parameters between the target and the vehicle body; The three-level early warning determination module has built-in red, yellow and green three-level lateral distance thresholds, which are used to dynamically adjust the thresholds in combination with the vehicle's real-time driving parameters, and to determine the early warning level and braking intervention timing in combination with the collision time and cargo type. The graded braking execution module is communicatively connected to the electric braking system and hydraulic braking system of new energy vehicles, and has a built-in cargo-adaptive braking unit for outputting differentiated braking deceleration curves according to the warning level and cargo type. The multimodal interaction module is used to synchronously project and display the early warning information on the screen and output multi-channel early warning signals.

2. The pedestrian and vehicle early warning and automatic braking control system adapted to new energy vehicles according to claim 1, characterized in that: The image acquisition module includes two sets of binocular camera modules integrated into the housings of the left and right electronic rearview mirrors, respectively. Each set of binocular camera modules is arranged along the front-rear direction of the vehicle body. The field of view of the two sets of binocular cameras together cover the side blind spot, the rear side blind spot and the inner wheel difference area when turning, with an effective lateral distance measurement range of 0.5~10m.

3. The pedestrian and vehicle early warning and automatic braking control system adapted to new energy vehicles according to claim 1, characterized in that: The real-time vehicle driving parameters obtained by the three-level early warning judgment module include vehicle speed, steering wheel angle, and body yaw rate; the three-level lateral distance threshold increases with vehicle speed, decreases the early warning trigger distance on the inside of the turn with the increase of steering wheel angle, and simultaneously increases the detection priority of the inside target.

4. The pedestrian and vehicle early warning and automatic braking control system adapted to new energy vehicles according to claim 1, characterized in that: The target recognition and ranging module calculates the initial lateral distance between the target and the vehicle body based on the principle of binocular vision parallax, and performs kinematic correction on the initial distance by combining the motion parameters output by the vehicle wheel speed sensor and steering angle sensor, eliminating static interference objects and false detection targets, and outputting the target's relative motion speed and trajectory direction.

5. A pedestrian and vehicle early warning and automatic braking control system adapted to new energy vehicles according to claim 1, characterized in that: The basic differentiated braking control commands of the graded braking execution module are as follows: the green level corresponds to the passage release state, which does not trigger any braking intervention; the yellow level corresponds to the detection and warning state, which triggers the hydraulic braking system to pre-charge and simultaneously calls the electric braking system of the new energy vehicle to perform mild kinetic energy recovery deceleration; the red level corresponds to the emergency danger state, which triggers the electric braking system and the hydraulic braking system to work together and perform automatic braking according to the corresponding deceleration curve.

6. A pedestrian and vehicle early warning and automatic braking control system adapted to new energy vehicles according to claim 1, characterized in that: The tactile warning signal of the multimodal interaction module is output through the vibration motor built into the driver's seat and steering wheel. The yellow warning corresponds to low-frequency intermittent vibration, and the red warning corresponds to high-frequency continuous vibration. The audible and visual warning signals are output through the in-vehicle buzzer and instrument panel warning lights.

7. A pedestrian and vehicle early warning and automatic braking control system adapted to new energy vehicles according to claim 3, characterized in that: The three-level early warning judgment module has a built-in inner wheel difference calculation unit, which is used to calculate the inner wheel difference trajectory area in real time based on the steering wheel angle, vehicle wheelbase and track parameters, and set this area as a high-priority detection area, and the overall early warning threshold in the area is reduced by 20%.

8. A pedestrian and vehicle early warning and automatic braking control system adapted to new energy vehicles according to claim 5, characterized in that: The cargo-adaptive braking unit has a pre-stored braking strategy library and collision time threshold for different cargo types, and it will obtain the type and load parameters of the cargo on the vehicle; when the red warning triggers emergency braking, for rigid heavy-load cargo, the braking intervention time is advanced, and a constant deceleration uniform speed braking mode is adopted to limit the maximum deceleration and the rate of change of deceleration; during the braking process, it is linked with the vehicle controller (VCU) and battery management system (BMS), and prioritizes the use of electric braking for kinetic energy recovery. When the braking force is insufficient, hydraulic braking is added to supplement it; when the lateral distance of the target is less than the extremely close safety threshold, the maximum braking force is executed to break through the cargo constraint.

9. A pedestrian and vehicle early warning and automatic braking control system adapted to new energy vehicles according to claim 1, characterized in that: The implementation steps include the following: S1. System initialization, complete the binocular camera calibration, vehicle CAN bus signal docking and cargo parameter input; S2. The dual-lens cameras of the left and right electronic rearview mirrors collect real-time side image data of the vehicle and simultaneously obtain real-time driving parameters of the vehicle. S3. Perform target detection and classification on the image data, identify pedestrians, motor vehicles and non-motor vehicles, and calculate the lateral distance and relative motion parameters between the target and the vehicle body based on binocular vision; S4. Correct the target lateral distance by combining the vehicle's real-time driving parameters, and dynamically calculate the red, yellow and green three-level lateral distance thresholds under the current working conditions. S5. Compare the corrected target lateral distance with the thresholds at each level, and combine the collision time prediction results with the cargo type to determine the warning level and braking intervention timing corresponding to the target. S6. Output the corresponding warning signal to the multimodal interaction module according to the warning level, and simultaneously project and display the target location and warning level on the screen; S7. Match the braking deceleration curve of the corresponding cargo type, and output the corresponding control command to the graded braking execution module in combination with the warning level to perform differentiated braking intervention. S8. Continuously monitor the target status. Once the target leaves the danger zone, gradually release the warning and braking intervention to restore normal driving status.

Images