Electric vehicle kinetic energy recovery system and method based on traffic signal lamp adaptive adjustment

Abstract

The invention discloses an electric vehicle kinetic energy recovery system and method based on traffic signal lamp adaptive adjustment. The system comprises a signal lamp sensing module, an environment sensing module, a driver intention algorithm module, a kinetic energy recovery decision module and a kinetic energy recovery execution module. Acquiring front traffic signal lamp data in real time through a camera, and analyzing to obtain the traffic signal lamp state of the current intersection; the distance between the front vehicles and the front stop line distance are obtained through a camera, a millimeter wave radar and a laser radar; judging and outputting a driver intention according to the driver behavior action; outputting a kinetic energy recovery strategy according to the current signal lamp state, the environment data and the intention of the driver; and the kinetic energy recovery action is executed according to the strategy output by the kinetic energy recovery decision module. According to front real-time traffic signal lamp information, the kinetic energy recovery force and time are adjusted in a self-adaptive mode, and therefore the purposes of efficiently recovering energy and improving the driving performance are achieved.

Description

Technical Field

[0001] The present invention relates to electric vehicle driving and energy management technology, and in particular to an electric vehicle kinetic energy recovery system and method based on traffic signal light adaptive regulation. Background Art

[0002] At present, the kinetic energy recovery system of electric vehicles usually operates according to preset rules, but does not fully consider the real-time road traffic conditions, especially the impact of traffic lights on vehicle deceleration and stopping. If the kinetic energy recovery strength can be accurately adjusted according to the remaining time of the red light when the vehicle approaches the red light, it can not only effectively improve the energy recovery rate, but also improve the driving experience and reduce unnecessary emergency braking.

[0003] The problems existing in the prior art are: Fixed-mode operation: Most current electric vehicle kinetic energy recovery systems use fixed control strategies, which fail to fully integrate real-time traffic conditions for dynamic adjustments, and especially ignore the impact of traffic light status on vehicle driving rhythm and deceleration process.

[0004] Limitations in energy recovery efficiency: Due to the lack of utilization of real-time red light countdown information, the kinetic energy recovery system fails to start at the most appropriate time and intensity, which may result in missing a large amount of energy recovery opportunities when stopping near a red light, reducing the overall energy recovery efficiency.

[0005] Driving comfort is affected: Under current technology, the kinetic energy recovery process of electric vehicles when facing red lights may not be smooth enough, causing unnecessary rapid deceleration or braking, which not only has a negative impact on the driver's comfort, but may also increase the wear of the braking system.

[0006] Insufficient intelligence: Existing technologies fail to fully utilize the data resources of the electric vehicle perception system to perform adaptive regulation in real time based on the red light countdown information of traffic lights, thus failing to fully tap the potential of the kinetic energy recovery system to improve overall driving performance and energy-saving effects. Summary of the invention

[0007] In order to solve the deficiencies in the prior art, the object of the present invention is to provide an electric vehicle kinetic energy recovery system and method based on adaptive adjustment of traffic lights.

[0008] To achieve the purpose of the present invention, the technical solution adopted by the present invention is: An electric vehicle kinetic energy recovery system based on adaptive regulation of traffic lights, comprising a traffic light perception module, an environment perception module, a driver intention algorithm module, a kinetic energy recovery decision module, and a kinetic energy recovery execution module; The traffic light sensing module uses the camera to obtain the real-time data of the traffic light ahead and analyzes the status of the traffic light at the current intersection; The environment perception module uses cameras, millimeter-wave radars, and lidars to obtain the distance to the vehicle in front and the distance to the stop line ahead; The driver intention algorithm module determines and outputs the driver's intention based on the driver's behavior; The kinetic energy recovery decision module outputs the kinetic energy recovery strategy based on the current traffic light status, environmental data and driver intention; The kinetic energy recovery execution module executes the kinetic energy recovery action according to the strategy output by the kinetic energy recovery decision module.

[0009] Furthermore, the implementation of the traffic light perception module: Collect the video stream of the traffic light ahead and convert it into a series of static images; pre-process the traffic light image, including enhancing the original image and locating the target; extract the color, shape and size features of the traffic light; train the classifier to identify the state of the traffic light; and output the recognition results to the upper-level application or control system.

[0010] Furthermore, the environment perception module recognizes the implementation of the stop line ahead: Collect images of the road ahead in real time and transmit them to the processing unit; perform image preprocessing, including using filters to remove noise in the image, converting RGB images into grayscale images, and adjusting the contrast and brightness of the image; perform edge detection and use the Canny edge detection algorithm to extract edge information in the image; use Hough transform to detect straight lines in the image; identify stop lines through color features, shape features or position constraints; once the stop line is detected, calculate the actual distance from the vehicle to the stop line based on the camera's internal parameters and the vehicle's position, and output the detection results to the upper-level application.

[0011] Furthermore, the environment perception module calculates the distance to the preceding vehicle: Perform sensor data collection, including camera image collection, millimeter-wave radar data collection, and lidar data collection; perform data preprocessing, including camera image processing, millimeter-wave radar signal processing, and lidar point cloud processing; perform target detection and tracking based on camera images, millimeter-wave radar, and lidar point clouds; perform distance calculation, including calculating the distance to the vehicle in front based on image ranging by the camera, calculating the distance to the vehicle in front based on direct ranging by the millimeter-wave radar, and calculating the distance to the vehicle in front based on the lidar point cloud ranging; perform multi-sensor data fusion, and output the calculated distance to the vehicle in front to the decision module of the autonomous driving system.

[0012] Furthermore, the driver intention algorithm implements: Collect vehicle body data, including steering wheel angle data; collect accelerator and brake pedal position data; collect gear-related data; extract vehicle status features, including steering wheel angle, accelerator and brake pedal position, vehicle speed, and acceleration; use historical data to train classifiers and map features to the driver's intentions; perform real-time predictions, input real-time collected sensor data into the trained model, and predict the driver's current intentions.

[0013] Furthermore, the kinetic energy recovery strategy includes: Based on the current traffic light status and switching time, evaluate whether the vehicle needs to slow down or stop; based on the distance to the vehicle ahead and traffic density information, evaluate whether the vehicle needs to take braking measures; based on the driver's behavior and vehicle status, predict the driver's next move.

[0014] Furthermore, the kinetic energy recovery performs the following operation: The sensor data collected in real time is input into the decision-making model to generate the corresponding kinetic energy recovery strategy; the generated control instructions are sent to the vehicle's braking system and energy management system to execute the kinetic energy recovery operation.

[0015] A method for recovering kinetic energy of an electric vehicle based on adaptive regulation of traffic lights comprises the following steps: (1) Obtain the traffic light data ahead in real time through the camera and analyze the traffic light status at the current intersection; (2) Obtain the distance to the vehicle in front and the distance to the stop line ahead through cameras, millimeter-wave radars, and lidars; (3) Determine and output the driver’s intention based on the driver’s behavior; (4) Output kinetic energy recovery strategy based on current traffic light status, environmental data and driver intention; (5) Execute kinetic energy recovery actions according to the strategy output by the kinetic energy recovery decision module.

[0016] The beneficial effects of the present invention are that, compared with the prior art, the present invention has the following advantages: Improve energy recovery efficiency: The system can obtain traffic light information in real time, and accurately calculate the best time and intensity of kinetic energy recovery based on factors such as the vehicle's current speed and distance from the red light, significantly improving the energy recovery rate of electric vehicles before parking.

[0017] Optimize driving experience: Through adaptive adjustment of kinetic energy recovery, abrupt deceleration caused by sudden increase in kinetic energy recovery is effectively avoided, so that the vehicle can achieve a smoother braking process when approaching a red light, greatly improving driving comfort.

[0018] Reduce brake system wear: When the vehicle approaches a red light, intelligent kinetic energy recovery replaces part of the mechanical braking, reducing the burden on the traditional brake system, thereby helping to reduce wear on brake system components and extend their service life.

[0019] Strengthening the integration of intelligent transportation: This invention integrates the information of the vehicle-mounted system and the intelligent traffic signal system, realizes the linkage between vehicles and traffic facilities, demonstrates the application of advanced vehicle networking technology, and provides strong support for future autonomous driving.

[0020] Energy saving and environmental protection: By optimizing the kinetic energy recovery strategy, the waste kinetic energy during the vehicle's deceleration process is effectively utilized and converted into electrical energy storage, further improving the overall energy efficiency of electric vehicles and helping to save energy and reduce emissions.

[0021] Improve driving safety: Intelligent kinetic energy recovery strategy helps the vehicle to decelerate smoothly in advance when approaching a red light, reducing the risk of traffic accidents caused by sudden braking, thereby enhancing driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a schematic diagram of an electric vehicle kinetic energy recovery system based on adaptive regulation of traffic lights according to the present invention; Figure 2 It is a schematic diagram of a practical scenario; Figure 3 This is a schematic diagram of practical scenario 2. DETAILED DESCRIPTION

[0023] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of this application.

[0024] like Figure 1 As shown, the electric vehicle kinetic energy recovery system based on adaptive adjustment of traffic lights described in the present invention includes a traffic light perception module, an environment perception module, a driver intention algorithm module, a kinetic energy recovery decision module, and a kinetic energy recovery execution module.

[0025] Traffic light perception module (module 1): obtains the traffic light data ahead in real time through the camera, and analyzes the traffic light status of the current intersection through the collected traffic light data information, such as whether the light is red for going straight, green for turning left, whether there is a red light for turning right, etc.

[0026] The implementation process of the traffic light perception module is as follows: 1. Collect traffic light data ahead; specifically, use a camera mounted on the vehicle (such as a monocular or binocular camera) to capture a video stream of the traffic light ahead. The position and angle of the camera need to be precisely calibrated to ensure that the image of the traffic light can be clearly captured. Convert the captured video stream into a series of static images, each of which represents a snapshot of the scene at a point in time.

[0027] 2. Preprocess the traffic light image; including enhancing the original image, such as adjusting brightness, contrast, and removing noise, to improve the accuracy of subsequent processing. Perform target positioning and locate the traffic light in the image through image processing techniques (such as edge detection, color segmentation, etc.). Common methods include color-based segmentation (specific wavelengths of red, yellow, and green light), shape matching (circles or arrows), etc.

[0028] 3. Extract the color, shape and size features of the traffic light; this can be achieved through methods such as color histogram and HSV color space.

[0029] 4. Train a classifier to identify the state of traffic lights; the training set should contain images of traffic lights under different lighting conditions and weather conditions. For some simple cases, you can also design rules to directly determine the state of traffic lights. For example, when a red area is detected and the area reaches a certain threshold, it can be determined as a red light.

[0030] 5. Output the recognition results (red light, yellow light, green light) to the upper-level application or control system. In order to ensure the reliability and security of the system, it is usually necessary to add additional security verification mechanisms, such as multi-frame continuous verification, to ensure that the final judgment result is accurate.

[0031] Environmental perception module (module 2): uses cameras, millimeter-wave radars, and lidars to obtain the distance to the vehicle in front, the distance to the stop line ahead, etc.

[0032] The implementation process of the environment perception module to identify the stop line ahead is as follows: 1. Collect images of the road ahead in real time. The camera is installed above the front windshield or on the bumper of the vehicle. The camera is used to capture images of the road ahead in real time and transmit them to the processing unit.

[0033] 2. Perform image preprocessing, including using filters to remove noise from the image and improve image quality. Convert RGB images to grayscale images or HSV color space for subsequent processing. Adjust the contrast and brightness of the image to make the stop line more obvious.

[0034] 3. Perform edge detection and use the Canny edge detection algorithm to extract edge information in the image. The Canny algorithm can detect strong and weak edges in an image and is suitable for identifying lines.

[0035] 4. Use Hough transform to detect straight lines in the image. Hough transform maps the straight lines in the image to parameter space and finds the most likely straight line through the accumulator matrix.

[0036] 5. Stop line recognition is performed through color features, shape features or position constraints. Specifically, color features: stop lines are usually solid or dotted lines in white or yellow, and possible stop line areas can be preliminarily screened out through color segmentation. Position constraints: According to the driving direction of the vehicle and the installation position of the camera, a reasonable search area is set. For example, the stop line is usually located in the center or right side of the road in front of the vehicle. Shape features: The stop line is usually a horizontal straight line, which can be further confirmed by detecting the direction and length of the straight line.

[0037] 6. To verify the results, in order to avoid false detection, stop line detection can be performed in multiple consecutive frames to ensure the consistency of the detection results. The detection results can also be further verified by combining the data of millimeter wave radar or lidar. For example, radar can provide distance information of obstacles ahead to help confirm whether the vehicle is close to the stop line.

[0038] 7. Once the stop line is detected, the actual distance from the vehicle to the stop line can be calculated based on the camera's internal parameters (focal length, pixel size, etc.) and the vehicle's position, and the detection results can be output to the upper-level application.

[0039] The process of calculating the distance between the front vehicle and the environment perception module is as follows: 1. Collect sensor data, including camera image acquisition, millimeter wave radar data acquisition, and lidar data acquisition; The camera is installed above the front windshield or on the bumper of the vehicle to ensure that it can clearly capture the image of the road ahead. The camera is used to capture the image of the road ahead in real time and transmit it to the processing unit.

[0040] Millimeter-wave radar is installed inside the front bumper of the vehicle and can penetrate adverse weather conditions such as rain and fog to provide long-range target detection capabilities. The radar obtains the distance, relative speed and angle information of the object in front by emitting electromagnetic waves and receiving reflected waves.

[0041] LiDAR is installed on the top or front of the vehicle and provides high-precision 3D point cloud data. It generates a 3D point cloud map of the environment ahead by emitting laser beams and measuring the reflection time.

[0042] 2. Perform data preprocessing; Camera image processing: Use filters to remove noise from camera images; convert RGB images to grayscale images or HSV color space for subsequent processing; adjust the contrast and brightness of the image to make the front vehicle more visible.

[0043] Millimeter-wave radar signal processing: Filter and remove clutter from the signal returned by the radar to extract useful target information.

[0044] LiDAR point cloud processing: downsample and filter point cloud data to reduce data volume while retaining key features.

[0045] 3. Perform target detection and tracking; Camera image object detection: Use deep learning models to identify the front vehicle in the image. Extract the front vehicle’s bounding box, color, shape and other features for subsequent tracking and distance calculation.

[0046] Millimeter wave radar target detection: Determine the distance, relative speed and direction of the target through information such as the Doppler effect and radar echo intensity. Use methods such as Kalman filters to track the detected targets to ensure continuous detection results.

[0047] LiDAR point cloud target detection: Use clustering algorithms or other machine learning methods to separate the front vehicle from the point cloud data. Use methods such as particle filters to track the detected targets to ensure continuous detection results.

[0048] 4. Perform distance calculation; Camera-based image ranging: Monocular cameras use the known camera internal parameters (focal length, pixel size, etc.) and the size change of the front vehicle to calculate the distance using the triangulation principle. Binocular cameras use the principle of stereo vision to calculate the depth information from the images taken by two cameras, and then obtain the distance to the front vehicle.

[0049] Millimeter-wave radar direct ranging: The radar directly calculates the distance to the vehicle in front by measuring the round-trip time of electromagnetic waves.

[0050] LiDAR point cloud ranging: Calculate the distance from the center point or the nearest point of the front vehicle to the vehicle through the point cloud distribution of the front vehicle in the point cloud data.

[0051] 5. Perform multi-sensor data fusion; Combining information from cameras, millimeter-wave radars, and lidars, the accuracy and robustness of target detection are improved through technologies such as Kalman filtering and particle filtering. A consistency check is also performed to ensure that target information detected by different sensors is consistent, eliminating false detections and missed detections.

[0052] 6. The calculated distance to the vehicle ahead is output to the decision module of the autonomous driving system for controlling the speed and braking of the vehicle. Based on the distance to the vehicle ahead and the current speed of the vehicle, it is decided whether to slow down, speed up, or maintain the current speed.

[0053] Driver intention algorithm module (module three): judges and outputs the driver's intention based on the driver's behavior: turning on the turn signal, releasing the accelerator pedal, pressing the accelerator pedal, pressing the brake pedal, adjusting the gear, etc.

[0054] The driver intention algorithm implementation process is as follows: 1. Collect vehicle body data, including steering wheel rotation angle data; accelerator and brake pedal position data; gear position related data.

[0055] 2. Extract vehicle status features, including steering wheel angle, accelerator and brake pedal positions, vehicle speed, acceleration, etc.

[0056] 3. Use historical data to train a classifier to map features to the driver’s intentions (such as changing lanes, accelerating, decelerating, stopping, etc.).

[0057] 4. Perform real-time predictions by inputting sensor data collected in real time into the trained model to predict the driver’s current intentions.

[0058] Combine data from multiple sensors and use multimodal fusion techniques (such as attention mechanism, weighted average, etc.) to improve prediction accuracy. Consider the current traffic environment and vehicle status, dynamically adjust the weight of the prediction model, and improve the robustness of the prediction.

[0059] 5. Decision-making and control; for example: judging that the driver wants to go straight based on the driver’s turn signal, steering wheel angle and historical driving habits; judging that the driver wants to turn left based on the driver’s turning on the left turn signal; judging that the driver wants to turn right based on the driver’s turning on the right turn signal.

[0060] Kinetic energy recovery decision module (module four): outputs kinetic energy recovery strategy based on the current traffic light status, environmental data and driver intention.

[0061] The kinetic energy recovery strategy is as follows: 1. Traffic light status assessment: Based on the current traffic light status and switching time, assess whether the vehicle needs to slow down or stop.

[0062] 2. Environmental assessment: Based on information such as the distance to the vehicle ahead and traffic density, assess whether the vehicle needs to take braking measures.

[0063] 3. Driver intention assessment: Based on the driver’s behavior and vehicle status, predict the driver’s next action (such as changing lanes, accelerating, decelerating, stopping, etc.).

[0064] 4. Strategy generation, including: Driving at low speed: When driving at low speed, you can appropriately increase the kinetic energy recovery force to improve energy recovery efficiency.

[0065] High-speed driving: When driving at high speed, appropriately reduce the kinetic energy recovery force to avoid affecting the stability and comfort of the vehicle.

[0066] Approaching a red light: When the vehicle approaches a red light, kinetic energy recovery is activated in advance to reduce the use of mechanical brakes and extend the life of the brake system.

[0067] Distance to vehicle ahead: Dynamically adjust kinetic energy recovery based on the distance and relative speed of the vehicle ahead to ensure a safe distance.

[0068] Driver Intention: Dynamically adjust the kinetic energy recovery strategy based on the driver's intention. For example, if the driver intends to slow down or stop, increase the kinetic energy recovery; if the driver intends to accelerate, reduce the kinetic energy recovery.

[0069] Kinetic energy recovery execution module (module five): executes kinetic energy recovery actions according to the strategy output by the kinetic energy recovery decision module.

[0070] The control operation process is as follows: the sensor data collected in real time is input into the decision model to generate the corresponding kinetic energy recovery strategy. The generated control instructions are sent to the vehicle's braking system and energy management system to execute the kinetic energy recovery operation. According to the actual response of the vehicle and environmental changes, the kinetic energy recovery strategy is dynamically adjusted to ensure the optimal energy utilization efficiency.

[0071] The present invention also provides an electric vehicle kinetic energy recovery method based on adaptive regulation of traffic lights, comprising the steps of: Includes steps: (1) Obtain the traffic light data ahead in real time through the camera and analyze the traffic light status at the current intersection; (2) Obtain the distance to the vehicle in front and the distance to the stop line ahead through cameras, millimeter-wave radars, and lidars; (3) Determine and output the driver’s intention based on the driver’s behavior; (4) Output kinetic energy recovery strategy based on current traffic light status, environmental data and driver intention; (5) Execute kinetic energy recovery actions according to the strategy output by the kinetic energy recovery decision module.

[0072] Practical scenario 1 like Figure 2As shown in the figure, the driver released the accelerator pedal and did not step on the brake pedal (in the sliding state), did not turn on the turn signal, and the steering wheel angle did not turn significantly. When the current vehicle speed is greater than the speed of kinetic energy recovery, the traffic light ahead is red, and the kinetic energy recovery strategy is decided.

[0073] 1. The formula for calculating the kinetic energy generated from the current vehicle speed to the vehicle stopping is: KE =½ mv² in, m is the mass of the vehicle in kilograms (kg); v is the initial velocity of the vehicle in meters per second (m / s).

[0074] 2. The formula for calculating the kinetic energy required for the vehicle to reach the stop line is: Friction = μmgS in, Friction is the work done by friction, which is also the total energy consumed by the vehicle, measured in joules (J); m is the coefficient of rolling friction; m is the mass of the vehicle in kilograms (kg); g is the acceleration due to gravity, which is about 9.8 m / s²; S is the distance the vehicle slides, in meters (m).

[0075] 3. The formula for calculating the kinetic energy that can be used for kinetic energy recovery of the vehicle is: Erecycle=½mv²-μmgS in, Erecycle is the total energy available for kinetic energy recovery.

[0076] Practical scenario 2 like Figure 3 As shown in the figure, the driver released the accelerator pedal and did not step on the brake pedal (in the coasting state), did not turn on the turn signal, and the steering wheel angle did not turn significantly. When the current vehicle speed is greater than the speed of kinetic energy recovery, the traffic light ahead is red and the red light countdown is less than 10 seconds, the kinetic energy recovery strategy is decided.

[0077] 1. The formula for calculating the kinetic energy generated from the current vehicle speed to the set vehicle speed for kinetic energy recovery is: KE =½ m(vi²- vf²) in, m is the mass of the vehicle in kilograms (kg); vi is the initial speed of the vehicle in meters per second (m / s); vf is the vehicle's kinetic energy recovery set speed, measured in meters per second (m / s).

[0078] 2. The formula for calculating the kinetic energy required for the vehicle to reach the stop line is: Fr = μrN

[0079] in, Fr is the rolling resistance (here it is the force that hinders the vehicle from moving forward during deceleration, that is, the friction force that converts kinetic energy into heat energy), μr is the coefficient of rolling friction, N is the support force perpendicular to the road surface. For horizontal roads, it is approximately equal to the gravity component of the vehicle (i.e., mass multiplied by gravitational acceleration g). Friction It represents the total amount of work done by friction, which represents the reduction in the kinetic energy of the system because friction plays the role of consuming kinetic energy in this process.

[0080] ∫ vivf is the integral symbol, indicating that we want to solve the speed from vi Change to vf The cumulative value of the work done by friction in this range. Fr is the rolling friction, which is related to the speed v The multiplication represents the power at each moment due to friction. v is the instantaneous velocity. In theory, the velocity function that changes with time should be considered in the integral v ( t ), but since it is not specified whether it is uniformly accelerated motion, it is abbreviated here as v . dt is the time element, which represents the accumulation of work done on friction as time changes.

[0081] 2. The formula for calculating the kinetic energy that can be used for kinetic energy recovery of the vehicle is: Erecycle= ½ m(vstart²- vfinal²)-Wfriction in, Erecycle is the total energy available for kinetic energy recovery.

[0082] The beneficial effects of the present invention are that, compared with the prior art, the present invention has the following advantages: Improve energy recovery efficiency: The system can obtain traffic light information in real time, and accurately calculate the best time and intensity of kinetic energy recovery based on factors such as the vehicle's current speed and distance from the red light, significantly improving the energy recovery rate of electric vehicles before parking.

[0083] Optimize driving experience: Through adaptive adjustment of kinetic energy recovery, abrupt deceleration caused by sudden increase in kinetic energy recovery is effectively avoided, so that the vehicle can achieve a smoother braking process when approaching a red light, greatly improving driving comfort.

[0084] Reduce brake system wear: When the vehicle approaches a red light, intelligent kinetic energy recovery replaces part of the mechanical braking, reducing the burden on the traditional brake system, thereby helping to reduce wear on brake system components and extend their service life.

[0085] Strengthening the integration of intelligent transportation: This invention integrates the information of the vehicle-mounted system and the intelligent traffic signal system, realizes the linkage between vehicles and traffic facilities, demonstrates the application of advanced vehicle networking technology, and provides strong support for future autonomous driving.

[0086] Energy saving and environmental protection: By optimizing the kinetic energy recovery strategy, the waste kinetic energy during the vehicle's deceleration process is effectively utilized and converted into electrical energy storage, further improving the overall energy efficiency of electric vehicles and helping to save energy and reduce emissions.

[0087] Improve driving safety: Intelligent kinetic energy recovery strategy helps the vehicle to decelerate smoothly in advance when approaching a red light, reducing the risk of traffic accidents caused by sudden braking, thereby enhancing driving safety.

[0088] The applicant of the present invention has made a detailed explanation and description of the implementation examples of the present invention in conjunction with the drawings in the specification. However, those skilled in the art should understand that the above implementation examples are only preferred implementation schemes of the present invention, and the detailed description is only to help readers better understand the spirit of the present invention, but not to limit the scope of protection of the present invention. On the contrary, any improvements or modifications based on the inventive spirit of the present invention should fall within the scope of protection of the present invention.

Claims

1. An electric vehicle kinetic energy recovery system based on adaptive regulation of traffic lights, characterized in that: It includes signal light perception module, environment perception module, driver intention algorithm module, kinetic energy recovery decision module and kinetic energy recovery execution module; The traffic light sensing module uses the camera to obtain the real-time data of the traffic light ahead and analyzes the status of the traffic light at the current intersection; The environment perception module uses cameras, millimeter-wave radars, and lidars to obtain the distance to the vehicle in front and the distance to the stop line ahead; The driver intention algorithm module determines and outputs the driver's intention based on the driver's behavior; The kinetic energy recovery decision module outputs the kinetic energy recovery strategy based on the current traffic light status, environmental data and driver intention; The kinetic energy recovery execution module executes the kinetic energy recovery action according to the strategy output by the kinetic energy recovery decision module.

2. The electric vehicle kinetic energy recovery system based on traffic light adaptive adjustment according to claim 1 is characterized in that: Implementation of the traffic light perception module: Collect the video stream of the traffic light ahead and convert it into a series of static images; pre-process the traffic light image, including enhancing the original image and locating the target; extract the color, shape and size features of the traffic light; Train the classifier to identify the state of traffic lights; output the identification results to the upper-level application or control system.

3. The electric vehicle kinetic energy recovery system based on traffic light adaptive adjustment according to claim 1 is characterized in that: Implementation of the environment perception module to identify the stop line ahead: Collect images of the road ahead in real time and transmit them to the processing unit; perform image preprocessing, including using filters to remove noise in the image, converting RGB images into grayscale images, and adjusting the contrast and brightness of the image; perform edge detection and use the Canny edge detection algorithm to extract edge information in the image; use Hough transform to detect straight lines in the image; identify stop lines through color features, shape features or position constraints; once the stop line is detected, calculate the actual distance from the vehicle to the stop line based on the camera's internal parameters and the vehicle's position, and output the detection results to the upper-level application.

4. The electric vehicle kinetic energy recovery system based on traffic light adaptive adjustment according to claim 1 is characterized in that: The environment perception module calculates the distance to the vehicle in front: Perform sensor data collection, including camera image collection, millimeter-wave radar data collection, and lidar data collection; perform data preprocessing, including camera image processing, millimeter-wave radar signal processing, and lidar point cloud processing; Perform target detection and tracking based on camera images, millimeter-wave radar, and lidar point clouds; perform distance calculations, including using the camera to calculate the distance to the vehicle in front based on image distance measurement, using the millimeter-wave radar to directly measure the distance to the vehicle in front, and using the lidar point cloud to measure the distance to the vehicle in front; perform multi-sensor data fusion and output the calculated distance to the vehicle in front to the decision module of the autonomous driving system.

5. The electric vehicle kinetic energy recovery system based on traffic light adaptive adjustment according to claim 1 is characterized in that: Driver Intent Algorithm Implementation: Collect vehicle body data, including steering wheel rotation angle data; accelerator and brake pedal position data; gear position related data; Extract vehicle status features, including steering wheel angle, accelerator and brake pedal positions, vehicle speed, and acceleration; use historical data to train a classifier to map features to the driver’s intent; Perform real-time predictions by inputting real-time collected sensor data into the trained model to predict the driver’s current intentions.

6. The electric vehicle kinetic energy recovery system based on traffic light adaptive adjustment according to claim 1 is characterized in that: Energy recovery strategies include: Based on the current traffic light status and switching time, evaluate whether the vehicle needs to slow down or stop; based on the distance to the vehicle ahead and traffic density information, evaluate whether the vehicle needs to take braking measures; based on the driver's behavior and vehicle status, predict the driver's next move.

7. The electric vehicle kinetic energy recovery system based on traffic light adaptive adjustment according to claim 1 is characterized in that: Kinetic energy recovery operation process: The sensor data collected in real time is input into the decision-making model to generate the corresponding kinetic energy recovery strategy; the generated control instructions are sent to the vehicle's braking system and energy management system to execute the kinetic energy recovery operation.

8. A method for kinetic energy recovery of electric vehicles based on adaptive regulation of traffic lights, based on the kinetic energy recovery system of electric vehicles based on adaptive regulation of traffic lights according to any one of claims 1 to 7, characterized in that: Includes steps: (1) Obtain the traffic light data ahead in real time through the camera and analyze the traffic light status at the current intersection; (2) Obtain the distance to the vehicle in front and the distance to the stop line ahead through cameras, millimeter-wave radars, and lidars; (3) Determine and output the driver’s intention based on the driver’s behavior; (4) Output kinetic energy recovery strategy based on current traffic light status, environmental data and driver intention; (5) Execute kinetic energy recovery actions according to the strategy output by the kinetic energy recovery decision module.

Images