Vehicle deceleration control method, device and equipment and storage medium

Abstract

The invention provides a vehicle deceleration control method, device and equipment and a storage medium, and the method comprises the steps: obtaining the current driving information and driving environment information of a vehicle, predicting an expected road traffic behavior in a future preset time period, and if a prediction result shows that deceleration needs to be carried out, carrying out the deceleration. If yes, determining target driving information which should be reached in a future preset time period according to the current driving information and the driving environment information, including a target driving speed, and generating a corresponding vehicle deceleration control instruction based on the determined target driving information, the strength of a kinetic energy recovery system and / or the strength of a mechanical brake of the vehicle are / is controlled according to the instruction, and it is ensured that the vehicle can be stably decelerated to the target speed according to a preset time node; the driving safety and comfort are effectively improved, and the energy use efficiency is optimized.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to a vehicle deceleration control method, device, equipment and storage medium. Background Art

[0002] With the continuous advancement of automotive technology, especially in the fields of intelligent driving and vehicle automation, the demand for improving driving safety and comfort is growing. The traditional driving method relies on the driver's immediate response to traffic conditions to operate the brake system, which may cause a series of problems in complex and changing road environments, such as driving fatigue, increased wear of the brake system, and potential safety risks. In modern traffic environments, especially in congested urban roads, frequent starting and stopping not only increases the driver's operating burden, but also may cause rapid wear of the brake system and increase the risk of traffic accidents. In addition, the kinetic energy generated during traditional braking is usually dissipated in the form of heat energy, which is not ideal for energy efficiency.

[0003] In recent years, with the popularity of electric vehicles (EV) and hybrid electric vehicles (HEV), kinetic energy recovery mechanisms or regenerative braking systems have become an important means of optimizing energy use. This system can convert part of the kinetic energy into electrical energy and store it in the battery when the vehicle decelerates, thereby reducing energy waste and extending the vehicle's range. However, how to efficiently integrate kinetic energy recovery with mechanical braking to achieve smooth and efficient deceleration control remains a challenge. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present application provides a vehicle deceleration control method, device, equipment and storage medium to solve the above-mentioned technical problems.

[0005] The present application provides a vehicle deceleration control method, the method comprising: obtaining current driving information and driving environment information of the vehicle, the current driving information being parameter information used to describe the current vehicle motion state, and the driving environment information being information about the driving scene in which the vehicle is located; predicting the expected road traffic behavior of the vehicle within a preset time period in the future based on the current driving information and the driving environment information; if the expected road traffic behavior includes vehicle deceleration, determining the target driving information of the vehicle within the preset time period in the future based on the current driving information and the driving environment information, the target driving information being parameter information used to describe the future motion state of the vehicle; generating a vehicle deceleration control instruction based on the target driving information, and controlling the strength of the vehicle's kinetic energy recovery system and / or the strength of the mechanical brake according to the vehicle deceleration control instruction, so that the vehicle decelerates to the target driving speed in the target driving information after reaching a preset time.

[0006] In one embodiment of the present application, the target driving information of the vehicle within a future preset time period is determined based on the current driving information and the driving environment information, including: taking the current driving information and the driving environment information as input information; inputting the input information into a trained vehicle driving information determination model to obtain an output result of the vehicle driving information determination model, and determining the output result as the target driving information; the vehicle driving information determination model includes: a driving state prediction sub-model, which is used to predict the expected road traffic behavior of the vehicle within a future preset time period based on the current driving information and the driving environment information; and a driving information generation sub-model, which is used to determine the target driving information of the vehicle within a future preset time period based on the current driving information and the driving environment information, and the output of the driving state prediction sub-model.

[0007] In one embodiment of the present application, the road traffic behavior of the vehicle within a preset time is predicted based on the current driving information and the driving environment information, including: based on the current driving speed and the current position of the vehicle, predicting the target position of the vehicle after the preset time, the current position being obtained based on the current driving information; identifying the road traffic status of the interval section between the current position and the target position, the road traffic status at least including a smooth state and a blocked state; if the road traffic status is a smooth state, the driving behavior of the vehicle through the interval section does not include deceleration; if the road traffic status is a blocked state, the driving behavior of the vehicle through the interval section includes deceleration.

[0008] In one embodiment of the present application, the target driving information includes a target speed and a target deceleration, and determining the target driving information of the vehicle within a preset time period in the future includes: identifying obstacles between the current position and the target position based on the environmental information, and recording the position and speed of each obstacle; wherein the obstacles include dynamic obstacles and stationary obstacles; calculating the interval distance between the vehicle and each obstacle based on the current position of the vehicle and the position of each obstacle, and calculating the expected collision time between the vehicle and each obstacle based on the interval distance, the speed of the obstacle, and the current driving speed; if the expected collision time is less than a preset safety time threshold, calculating the target speed based on the current speed and the interval distance to ensure that the vehicle stops before reaching the obstacle; and calculating the target deceleration to decelerate the vehicle to the target speed based on the current speed and the target speed.

[0009] In one embodiment of the present application, a vehicle deceleration control instruction is generated based on the target driving information, including: obtaining a deceleration threshold of a vehicle kinetic energy recovery system; if the target deceleration is less than or equal to the deceleration threshold of the kinetic energy recovery system, calculating the kinetic energy recovery intensity based on the target deceleration; if the target deceleration is greater than the deceleration threshold of the kinetic energy recovery system, calculating the deceleration difference between the deceleration threshold of the kinetic energy recovery system and the target deceleration, and determining the mechanical braking force based on the deceleration difference; converting the calculated mechanical braking force and kinetic energy recovery intensity into a control instruction, and performing deceleration control on the vehicle based on the control instruction to avoid a collision between the vehicle and an obstacle.

[0010] In one embodiment of the present application, the vehicle deceleration control method also includes: constructing an initial neural network model, wherein the initial neural network model is implemented based on a preset machine learning algorithm; training the initial neural network model based on historical driving information and corresponding driving environment information to generate a vehicle driving information determination model; inputting the current driving information into the vehicle driving information determination model to predict the road condition of the vehicle within a preset time period in the future; generating corresponding target driving information based on the predicted road condition, wherein the target driving information includes at least a target speed and a target deceleration; and gradually adjusting the strength of the mechanical brake and the strength of the kinetic energy recovery system to allow the vehicle to smoothly transition from the current driving state to the driving state indicated by the target driving information.

[0011] In one embodiment of the present application, after controlling the strength of the vehicle's kinetic energy recovery system and / or the strength of the mechanical brake according to the vehicle deceleration control instruction, it also includes at least one of the following: real-time monitoring of the vehicle's actual deceleration, including at least changes in vehicle speed, acceleration changes and the response of the braking system, and dynamically adjusting the braking force and / or kinetic energy recovery strength based on the difference between the monitored actual deceleration and the expected deceleration target to ensure that the actual deceleration process meets the expected deceleration curve; continuously evaluating changes in driving environment information during the deceleration process, such as changes in the position of obstacles ahead and sudden changes in road conditions, and dynamically adjusting the deceleration strategy based on the changes to ensure driving safety; when the vehicle reaches a preset target speed or stops completely, recording data from the entire deceleration process, including the deceleration control instructions issued, the actual deceleration effect and any adjustment measures, for optimizing the vehicle driving information determination model.

[0012] The present application provides a vehicle deceleration control device, the device comprising: an information acquisition module, used to acquire current driving information and driving environment information of the vehicle, the current driving information being parameter information used to describe the current vehicle motion state, and the driving environment information being information about the driving scene in which the vehicle is located; a target driving information generation module, used to predict the expected road traffic behavior of the vehicle within a future preset time period based on the current driving information and the driving environment information; if the expected road traffic behavior includes vehicle deceleration, the target driving information of the vehicle within a future preset time period is determined based on the current driving information and the driving environment information, the target driving information being parameter information used to describe the future motion state of the vehicle; a deceleration control module, used to generate a vehicle deceleration control instruction based on the target driving information, and to control the strength of the vehicle's kinetic energy recovery system and / or the strength of the mechanical brake according to the vehicle deceleration control instruction, so that the vehicle decelerates to the target driving speed in the target driving information after reaching a preset time.

[0013] The present application provides an electronic device, comprising a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the vehicle deceleration control method as described above.

[0014] The present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is used to enable a computer to execute the vehicle deceleration control method as described above.

[0015] Beneficial effects of the present application: The vehicle deceleration control method in the present application can accurately predict the road traffic behavior within a preset time period in the future by acquiring and analyzing the current driving information and driving environment information of the vehicle in real time. When the prediction result shows that deceleration is required, based on the current driving information and environmental information, a target driving information that should be achieved within a preset time period in the future is intelligently determined, especially the target driving speed, so as to generate corresponding vehicle deceleration control instructions, and accurately adjust the strength of the kinetic energy recovery system and / or the strength of the mechanical brake according to these instructions, to ensure that the vehicle can smoothly and efficiently decelerate to the target speed at a predetermined time node. This method not only reduces the driver's operating burden and fatigue caused by frequent braking, but also effectively slows down the wear of the brake system and extends its service life. More importantly, by rationally utilizing the kinetic energy recovery mechanism, the energy utilization efficiency is greatly improved, providing electric vehicles and hybrid vehicles with a longer range.

[0016] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0018] Figure 1 is a schematic diagram of an implementation environment of a vehicle deceleration control method shown in an exemplary embodiment of the present application;

[0019] Figure 2 is a flow chart of a vehicle deceleration control method shown in an exemplary embodiment of the present application;

[0020] Figure 3 is a schematic diagram of a vehicle driving information determination model training process shown in an exemplary embodiment of the present application;

[0021] Figure 4 is a schematic diagram of an overall scheme of a vehicle deceleration control method shown in an exemplary embodiment of the present application;

[0022] Figure 5 is a block diagram of a vehicle deceleration control device shown in an exemplary embodiment of the present application;

[0023] Figure 6 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0024] The following will describe the implementation methods of the present application with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, not for limiting the scope of protection of the present application.

[0025] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0026] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0027] First of all, it should be noted that the kinetic energy recovery mechanism is widely used as an innovative energy management technology. This mechanism, also known as a regenerative braking system, can convert kinetic energy that was originally dissipated in the form of heat energy into electrical energy during vehicle deceleration or braking, and store it in the battery for subsequent use. Specifically, when the driver steps on the brake pedal or releases the accelerator pedal, the action mode of the motor will change from driving mode to power generation mode. At this time, the motor works as a generator, converting rotational kinetic energy into electrical energy and feeding it back to the battery pack, which not only reduces energy waste but also extends the vehicle's range.

[0028] This application focuses on the discussion of how to obtain the target speed and acceleration of deceleration based on the prediction of future road conditions in the scenario of autonomous driving, so as to comprehensively adjust the strength of the vehicle's kinetic energy recovery system and the strength of the mechanical brake based on the target speed and target acceleration, so as to achieve vehicle deceleration based more on kinetic energy recovery and recover more mechanical kinetic energy, which is beneficial to energy recovery and improves energy utilization efficiency. In addition, energy recovery is achieved based on predicted road information, so that the vehicle has more reaction time, and the energy recovery intensity is adjusted according to the target deceleration, so that its energy recovery process is more accurate, efficient, and smoother. Therefore, not only the energy utilization efficiency is improved, but also the user's comfort and experience are improved.

[0029] DDPG (Deep Deterministic Policy Gradient) is a reinforcement learning algorithm that is particularly suitable for solving problems in continuous action spaces. It combines deep learning and traditional deterministic policy gradient methods.

[0030] ADAS system (Advanced Driver Assistance Systems) is a general term for a series of technologies and systems designed to improve vehicle safety and driving experience.

[0031] V2X (Vehicle to Everything) communication technology is a technology that enables vehicles to communicate with various entities in the surrounding environment, including other vehicles (V2V, Vehicle to Vehicle), infrastructure (V2I, Vehicle to Infrastructure), pedestrians (V2P, Vehicle to Pedestrian) and networks (V2N, Vehicle to Network). The purpose of V2X is to improve road safety, traffic efficiency and driving experience by exchanging information in real time.

[0032] Figure 1 It is a schematic diagram of an implementation environment of a vehicle deceleration control method shown in an exemplary embodiment of the present application.

[0033] like Figure 1 As shown, the implementation environment of the vehicle deceleration control method includes a data acquisition device 101 and a computer device 102. Among them, the data acquisition device 101 is responsible for collecting the current driving data and driving environment data during the operation of the vehicle, which data include but are not limited to the current position, current driving speed, navigation information, road conditions ahead, and the distance from the front vehicle. The current position can be determined by obtaining the real-time geographical location of the vehicle through GPS or other positioning systems; the current driving speed is obtained by the real-time speed information provided by the vehicle speed sensor or the vehicle-mounted system; the navigation information is generated from the route planning and road type information obtained from the vehicle-mounted navigation system or the external navigation service; the road conditions ahead are obtained through perception technologies such as vehicle-mounted cameras, radars or laser radars, and obstacles, traffic light status, road conditions, etc. can be identified; the distance from the front vehicle is measured by radar or ultrasonic sensors. Therefore, the data acquisition device can be various sensors installed on the vehicle (such as GPS modules, vehicle speed sensors, cameras, radars or ultrasonic sensors, etc.), or it can be an external device connected to the vehicle by wireless or wired means. This application does not impose any restrictions on the specific implementation form of the data acquisition device.

[0034] In addition, the computer device 102 is used to process the data collected by the data acquisition device 101, specifically including: by real-time acquisition and analysis of the current driving information and driving environment information of the vehicle, it can accurately predict the road traffic behavior within the future preset time period. When the prediction result shows that deceleration is required, based on the current driving information and environmental information, intelligently determine a target driving information that should be achieved within the future preset time period, especially the target driving speed, so as to generate corresponding vehicle deceleration control instructions, and accurately adjust the strength of the kinetic energy recovery system and / or the strength of the mechanical brake according to these instructions, to ensure that the vehicle can smoothly and efficiently decelerate to the target speed at a predetermined time node. Therefore, the computer device 102 can be a high-performance processor, a dedicated vehicle-mounted computing platform, or an embedded system integrated in the vehicle, etc. This application also does not impose any restrictions on the specific type of computer equipment.

[0035] It should be noted that through such an implementation environment configuration, the vehicle deceleration control system can effectively use the real-time collected data to make intelligent decisions, thereby improving driving safety and providing the driver with a more comfortable driving experience.

[0036] Figure 2 It is a flow chart of a vehicle deceleration control method shown in an exemplary embodiment of the present application.

[0037] like Figure 2 As shown, in an exemplary embodiment, the vehicle deceleration control method includes at least steps S210 to S230, which are described in detail as follows:

[0038] Step S210, obtaining the current driving information and driving environment information of the vehicle, wherein the current driving information is parameter information used to describe the vehicle's motion state, and the driving environment information is information about the driving scene in which the vehicle is located.

[0039] In one embodiment of the present application, current driving information is obtained, which is used to describe the current driving information of the vehicle's motion state, and is generally used to characterize driving-related information on the vehicle side, including but not limited to vehicle speed data, acceleration data, steering angle, accelerator pedal position, and brake pedal position. The data is collected in the following manner:

[0040] Vehicle speed data: Use the wheel speed sensor installed on the vehicle's wheel hub to monitor the vehicle speed in real time. For example, assume the current vehicle speed is 60km / h.

[0041] Acceleration data: The vehicle's longitudinal acceleration is measured by the built-in three-axis accelerometer. For example, the current acceleration is 0.5m / s 2 .

[0042] Steering angle: The driver's steering operation is obtained through the steering wheel angle sensor. For example, the current steering angle is +10 degrees (turn right).

[0043] Accelerator pedal position: The accelerator pedal position is monitored by the accelerator position sensor to understand the driver's need for acceleration. For example, the current accelerator pedal position is 30%.

[0044] Brake pedal position: The brake pedal position is monitored by the brake position sensor to determine the degree of braking. For example, the current brake pedal position is 0% (not braking).

[0045] In addition, the driving environment information used to describe the driving scene in which the vehicle is located is generally represented by environmental information in nature and environmental information related to human activities, including but not limited to the road conditions ahead, weather conditions, traffic flow, road condition information, and road surface type. The collection method of each information is as follows:

[0046] Road conditions ahead: Front camera: Use the front camera to capture images of the road ahead and use image processing algorithms to identify lane lines, traffic signs, and other obstacles. For example, a stop line is detected 50 meters ahead. Radar / LiDAR: Use radar or LiDAR to detect the distance and relative speed of obstacles ahead. For example, a stationary vehicle is detected 100 meters ahead.

[0047] Weather conditions: The current weather conditions are obtained through the on-board weather sensors or the weather forecast data received from the external service provider. For example, the current weather is sunny and the temperature is 25°C.

[0048] Traffic flow: Through V2X (Vehicle to Everything) communication technology, data from other surrounding vehicles and infrastructure is received to obtain real-time traffic flow information. For example, the traffic flow on the current road section is medium.

[0049] Get the latest traffic updates through the built-in navigation system, for example, the navigation system shows that there is a minor congestion 2 kilometers ahead.

[0050] Road surface type: The image captured by the front camera is analyzed through image processing technology to determine the type of road surface currently being driven on. For example, the current road surface is dry asphalt road surface.

[0051] It should be noted that after the above data information is collected, it is necessary to further perform data preprocessing and data fusion, so that the sorted data can be used for road condition prediction and target driving information determination. Similar to the current driving information, the target driving information is the parameter information used to describe the vehicle motion state, and its data category is the same as the above current driving information. The data preprocessing includes preprocessing the collected data, including filtering and denoising, unit conversion, etc., to ensure the quality of the data. Data fusion includes using multi-sensor data fusion algorithms, such as Kalman filters or other fusion algorithms, to combine information from different sources to form comprehensive driving information and environmental information.

[0052] Step S220, based on the current driving information and driving environment information, predict the expected road traffic behavior of the vehicle within a future preset time period; if the expected road traffic behavior includes vehicle deceleration, determine the target driving information of the vehicle within the future preset time period based on the current driving information and driving environment information.

[0053] In one embodiment of the present application, a vehicle deceleration control method includes constructing an initial neural network model, which is implemented based on a preset machine learning algorithm; training the initial neural network model based on historical driving information and corresponding driving environment information to generate a vehicle driving information determination model; inputting current driving information into the vehicle driving information determination model to predict the road condition of the vehicle within a preset time period in the future; generating corresponding target driving information based on the predicted road condition, the target driving information including at least a target speed and a target deceleration; and gradually adjusting the strength of the mechanical brake and the strength of the kinetic energy recovery system to allow the vehicle to smoothly transition from the current driving state to the driving state indicated by the target driving information.

[0054] Before building the neural network model, it is necessary to initialize the vehicle's related systems and obtain relevant parameters, as well as environmental perception and data fusion.

[0055] The relevant systems of the vehicle are initialized and relevant parameters are obtained, as follows:

[0056] When the vehicle is started, the intelligent braking system will first initialize itself, check and ensure that all sensors and actuators are in normal working condition. Then, the system will communicate with the vehicle's CAN bus to obtain key driving parameters in real time. These parameters include but are not limited to the current vehicle speed, acceleration, deceleration, steering wheel angle, gear information, and various status signs of the vehicle. In addition, the system will obtain location information in real time through the car map navigation. This information not only includes the current precise location of the vehicle, but also includes important data such as route planning, road type, traffic signs, etc. With this data, the system can determine whether the vehicle is about to enter an area that requires special attention, such as a school, a crosswalk, a sharp turn, or a construction area.

[0057] Environmental perception and data fusion are as follows:

[0058] During the driving process, the intelligent braking system will continue to use multiple sensors to perceive the environment. The driving camera is responsible for capturing visual information in front of the vehicle and identifying road signs, pedestrians and other obstacles. Millimeter wave radar, ultrasonic radar and lidar provide accurate data on the distance, speed and direction of the objects in front.

[0059] Different data fusion strategies will produce different results. Here we mainly introduce the fusion strategy based on complementary features. Because different sensors can capture non-correlated size features of the same target, this provides better recognition capabilities for target recognition and detection. The features extracted in the ADAS system include target parameters and feature data extraction.

[0060] Target parameter extraction: includes target information such as size, distance, direction, speed and acceleration of the target extracted from the preprocessed data. Many studies extract the position features of radar or lidar targets and assist image recognition by generating regions of interest (ROIs). ROIs directly convert the position of radar-detected targets into images to form regions.

[0061] Data feature extraction: Data features are features extracted from images or other processed data, such as target contours, textures, time-frequency features, and color distribution, for classification and recognition. In computer vision, a large number of regions of interest (ROIs) that may contain targets are usually generated in the image, and these ROIs are classified by pre-trained classification models. In addition, the ROI with the highest confidence is where the target is located. Determining the location of the target in this way requires a lot of calculations. Due to the advantages of lidar and millimeter-wave radar in detecting the location of the target, the amount of calculation is relatively small. Therefore, many studies first use radar and lidar to extract the distance and orientation information of the target, and then map the location information into the image data to generate fewer ROIs. Finally, the pre-trained model is used to further identify these regions of interest and accurately classify the target category. After extracting the ROI, many apply machine learning methods to further perception tasks. Traditional machine learning methods usually require the extraction of standard features, such as Haar operators, HOG operators, and gray-level co-occurrence matrices, to extract features from images, and then apply SVM, Adaboost, and other methods to classify these features.

[0062] Then, an initial neural network model is constructed based on all the data obtained above, and a vehicle driving information determination model is obtained by training. This embodiment takes a deep deterministic policy gradient algorithm as an example, and the specific steps are as follows:

[0063] Based on the fused perception data, the intelligent braking system uses prediction algorithms to make short-term predictions about the road conditions ahead. These predictions include but are not limited to the speed changes of vehicles ahead, the movement trajectories of pedestrians, and the development trend of potential risks.

[0064] In order to achieve a better kinetic energy recovery strategy, we can consider using reinforcement learning algorithms to achieve this goal. Because driving behavior in the real world is a continuous action, traditional reinforcement learning has steadily improved and surpassed human performance in a number of games.

[0065] However, these achievements are not easy to replicate in the field of autonomous driving, because the spatial state of the real world is more complex, the action space is continuous, and precise control is required. Autonomous vehicles should also give priority to functional safety in complex environments. In summary, the deep deterministic policy gradient algorithm can be considered here, which uses deterministic rather than random action functions. In particular, DDPG combines the advantages of deterministic policy gradient algorithms, actor-critic, and deep Q networks.

[0066] The DDPG network structure consists of four networks: the current Actor network, the target Actor network, the current Q network, and the target Q network. The inputs of the current Actor and target Actor networks are states, and the outputs are actions. Figure 3 As shown in the figure, the current Actor network inputs the current state S and outputs the action a under the current state S, while the target Actor network inputs the next state S′ and outputs the action a′ under the next state S′. The inputs of the Critic network are all states and actions, and it outputs the Q value of taking this action under this state. Among them, the current Critic network inputs the current state S and the current action a, and outputs the Q value under the current state S and the current action a; while the target Critic network inputs the next state S′ and the next action a′, and outputs the Q value of the next state S′ and the next action a′.

[0067] Therefore, the specific construction process of the DDPG network is as follows:

[0068] Step 1: Data collection. Since we need a lot of real data to train our algorithm model, we will mainly use HighD and Ngsim datasets here. Both of them are open source and based on real road conditions (highways, expressways, urban roads, etc.), they use advanced cameras and other sensors to obtain various dynamic parameters of driving vehicles, so that researchers can build and use algorithm models more quickly. Randomly select 5,000 complete data sets, and divide the data sets into training sets, validation sets, and test sets in a ratio of 7:2:1. After that, we will process the original data and process the missing and outlier data (random forest completion, deletion, etc.). Then perform feature extraction to select useful features, such as the position parameters of the target object, vehicle speed, vehicle acceleration, data frame number, and total data frame number as the input state of the DDPG model. Then define the state and action parameters. We can set the state vector to the current vehicle's speed, position, and distance to the front vehicle. The action vector is the vehicle's acceleration, the intensity of kinetic energy recovery, etc. Finally, the previous data is normalized to obtain the processed data, which will be in the range of [0, 1] or [-1, 1]. It should be noted that the processed data can accelerate convergence, avoid numerical instability and improve model performance.

[0069] Step 2: Build the environment. Use OpenAi Gym to build a simulated environment model, and construct the Actor network, Critic network, experience replay pool, and target network. The details are as follows:

[0070] Step a, in the Actor network, the input layer is state S, and the two hidden layers use two Relu functions and the last Tanh activation function to connect the output layer action a.

[0071] Step b, set the input layer of the Critic network to the state S and the action a just obtained, and the output to the target value Q.

[0072] Step c, define the experience replay pool. The experience replay pool is used to store the historical experience of the agent interacting with the environment. Its main purpose is to break the correlation between data by randomly extracting samples from the stored experience, thereby improving the stability and efficiency of training. Its experience replay pool usually contains the following information: current state S, execution action a, reward r, next state S′, whether it is the terminal state done (Boolean value, whether the end point is reached). In addition, the storage capacity N and the random extraction batch (mini-batch) can also be defined.

[0073] Step d, define the DDPG agent. The DDPG Agent is the subject used for training and execution in reinforcement learning. It includes Actor and Critic networks, as well as components such as the experience replay pool and the target network. The DDPG workflow includes: obtaining the current state from the environment; selecting actions using the Actor network and the noise generator; executing actions and observing new states and rewards; storing experiences (states, actions, rewards, next states, completion flags) in the experience replay pool; randomly extracting batch samples from the experience replay pool, and using the Critic network to calculate the target Q value; updating the Critic network, and training the Critic network by minimizing the mean square error of the Q value; using the gradient of the Critic network to update the Actor network to optimize the strategy; soft updating the weights of the target network (i.e. gradually transferring the weights of the main network to the target network).

[0074] Assuming the target value Q′ is the result we want, we can get the following according to the Bellman equation:

[0075] Q′=r+γ.Q(s′,a′) Formula (1)

[0076] Where r is the current reward, γ\gammaγ is a discount factor (usually in the range [0,1][0,1][0,1]), and Q′(s′,a′) is the target Critic network’s prediction of the Q value for the next state s′ and the next action a′.

[0077] Then, all parameters θ are updated using the gradient descent method to update the critic network as follows:

[0078]

[0079] Where N is the batch size, (si,ai,ri,si′,di) is the i-th sample drawn from the experience replay pool, and yi is the corresponding target Q value.

[0080] Finally, use the optimization function (Adam) to minimize the loss function and update the parameters of the critic network to complete the model construction.

[0081] Step 3: Model training. Import data based on the built environment model to start model training, and finally get the model we want by continuously optimizing and adjusting parameters. After the model training is completed, the data obtained by fusion perception and the location information obtained by GPS can be used as input to the trained DDPG model to determine whether the current vehicle status requires the activation of the intelligent braking system and the intensity of kinetic energy recovery.

[0082] In another embodiment of the present application, the target driving information of the vehicle within a future preset time period is determined based on the current driving information and the driving environment information, including: taking the current driving information and the driving environment information as input information; inputting the input information into a trained vehicle driving information determination model to obtain an output result of the vehicle driving information determination model, and determining the output result as the target driving information; the vehicle driving information determination model includes: a driving state prediction sub-model, which is used to predict the expected road traffic behavior of the vehicle within a future preset time period based on the current driving information and the driving environment information; and a driving information generation sub-model, which is used to determine the target driving information of the vehicle within a future preset time period based on the current driving information and the driving environment information, and the output of the driving state prediction sub-model.

[0083] In one embodiment of the present application, the road traffic behavior of a vehicle within a preset time is predicted based on current driving information and driving environment information, including: based on the current driving speed and the current position of the vehicle, predicting the target position of the vehicle after the preset time, the current position being obtained based on the current driving information; identifying the road traffic status of the interval section between the current position and the target position, the road traffic status including at least a smooth state and a blocked state; if the road traffic status is a smooth state, the driving behavior of the vehicle through the interval section does not include deceleration; if the road traffic status is a blocked state, the driving behavior of the vehicle through the interval section includes deceleration.

[0084] In a specific embodiment of the present application, the vehicle speed sensor is used to obtain the current vehicle speed in real time, for example, the current vehicle speed is 60 km / h, and the GPS module or the vehicle navigation system is used to determine the current location of the vehicle, for example, the vehicle is currently located at point A on a main road in a city. Take the preset time of 30 seconds as an example. The specific implementation is as follows:

[0085] First, based on the current driving speed (60km / h) and the preset time period (30 seconds), the expected position of the vehicle in 30 seconds is calculated. Assuming there is no deceleration or other interference factors, the vehicle will travel about 500 meters. Therefore, the target position point B is about 500 meters ahead of point A.

[0086] Then, the front camera captures images of the road ahead to identify lane lines, traffic signs, and other obstacles; the radar / lidar detects the distance and relative speed of obstacles ahead; genjuV2X communication receives data from other vehicles or infrastructure to understand the traffic conditions ahead; and, based on the navigation system, provides real-time traffic updates, including congestion information, construction areas, etc.

[0087] Then, the road traffic status is further classified. Specifically, if there are no obstacles or traffic jams between point A and point B, and the road conditions ahead are good, the road section is considered to be unobstructed. If one of the following conditions exists between point A and point B, the road section is considered to be blocked: there are stationary or slow-moving vehicles ahead, in a traffic accident or construction area, the traffic light is red or is about to turn red, the road is slippery or other bad weather conditions require slowing down.

[0088] Next, the driving behavior is predicted based on the road traffic status. If the road traffic status between point A and point B is unobstructed, the system will predict that the vehicle does not need to slow down in the next 30 seconds, and the vehicle will continue to drive at the current speed (60km / h) until it reaches point B; if the road traffic status between point A and point B is blocked, the system will predict that the vehicle will need to slow down in the next 30 seconds. For example, if there is a red light 500 meters ahead, the system will predict that the vehicle needs to slow down and stop safely at the red light.

[0089] Finally, if the system predicts that deceleration is needed, it will generate deceleration control instructions based on the current driving information and driving environment information. Its deceleration control instructions may include adjusting the strength of the kinetic energy recovery system and / or the strength of the mechanical brakes to ensure that the vehicle can smoothly decelerate to a safe speed and come to a complete stop if necessary. In addition, the system will continuously monitor the actual deceleration of the vehicle and dynamically adjust the deceleration strategy based on the actual situation to ensure that the vehicle smoothly decelerates to the target speed at the predetermined time node.

[0090] It should be noted that based on the method proposed in the above embodiment, the vehicle system can accurately predict the vehicle's road traffic behavior within a preset time period in the future based on current driving information and driving environment information, and take corresponding deceleration control measures according to the prediction results, thereby improving driving safety and comfort.

[0091] In one embodiment of the present application, the target driving information includes a target speed and a target deceleration, and determining the target driving information of the vehicle within a preset time period in the future includes: identifying obstacles between the current position and the target position based on environmental information, and recording the position and speed of each obstacle; wherein the obstacles include dynamic obstacles and stationary obstacles; calculating the interval distance between the vehicle and each obstacle based on the current position of the vehicle and the position of each obstacle, and calculating the expected collision time between the vehicle and each obstacle based on the interval distance, the speed of the obstacle, and the current driving speed; if the expected collision time is less than a preset safety time threshold, calculating the target speed based on the current speed and the interval distance to ensure that the vehicle stops before reaching the obstacle; and calculating the target deceleration to decelerate the vehicle to the target speed based on the current speed and the target speed.

[0092] In a specific embodiment of the present application, sensors such as a front camera, radar, and laser radar (LiDAR) are used to detect and identify obstacles between the current position and the target position; dynamic obstacles (such as other vehicles, pedestrians, etc.) and static obstacles (such as parked vehicles, roadblocks, etc.) are distinguished through image processing and object recognition algorithms; and the position coordinates and speed of each obstacle are recorded. For example, a vehicle traveling at a speed of 30km / h is detected 200 meters ahead, and a static roadblock is detected 150 meters ahead. Then, based on the current position of the vehicle and the position coordinates of each obstacle, the interval distance between the vehicle and each obstacle is calculated. For example, the vehicle is currently located at point A, and the interval distance at point B of the vehicle 200 meters ahead is 200 meters; the interval distance at point C of the roadblock 150 meters ahead is 150 meters, the current vehicle speed is 60km / h, and the preset safety time threshold is 3 seconds. The specific steps are as follows:

[0093] First, the estimated collision time between the vehicle and each obstacle is calculated based on the interval distance, the speed of the obstacle, and the current driving speed. The vehicle speed 200 meters ahead is 30km / h, so the relative speed between the current vehicle and the vehicle ahead is calculated to be 8.34m / s, and the estimated collision time is calculated to be 24 seconds. In addition, for the stationary roadblock 150 meters ahead, the estimated collision time is about 9 seconds.

[0094] Then, if the estimated collision time is less than the preset safety time threshold, the target speed is calculated based on the current speed and the interval distance to ensure that the vehicle stops before reaching the obstacle. In this embodiment, for the stationary roadblock 150 meters ahead, the estimated collision time is 9 seconds, which is greater than the safety time threshold of 3 seconds, so there is no need to stop immediately. However, for the vehicle 200 meters ahead, the estimated collision time is 24 seconds. Although it is greater than the safety time threshold of 3 seconds, if the vehicle continues to travel at the current speed, it may be necessary to slow down in advance to maintain a safe distance. Assuming that it is necessary to slow down to a safe speed within 10 seconds, the target speed = (interval distance-safety distance) / 10 seconds. Assuming the safety distance is 50 meters, the target speed = (200m-50m) / 10s = 15m / s (about 54km / h).

[0095] Then, based on the current speed and the target speed, the target deceleration is calculated to decelerate the vehicle to the target speed. The current speed is 16.67 m / s, the target speed is 15 m / s, and the deceleration needs to be completed within 10 seconds, so the target deceleration is 0.167 m / s. 2 .

[0096] Finally, the corresponding deceleration control instructions are generated based on the calculated target deceleration. The deceleration control instructions include adjusting the strength of the kinetic energy recovery system and / or the strength of the mechanical brake to ensure that the vehicle can smoothly decelerate to the target speed and stop completely when necessary. In addition, the system will continuously monitor the actual deceleration of the vehicle and dynamically adjust the deceleration strategy according to the actual situation to ensure that the vehicle can smoothly decelerate to the target speed at the predetermined time node.

[0097] It should be noted that based on the method proposed in this embodiment, the vehicle system can accurately predict the vehicle's target driving information within a preset time period in the future based on current driving information and driving environment information, and take corresponding deceleration control measures, thereby improving driving safety and comfort.

[0098] Step S230, generating a vehicle deceleration control instruction based on the target driving information, and controlling the strength of the vehicle's kinetic energy recovery system and / or the strength of the mechanical brake according to the vehicle deceleration control instruction, so that the vehicle decelerates to the target driving speed in the target driving information after reaching a preset time.

[0099] In one embodiment of the present application, a vehicle deceleration control instruction is generated based on target driving information, including: obtaining a deceleration threshold of a kinetic energy recovery system of the vehicle; if the target deceleration is less than or equal to the deceleration threshold of the kinetic energy recovery system, calculating the kinetic energy recovery intensity based on the target deceleration; if the target deceleration is greater than the deceleration threshold of the kinetic energy recovery system, calculating the deceleration difference between the deceleration threshold of the kinetic energy recovery system and the target deceleration, and determining the mechanical braking force based on the deceleration difference; converting the calculated mechanical braking force and kinetic energy recovery intensity into a control instruction, and performing deceleration control on the vehicle based on the control instruction to avoid a collision between the vehicle and an obstacle.

[0100] In one embodiment of the present application, the deceleration threshold of the kinetic energy recovery system is read from the electronic control unit (ECU) or related sensors of the vehicle. Taking the deceleration threshold of the kinetic energy recovery system as 0.3g as an example, the kinetic energy recovery intensity and mechanical braking force are calculated based on the target deceleration as follows:

[0101] When the target deceleration is less than or equal to the deceleration threshold of the kinetic energy recovery system. Assuming the target deceleration (for example, 0.25g, about 2.45m / s 2 ) is less than or equal to the deceleration threshold of the kinetic energy recovery system (0.3g), then only the kinetic energy recovery system is used for deceleration. Kinetic energy recovery intensity = target deceleration / deceleration threshold of the kinetic energy recovery system * 100%. For example, kinetic energy recovery intensity = 0.25g / 0.3g*100%≈83.33%.

[0102] When the target deceleration is greater than the deceleration threshold of the kinetic energy recovery system. Assuming the target deceleration (for example, 0.4g, about 3.92m / s 2 ) is greater than the deceleration threshold of the kinetic energy recovery system (0.3g), then the kinetic energy recovery system and mechanical brakes need to be combined to achieve the target deceleration. The calculated deceleration difference = target deceleration - kinetic energy recovery system deceleration threshold 0.1g. Therefore, mechanical brake force = deceleration difference / mechanical brake deceleration threshold * 100%. Assuming the mechanical brake deceleration threshold is 0.5g (about 4.9m / s 2 ), then the mechanical braking force = 0.1g / 0.5g*100% = 20%.

[0103] Finally, the calculated mechanical braking force and kinetic energy recovery intensity are converted into specific control instructions. For example, if the kinetic energy recovery intensity is 83.33%, the corresponding kinetic energy recovery system control instruction is generated; if the mechanical braking force is 20%, the corresponding mechanical brake control instruction is generated. Then, the ECU sends the generated control instruction to the kinetic energy recovery system and the mechanical brake system. The kinetic energy recovery system adjusts its working intensity according to the control instruction to achieve the required deceleration, and the mechanical brake system adjusts the braking force according to the control instruction to supplement the target deceleration part that the kinetic energy recovery system cannot achieve. In addition, the system will continuously monitor the actual deceleration effect and dynamically adjust the control instruction according to the actual situation to ensure that the vehicle decelerates smoothly to the target speed and avoids collision with obstacles.

[0104] It should be noted that through the above steps, the vehicle system can generate accurate vehicle deceleration control instructions based on the target driving information, and through the rational use of the kinetic energy recovery system and the mechanical braking system, ensure that the vehicle decelerates smoothly within a safe time, thereby improving driving safety and comfort.

[0105] In one embodiment of the present application, after controlling the strength of the vehicle's kinetic energy recovery system and / or the strength of the mechanical brake according to the vehicle deceleration control instruction, it also includes at least one of the following: real-time monitoring of the vehicle's actual deceleration, including at least changes in vehicle speed, acceleration and the response of the braking system, and dynamically adjusting the braking force and / or kinetic energy recovery strength according to the difference between the monitored actual deceleration and the expected deceleration target to ensure that the actual deceleration process meets the expected deceleration curve; continuously evaluating changes in driving environment information during the deceleration process, such as changes in the position of obstacles ahead and sudden changes in road conditions, and dynamically adjusting the deceleration strategy according to the changes to ensure driving safety; when the vehicle reaches the preset target speed or stops completely, recording the data of the entire deceleration process, including the deceleration control instructions issued, the actual deceleration effect and any adjustment measures, for optimizing the vehicle driving information determination model.

[0106] In one embodiment of the present application, when the decision algorithm determines that action is needed, the intelligent braking system accurately adjusts the working state of the kinetic energy recovery system through the execution unit. This may include increasing or decreasing the strength of kinetic energy recovery to achieve fine control of the vehicle speed. When necessary, the system will also trigger the mechanical brake to ensure driving safety. In order to enhance the driver's perception of the system status, the intelligent braking system will provide real-time feedback through the user interface. This includes information such as the current strength of kinetic energy recovery, braking status, predicted driving conditions, and potential risks. Through these feedbacks, the driver can more intuitively understand the operating status of the system and intervene manually when necessary. In addition, when the vehicle stops and the ignition switch is turned off, the intelligent braking system will automatically execute the shutdown procedure. During this process, the system will save the current status and data so that it can be quickly restored at the next start. Subsequently, the system will enter a dormant state to reduce energy consumption and extend the service life of the equipment.

[0107] Figure 4 It is a schematic diagram of the overall scheme of a vehicle deceleration control method shown in an exemplary embodiment of the present application.

[0108] like Figure 4 As shown, in the overall process of the vehicle deceleration control method, the current driving information and driving environment information of the vehicle are first obtained, so as to predict the road traffic state of a certain stage in the future based on its current driving information and driving environment information, and the prediction result is obtained as a smooth state or a blocked state. If the traffic state of the road ahead is a blocked state, the deceleration condition is further calculated to obtain the corresponding target deceleration. If the target deceleration is greater than the deceleration threshold of the vehicle kinetic energy recovery system, the intensity of the power recovery system and the strength of the mechanical brake are notified to achieve vehicle deceleration. If the target deceleration is not greater than the deceleration threshold of the vehicle kinetic energy recovery system, the intensity of the kinetic energy recovery system is directly controlled to achieve vehicle deceleration. Among them, the deceleration threshold of the kinetic energy recovery system is a preset value, and its value is different based on different kinetic energy recovery systems and different vehicle driving modes. Therefore, this application does not make any restrictions on the value method and specific value of the deceleration threshold of its recovery system.

[0109] In one embodiment of the present application, based on the current driving information and driving environment information, a machine learning model (such as a deep deterministic policy gradient DDPG algorithm) is used to predict the road traffic state at a certain stage in the future (for example, 30 seconds later). The prediction result can be a smooth state or an obstructed state. Take the prediction result showing that there is a stationary vehicle 100 meters ahead, resulting in an obstructed road traffic state as an example. According to the current position and the position of the obstacle, the interval distance between the vehicle and the obstacle is calculated, and the estimated collision time is calculated based on the interval distance, the speed of the obstacle, and the current driving speed, so as to obtain the corresponding deceleration requirement, and then obtain the target deceleration. If the target deceleration is greater than the maximum deceleration of the kinetic energy recovery system, it is necessary to combine the kinetic energy recovery system and the mechanical brake to achieve the target deceleration, realize the calculation of the maximum deceleration that the kinetic energy recovery system can provide, and calculate the strength of the mechanical brake based on the difference in deceleration.

[0110] It should be noted that in the vehicle deceleration control method proposed in this application, an advanced vehicle deceleration control method is provided, which significantly improves energy efficiency by dynamically adjusting the kinetic energy recovery force and the use of mechanical brakes. In different driving scenarios, the system can maximize energy recovery, especially when downhill and turning sharply, converting more kinetic energy into electrical energy storage, thereby extending the battery's range. In addition, in a safe, straight and obstacle-free situation, the system can adjust the kinetic energy recovery force to low, providing a driving experience close to that of a traditional fuel vehicle, increasing driving comfort. The method also significantly enhances driving safety. In the case of congestion or waiting for traffic lights, the system can dynamically adjust the resistance of the brake and accelerator pedals according to the distance and waiting position of the vehicle in front, avoiding the risk of collision caused by rapid acceleration of electric vehicles. At the same time, by intelligently adjusting kinetic energy recovery, the dependence on the traditional brake system is reduced, thereby reducing the wear of the brake system, extending its service life, and thus reducing long-term maintenance and replacement costs. This not only saves economic costs, but also further improves energy efficiency and reduces energy consumption. In addition, the intelligent system has a high degree of real-time and responsiveness, and can quickly identify driving situations and respond to ensure driving safety. The system that automatically adjusts braking and acceleration reduces the driver's operating burden in complex driving environments, reduces driving stress, and makes the driving process easier and more secure. In summary, the method of the present application not only improves energy efficiency and driving experience, but also enhances driving safety and system durability, and has significant comprehensive benefits.

[0111] Figure 5 is a block diagram of a vehicle deceleration control device shown in an exemplary embodiment of the present application. The device can be applied to Figure 1The device may also be applicable to other exemplary implementation environments and specifically configured in other devices, and this embodiment does not limit the implementation environment to which the device is applicable.

[0112] like Figure 5 As shown, the exemplary vehicle deceleration control device includes: an information acquisition module 510 , a target driving information generation module 520 , and a deceleration control module 530 .

[0113] Among them, the information acquisition module 510 is used to obtain the current driving information and driving environment information of the vehicle, the current driving information is parameter information used to describe the vehicle's motion state, and the driving environment information is information about the driving scene in which the vehicle is located; the target driving information generation module 520 is used to predict the vehicle's expected road traffic behavior within a future preset time period based on the current driving information and driving environment information; if the expected road traffic behavior includes vehicle deceleration, the target driving information of the vehicle within the future preset time period is determined based on the current driving information and driving environment information, and the target driving information is parameter information used to describe the vehicle's future motion state; the deceleration control module 530 is used to generate a vehicle deceleration control instruction based on the target driving information, and control the strength of the vehicle's kinetic energy recovery system and / or the strength of the mechanical brake according to the vehicle deceleration control instruction, so that the vehicle decelerates to the target driving speed in the target driving information after reaching the preset time.

[0114] It should be noted that the vehicle deceleration control device provided in the above embodiment and the vehicle deceleration control method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment, and will not be repeated here. In practical applications, the vehicle deceleration control device provided in the above embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0115] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the electronic device implements the vehicle deceleration control method provided in the above-mentioned embodiments.

[0116] Figure 6 The structure diagram of the computer system suitable for implementing the electronic device of the embodiment of the present application is shown. It should be noted that: Figure 6 The computer system 600 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0117] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 602 or the program loaded from the storage part 608 to the random access memory (RAM) 603, such as executing the method described in the above embodiment. In the RAM 603, various programs and data required for system operation are also stored. The CPU 601, ROM 602 and RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0118] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed so that a computer program read therefrom is installed into the storage section 608 as needed.

[0119] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 609, and / or installed from a removable medium 611. When the computer program is executed by a central processing unit (CPU) 601, various functions defined in the system of the present application are executed.

[0120] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0121] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0122] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0123] Another aspect of the present application also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor of a computer, the computer executes the vehicle deceleration control method as described above. The computer-readable storage medium may be included in the electronic device described in the above embodiment, or may exist independently without being assembled into the electronic device.

[0124] Another aspect of the present application also provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the vehicle deceleration control method provided in each of the above embodiments.

[0125] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A vehicle deceleration control method, characterized in that: The method comprises: Acquiring current driving information and driving environment information of the vehicle, wherein the current driving information is parameter information used to describe the current motion state of the vehicle, and the driving environment information is information about the driving scene in which the vehicle is located; Based on the current driving information and the driving environment information, predict the expected road traffic behavior of the vehicle within a future preset time period; if the expected road traffic behavior includes vehicle deceleration, determine the target driving information of the vehicle within the future preset time period according to the current driving information and the driving environment information, the target driving information being parameter information used to describe the future motion state of the vehicle; A vehicle deceleration control instruction is generated based on the target driving information, and the strength of the vehicle's kinetic energy recovery system and / or the strength of the mechanical brakes are controlled according to the vehicle deceleration control instruction, so that the vehicle decelerates to the target driving speed in the target driving information after a preset time.

2. The vehicle deceleration control method according to claim 1, characterized in that: Determining target driving information of the vehicle within a future preset time period according to the current driving information and the driving environment information includes: using the current driving information and the driving environment information as input information; Inputting the input information into a trained vehicle driving information determination model to obtain an output result of the vehicle driving information determination model, and determining the output result as target driving information; The vehicle driving information determination model includes: A driving state prediction sub-model, used to predict the expected road traffic behavior of the vehicle within a future preset time period based on the current driving information and the driving environment information; The driving information generation sub-model is used to determine the target driving information of the vehicle within a preset time period in the future based on the current driving information and the driving environment information, as well as the output of the driving state prediction sub-model.

3. The vehicle deceleration control method according to claim 2, characterized in that: Predicting the road traffic behavior of the vehicle within a preset time based on the current driving information and the driving environment information includes: Predicting a target position of the vehicle after a preset time based on a current driving speed and a current position of the vehicle, the current position being obtained based on the current driving information; Identify the road traffic status of the interval section between the current position and the target position, wherein the road traffic status includes at least a smooth state and a blocked state; If the road traffic state is a smooth state, the driving behavior of the vehicle passing through the interval section does not include deceleration; if the road traffic state is a blocked state, the driving behavior of the vehicle passing through the interval section includes deceleration.

4. The vehicle deceleration control method according to claim 2, characterized in that: The target driving information includes a target speed and a target deceleration, Determine the target driving information of the vehicle within a preset time period in the future, including: Based on the environmental information, obstacles between the current position and the target position are identified, and the position and speed of each obstacle are recorded; wherein the obstacles include dynamic obstacles and stationary obstacles; Calculating the interval distance between the vehicle and each obstacle based on the current position of the vehicle and the position of each obstacle, and calculating the expected collision time between the vehicle and each obstacle based on the interval distance, the speed of the obstacle, and the current driving speed; If the estimated collision time is less than a preset safety time threshold, a target speed is calculated based on the current speed and the interval distance to ensure that the vehicle stops before reaching the obstacle; Based on the current speed and the target speed, a target deceleration for decelerating the vehicle to the target speed is calculated.

5. The vehicle deceleration control method according to claim 4, characterized in that: Generating a vehicle deceleration control instruction based on the target driving information includes: Obtaining a deceleration threshold of a vehicle kinetic energy recovery system; If the target deceleration is less than or equal to the deceleration threshold of the kinetic energy recovery system, calculating the kinetic energy recovery intensity based on the target deceleration; If the target deceleration is greater than the deceleration threshold of the kinetic energy recovery system, calculating the deceleration difference between the deceleration threshold of the kinetic energy recovery system and the target deceleration, and determining the mechanical braking force based on the deceleration difference; The calculated mechanical braking force and kinetic energy recovery intensity are converted into control instructions, and the vehicle is decelerated based on the control instructions to avoid collision between the vehicle and obstacles.

6. The vehicle deceleration control method according to claim 1, characterized in that: The method further comprises: Constructing an initial neural network model, wherein the initial neural network model is implemented based on a preset machine learning algorithm; Training the initial neural network model based on historical driving information and corresponding driving environment information to generate a vehicle driving information determination model; Inputting the current driving information into the vehicle driving information determination model to predict the road conditions of the vehicle within a future preset time period; Generate corresponding target driving information based on the predicted road conditions, the target driving information at least including a target speed and a target deceleration; The force of the mechanical brake and the strength of the kinetic energy recovery system are gradually adjusted to smoothly transition the vehicle from the current driving state to the driving state indicated by the target driving information.

7. The vehicle deceleration control method according to any one of claims 2 to 6, characterized in that: After controlling the strength of the vehicle's kinetic energy recovery system and / or the strength of the mechanical brake according to the vehicle deceleration control instruction, the method further includes at least one of the following: Monitor the actual deceleration of the vehicle in real time, including at least the change in vehicle speed, acceleration and the response of the braking system, and dynamically adjust the braking force and / or kinetic energy recovery intensity according to the difference between the actual deceleration monitored and the expected deceleration target, so as to ensure that the actual deceleration process meets the expected deceleration curve; During deceleration, the system continuously evaluates changes in driving environment information, such as changes in the position of obstacles ahead and sudden changes in road conditions, and dynamically adjusts deceleration strategies based on the changes to ensure driving safety. When the vehicle reaches a preset target speed or comes to a complete stop, data from the entire deceleration process is recorded, including the deceleration control instructions issued, the actual deceleration effect, and any adjustment measures, for optimizing the vehicle driving information determination model.

8. A vehicle deceleration control device, characterized in that: The device comprises: An information acquisition module, used to acquire current driving information and driving environment information of the vehicle, wherein the current driving information is parameter information used to describe the vehicle's motion state, and the driving environment information is information about the driving scene in which the vehicle is located; a target driving information generating module, configured to predict the expected road traffic behavior of the vehicle within a preset time period in the future based on the current driving information and the driving environment information; if the expected road traffic behavior includes vehicle deceleration, determine the target driving information of the vehicle within the preset time period in the future based on the current driving information and the driving environment information, wherein the target driving information is parameter information used to describe the future motion state of the vehicle; A deceleration control module is used to generate a vehicle deceleration control instruction based on the target driving information, and control the strength of the vehicle's kinetic energy recovery system and / or the strength of the mechanical brake according to the vehicle deceleration control instruction, so that the vehicle decelerates to the target driving speed in the target driving information after a preset time.

9. An electronic device, characterized in that: It comprises a processor, a memory and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the vehicle deceleration control method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is used to enable a computer to execute the vehicle deceleration control method as described in any one of claims 1-7.

Images