A method and system for controlling vehicle steering stability

Abstract

The present invention discloses a method and system for controlling vehicle steering stability, the method comprising: the cloud collects and processes data, optimizes the trajectory based on the MPC model, and sends the optimal trajectory of stability constraints to the vehicle end; the vehicle end perceives the environment in real time and controls the vehicle steering stability based on the trajectory sent by the cloud end. The present invention uses the MPC model to perform trajectory optimization and real-time control through the collaborative work of the cloud end and the vehicle end, effectively solving the problems of real-time performance, poor environmental adaptability and lack of global optimization in the prior art, and significantly improving the performance of vehicle steering stability control.

Description

Technical Field

[0001] The invention belongs to the technical field of vehicle control, and in particular relates to a method and system for controlling automobile steering stability. Background Art

[0002] With the rapid development of the automobile industry, the safety and stability of vehicles have become increasingly important. The traditional method of automobile steering stability control mainly relies on on-board sensors to collect road conditions and vehicle dynamic information in real time, and process and control it through an on-board computing unit. In this method, the vehicle is equipped with various sensors, such as accelerometers, gyroscopes, and wheel speed sensors, to monitor the vehicle's motion state and surrounding environment. The on-board computing unit receives these sensor data and uses a preset control algorithm to adjust the vehicle's steering angle and speed in real time to ensure the stability of the vehicle under various driving conditions. This traditional method has improved the vehicle's handling and safety to a certain extent and is widely used in modern vehicles. However, with the complexity of the driving environment and the improvement of users' requirements for driving safety, traditional on-board computing and control methods face challenges in dealing with complex road conditions and changeable weather conditions. Therefore, how to further improve the effect of steering stability control has become an important direction for current technological development.

[0003] In order to solve the above problems, the present invention proposes a vehicle steering stability control method based on model predictive control (MPC). Through the collaborative work of the cloud and the vehicle, the powerful computing power of the cloud is used to optimize the trajectory, and combined with the real-time perception and control of the vehicle, the global optimal control of steering stability is achieved. This method not only improves the real-time and accuracy of the control, but also enhances the system's adaptability to complex environments, significantly improving the safety and stability of the vehicle. Summary of the invention

[0004] In view of the defects existing in the above-mentioned prior art, the present invention provides a method for controlling the steering stability of an automobile, the method comprising:

[0005] The cloud collects and processes data, optimizes the trajectory based on the MPC model, and sends the optimal trajectory with stability constraints to the vehicle.

[0006] The vehicle side perceives the environment in real time and performs vehicle steering stability control based on the trajectory sent from the cloud.

[0007] The cloud collects data, and the data types include destination, estimated arrival time, weather, road type and road adhesion coefficient.

[0008] Among them, there is a unified database structure in the cloud, in which each road segment information corresponds to the location in the high-precision map.

[0009] Among them, the navigation trajectory is determined in the cloud, including determining all road segments through a path planning algorithm according to the destination and arrival time, and querying the corresponding adhesion coefficient correction value from the database according to the road segment ID and weather in the driving path.

[0010] Wherein, calculating the actual adhesion coefficient in the cloud may also include:

[0011] For each road segment, the weather correction value is applied to calculate the actual adhesion coefficient. The formula is as follows: .

[0012] in:

[0013] : Actual adhesion coefficient

[0014] : Adhesion coefficient under standard conditions

[0015] : Correction value of adhesion coefficient for weather conditions

[0016] Among them, nonlinear MPC trajectory optimization is performed in the cloud.

[0017] Among them, in the cloud-based MPC trajectory planning model, based on the vehicle kinematics and dynamics constraints, stability-related constraints such as maximum lateral acceleration limit, maximum yaw angular velocity limit and minimum curvature radius limit are introduced into the model; in the objective function, evaluation indicators related to steering stability are added to minimize the lateral acceleration change rate and the yaw angular velocity change rate to obtain control input.

[0018] Among them, the cloud sends the generated optimal trajectory to the vehicle-side control module; the vehicle-side control module receives and parses the optimal trajectory data from the cloud, including timestamp, position, speed, acceleration, heading angle, yaw angular velocity, steering angle, curvature, road adhesion coefficient and environmental information.

[0019] The vehicle-side MPC optimization process includes:

[0020] Construct a linearized state space model, including using the current state and reference trajectory points to linearize the vehicle dynamics model and discretize it into a state transfer matrix A k and the control input matrix B k .

[0021] Based on the linearized model and reference trajectory, the cost function and constraints of the optimization problem are constructed, and the control input sequence is solved through the optimization algorithm.

[0022] The resulting control input is applied to the vehicle.

[0023] The present invention also discloses a vehicle steering stability control system, the system includes a cloud end and a vehicle end, the cloud end and the vehicle end both include a memory and a processor, the memory stores a computer program, and the processor is used to execute the computer program to implement the aforementioned method.

[0024] The present invention uses the MPC model to perform trajectory optimization and real-time control through the collaborative work of the cloud and the vehicle, effectively solving the problems of real-time performance, poor environmental adaptability and lack of global optimization in the prior art, and significantly improving the performance of vehicle steering stability control. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present disclosure will become readily understood. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0026] Figure 1 is a flow chart showing a method for controlling vehicle steering stability according to an embodiment of the present invention;

[0027] Figure 2 is a flow chart showing a cloud execution method for automobile steering stability control according to an embodiment of the present invention;

[0028] Figure 3 The present invention is a flow chart showing a method for executing vehicle-side steering stability control according to an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0031] It should be understood that although the terms first, second, third, etc. may be used to describe ... in the embodiments of the present invention, these ... should not be limited to these terms. These terms are only used to distinguish .... For example, without departing from the scope of the embodiments of the present invention, the first ... may also be referred to as the second ..., and similarly, the second ... may also be referred to as the first ....

[0032] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0033] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0034] It should also be noted that the term "includes", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, so that a commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprising a ..." do not exclude the existence of other identical elements in the commodity or device including the elements.

[0035] Traditional vehicle steering control methods have difficulty processing a large amount of environmental information in real time and making optimal decisions, resulting in poor stability control of vehicles under complex road conditions. Existing control systems cannot dynamically adjust control strategies when facing different weather and road conditions, and lack pertinence and flexibility. Existing systems can usually only process local information, making it difficult to optimize trajectories on a global scale, and unable to fully utilize cloud computing resources for global optimization control.

[0036] The present invention achieves trajectory optimization and real-time control by using the MPC model through collaborative work between the cloud and the vehicle.

[0037] like Figure 1 As shown, the present invention discloses a method for controlling automobile steering stability, the method comprising:

[0038] The cloud collects and processes data, optimizes the trajectory based on the MPC model, and sends down the optimal trajectory with stability constraints, thus realizing cloud-based MPC trajectory planning and stability constraints.

[0039] The vehicle perceives the environment in real time and controls the vehicle based on the trajectory sent from the cloud, and performs stability control based on the vehicle-side control module.

[0040] The present invention optimizes the trajectory based on the MPC model in the cloud, and sends the optimal trajectory including stability constraints to the vehicle side, which perceives the environment in real time and controls based on the cloud trajectory. This can improve the real-time and accuracy of steering control and ensure that the vehicle can maintain stability under various road conditions.

[0041] The cloud first collects data related to steering stability such as destination, estimated arrival time, weather, road type, road adhesion coefficient, etc., and cleans and pre-processes this data.

[0042] In one embodiment, in order to achieve pre-planning and optimize the steering stability of the vehicle, it is first necessary to collect a variety of relevant data. The data includes the destination, estimated arrival time, weather, and data closely related to steering stability such as road type and road adhesion coefficient.

[0043] The collection process relies on multiple sources: the destination and estimated arrival time are obtained through user input in the navigation system, weather data can be obtained from the meteorological service platform, and historical route vehicle flow data is extracted from the road management system.

[0044] Data on road surface type and road adhesion coefficient are usually pre-stored in the system and stored in combination with high-precision map data.

[0045] The road surface type data is inferred based on the detailed road information of high-precision maps. These high-precision maps usually contain rich road attribute information, such as road grade (highway, urban road, rural road, etc.), slope, curvature, road width and surface material. By analyzing these attributes, the road surface type of different road sections can be inferred. For example, highways usually have better road surface quality and higher adhesion coefficient, while rural roads may have more muddy or gravel sections with lower adhesion coefficient.

[0046] In one embodiment, the data of the road adhesion coefficient also comes from historical measurement data and reports from road management departments. These data reflect the adhesion performance of different road sections under various weather and seasonal conditions, and are stored in the database in combination with specific road sections of high-precision maps. For example, for a specific road section, the system stores a baseline value of the adhesion coefficient and a correction value of the adhesion coefficient under different weather conditions (such as sunny, rainy, and snowy days).

[0047] In one embodiment, in order to realize the combined storage of data, the system needs to use a unified database structure, in which each road segment corresponds to a specific location in the high-precision map. Specifically, the database adopts the following structure:

[0048] Road segment identifier: Each road segment has a unique identifier (such as a road segment ID) in the HD map.

[0049] Road attributes: store the attribute information of road segments, including road grade, slope, curvature, road width and surface material.

[0050] Road surface type: Road surface type information inferred based on road attributes.

[0051] Adhesion coefficient: stores the road adhesion coefficient of the road section under standard conditions.

[0052] Adhesion coefficient correction value: Adhesion coefficient correction values ​​stored in different weather conditions, including correction values ​​in different weather conditions such as rain, snow and frost.

[0053] The collected data needs to be cleaned and pre-processed to ensure its accuracy and consistency in format.

[0054] In one embodiment, in order to realize the combined storage of data, the following database table structure is used to store road information:

[0055] Road segment table:

[0056] road_segment_id: Each road segment has a unique identifier in the HD map.

[0057] road_grade: Road grade (such as highway, city road, etc.).

[0058] slope: slope.

[0059] curvature: curvature.

[0060] width: Road width.

[0061] surface_material: Surface material.

[0062] Pavement type table:

[0063] road_segment_id: A unique identifier for each road segment.

[0064] surface_type: surface type (inferred based on road attributes).

[0065] Adhesion coefficient table:

[0066] road_segment_id: A unique identifier for each road segment.

[0067] standard_friction_coefficient: Road adhesion coefficient under standard conditions.

[0068] Adhesion coefficient correction table:

[0069] road_segment_id: A unique identifier for each road segment.

[0070] weather_condition: Weather conditions (e.g. rain, snow, frost).

[0071] correction_value: Correction value of the adhesion coefficient under different weather conditions.

[0072] In one embodiment, nonlinear MPC trajectory optimization is performed in the cloud.

[0073] Model predictive control (MPC) uses the mathematical model of the system to predict the future behavior of the system at each control moment, and finds the optimal control input sequence through optimization algorithms to optimize the system's performance indicators.

[0074] In one embodiment, in the cloud-based MPC trajectory planning model, the vehicle kinematic and dynamic constraints are considered, and stability-related constraints such as maximum lateral acceleration limit, maximum yaw rate limit, and minimum curvature radius limit are introduced into the model. In the objective function, evaluation indicators related to steering stability are added to minimize the lateral acceleration change rate and minimize the yaw rate change rate to ensure that the trajectory optimizes the vehicle's driving smoothness and speed while meeting the stability requirements. Using an optimization algorithm, the MPC optimization problem is solved under a predetermined driving time limit to generate an optimal trajectory that takes into account speed, smoothness, and steering stability, and the generated optimal trajectory is sent to the vehicle-side control module.

[0075] In one embodiment, the navigation trajectory is determined in the cloud, including determining all road segments based on the destination and arrival time, which is implemented by a path planning algorithm (such as Dijkstra algorithm or A* algorithm). This step outputs a list of road segment IDs representing the vehicle's travel path. Get current or predicted weather information. Assume that there is a function or API that returns the current weather conditions, such as "rain", "snow" or "frost". Based on the road segment ID and weather conditions in the travel path, query the corresponding road segment attributes and adhesion coefficient correction values ​​from the database.

[0076] The present invention collects and processes data including destination, estimated arrival time, weather, road type and road adhesion coefficient through the cloud, and dynamically adjusts trajectory planning and control strategies. By applying weather correction values ​​to calculate the actual adhesion coefficient, the vehicle can adjust the steering control strategy according to different weather and road conditions, enhancing the environmental adaptability of the system.

[0077] In one embodiment, calculating the actual adhesion coefficient includes: for each road segment, applying the weather correction value to calculate the actual adhesion coefficient. The formula is as follows: .

[0078] in:

[0079] : Actual adhesion coefficient

[0080] : Adhesion coefficient under standard conditions

[0081] : Adhesion coefficient correction value for weather conditions

[0082] In one embodiment, after obtaining the actual adhesion coefficients of all road segments, the cloud uses nonlinear MPC model predictive control MPC to perform trajectory optimization.

[0083] Among them, all parameters defined in the optimization process of the nonlinear MPC deployed in the cloud include:

[0084] : The longitudinal position of the vehicle at time t.

[0085] : The lateral position of the vehicle at time t.

[0086] : The heading angle of the vehicle at time t.

[0087] : The speed of the vehicle at time t.

[0088] : Acceleration of the vehicle at time t (control input).

[0089] : Steering angle of the vehicle at time t (control input).

[0090] : The longitudinal position of the reference trajectory at time t.

[0091] : The lateral position of the reference trajectory at time t.

[0092] : Reference speed.

[0093] : The weight coefficients of each item in the objective function.

[0094] : Vehicle wheelbase.

[0095] : Sampling time step.

[0096] : acceleration due to gravity.

[0097] : Actual adhesion coefficient.

[0098] : Speed ​​limit.

[0099] : Acceleration limit.

[0100] : Steering angle limit.

[0101] : lateral acceleration of the vehicle at time t.

[0102] : Radius of curvature.

[0103] In one embodiment, the steps of optimizing the nonlinear MPC deployed in the cloud are as follows:

[0104] Define the vehicle's state variables and control inputs:

[0105] : The longitudinal position of the vehicle at time t.

[0106] : The lateral position of the vehicle at time t.

[0107] : The heading angle of the vehicle at time t.

[0108] : The speed of the vehicle at time t.

[0109] : Acceleration of the vehicle at time t (control input).

[0110] : Steering angle of the vehicle at time t (control input).

[0111] In one embodiment, the dynamic model of the vehicle used in the nonlinear MPC deployed in the cloud is represented by the following difference equation:

[0112] ;

[0113] ;

[0114] ;

[0115] ;

[0116] in is the sampling time step, and L is the vehicle wheelbase.

[0117] In one embodiment, the objective function used in the nonlinear MPC deployed in the cloud aims to minimize the following:

[0118] Trajectory Error:

[0119] The error of the vehicle's longitudinal position relative to the reference trajectory.

[0120] : The error between the vehicle's lateral position and the reference trajectory.

[0121] Speed ​​error:

[0122] : The error between the vehicle speed and the reference speed.

[0123] Control Input:

[0124] : The square of the acceleration, representing energy consumption or smooth driving.

[0125] : The square of the steering angle, representing the steering smoothness.

[0126] In one embodiment, the objective function used in the nonlinear MPC deployed in the cloud is of the following form:

[0127] ;

[0128] in:

[0129] : Weight coefficient, used to balance various costs.

[0130] In one embodiment, the actual constraints used in the nonlinear MPC deployed in the cloud include physical constraints, specifically:

[0131] Speed ​​Limit: .

[0132] Acceleration limit: .

[0133] Steering angle limit: .

[0134] Effect of adhesion coefficient:

[0135] Maximum lateral acceleration constraint, subject to road adhesion coefficient Impact: .in is the radius of curvature and g is the acceleration due to gravity.

[0136] In one embodiment, the control problem used in the nonlinear MPC deployed in the cloud is formulated as the following optimization problem:

[0137] ,

[0138]

[0139]

[0140]

[0141]

[0142]

[0143]

[0144]

[0145]

[0146] .

[0147] In one embodiment, in a nonlinear MPC deployed in the cloud, the control input is obtained by solving a nonlinear constrained optimization problem. and Common methods include linear quadratic regulator , Quadratic Programming wait.

[0148] In one embodiment, performing MPC trajectory optimization in the cloud includes the following steps:

[0149] Using the path planning algorithm, all road segments are determined according to the destination and arrival time, and a list of road segment IDs is generated.

[0150] Get current or forecast weather information.

[0151] According to the road segment ID and weather conditions in the driving path, the corresponding road segment attributes and adhesion coefficient correction values ​​are queried from the database.

[0152] For each road segment, the actual adhesion coefficient is calculated by applying the weather correction value: .

[0153] Define the vehicle's state variables and control input .

[0154] The state update is described using the vehicle dynamics equations.

[0155] Define the objective function J to minimize the trajectory error and control input.

[0156] Define constraints, including dynamic constraints, physical constraints, and maximum lateral acceleration constraints.

[0157] Solve optimization problems and use methods such as quadratic programming (QP) to solve optimization problems and obtain control inputs and .

[0158] Through the above process, the present invention ensures that the vehicle can reach the destination safely and efficiently under different weather conditions by performing nonlinear MPC optimization in the cloud. The MPC method combines real-time optimization and system constraints, enabling the vehicle to perform intelligent driving in complex dynamic environments.

[0159] The present invention uses a unified database structure and high-precision map information in the cloud, combined with a path planning algorithm and actual adhesion coefficient calculation, to perform global optimal trajectory planning. Through a nonlinear MPC trajectory optimization model, the global optimal steering stability control trajectory is generated by considering vehicle kinematic and dynamic constraints and stability-related constraints.

[0160] The present invention optimizes the control input by minimizing the lateral acceleration change rate and the yaw angular velocity change rate in the cloud-based MPC trajectory planning model, thereby further improving the steering stability.

[0161] In a certain embodiment, in order to achieve trajectory optimization based on the MPC model and send the optimal trajectory with stability constraints, the present invention sends the trajectory to the vehicle control unit from the cloud based on segmented sending. The sending strategy selects the corresponding segmentation strategy according to the network mode supported by the vehicle, user settings or real-time network conditions.

[0162] In one embodiment, the system needs to detect the network mode currently supported by the vehicle. This obtains relevant information through the vehicle's communication module. For example, by calling the communication interface to obtain the current network type, the return value is '4G', '5G', 'WiFi', etc.

[0163] In one embodiment, the system should allow users to set segmentation strategies through the vehicle control interface or mobile application. Users select different segmentation strategies, such as by time, by distance, by priority, etc. Design a user setting interface to obtain segmentation strategies from user settings.

[0164] In one embodiment, the system also needs to obtain the current network status through the network monitoring module, including bandwidth, latency, and packet loss rate information. This information is obtained by calling the network monitoring interface and returning a dictionary containing network status, such as 'bandwidth', 'latency', 'packet_loss_rate'.

[0165] In one embodiment, after obtaining the network support information, user settings and real-time network status of the vehicle, the system selects the corresponding segmentation strategy based on this information. The specific selection logic is: if the user has a clear segmentation strategy setting, the user-set strategy is used first; if there is no user setting, the default strategy is selected according to the network type currently supported by the vehicle and the real-time network status, and the default strategy is a preset strategy. For example, the strategy may include: segmented transmission of the optimal trajectory based on the size of the data packet. For a 4G network, if the bandwidth is greater than 5Mbps and the delay is less than 50ms, a medium-sized data packet (such as 20KB) is selected to be sent; if the bandwidth is low or the delay is high, a smaller data packet (between 1KB and 5KB) is selected to be sent. For a 5G network, if the bandwidth is greater than 20Mbps and the delay is less than 20ms, a larger data packet (such as 50KB) is selected to be sent; if the network condition is general, a medium-sized data packet (such as 20KB) is selected to be sent. For a WiFi network, a suitable data packet size and transmission frequency are selected according to similar logic.

[0166] In one embodiment, for small data packet segmentation, the system divides the trajectory data into smaller data packets, each of which is between 1KB and 5KB in size. For medium data packet segmentation, the trajectory data is divided into medium-sized data packets of 5KB to 20KB. For large data packet segmentation, the trajectory data is divided into large data packets of 20KB to 50KB.

[0167] In one embodiment, for time-based segmentation, the trajectory data is segmented by time periods, such as one data segment per minute.

[0168] In one embodiment, for distance-based segmentation, the trajectory data is segmented by distance segments, such as one data segment every 1000 meters or 2000 meters. For priority segmentation.

[0169] In one embodiment, key points in the trajectory (such as turning points, acceleration points, deceleration points, etc.) are used as segmentation basis, and slices are sent after a preset distance before and after the key points, and high-priority data in a segment is sent first.

[0170] In one embodiment, the MPC model is first used to perform trajectory optimization to generate the optimal trajectory. Then, a suitable segmentation strategy is selected according to the network mode supported by the vehicle, user settings, or real-time network conditions. Next, the optimal trajectory data is segmented according to the selected segmentation strategy, and the segmented trajectory data is sent segment by segment. Finally, the segmentation strategy and sending frequency are adjusted in real time according to the network conditions to ensure the stability and efficiency of data transmission. In this way, by dynamically adjusting the segmentation strategy to adapt to different network conditions and user needs, intelligent and reliable trajectory optimization and data transmission can be achieved.

[0171] In one embodiment, the optimal trajectory received by the vehicle-side control module from the cloud is a segmented optimal trajectory.

[0172] The optimal trajectory data must at least include the following parameters:

[0173] Timestamp: A time stamp for each trajectory point, indicating that the vehicle should arrive at that point at a specific time.

[0174] Location: includes the longitude, latitude, and altitude of the vehicle at the track point.

[0175] Speed: The speed of the vehicle at the trajectory point.

[0176] Acceleration: The acceleration information of the vehicle at the trajectory point, including longitudinal acceleration and lateral acceleration.

[0177] Heading angle: The direction of the vehicle at a trajectory point, relative to a reference direction.

[0178] Yaw rate: The rate at which the vehicle rotates at a point in the trajectory.

[0179] Steering Angle: The steering angle of the vehicle's front wheels at a trajectory point.

[0180] Curvature: The curvature of the trajectory at this point, describing the degree of curvature of the trajectory.

[0181] Road adhesion coefficient: describes the friction between the vehicle and the road, affecting the vehicle's handling stability.

[0182] Environmental information: such as road type (straight road, curve, ramp), traffic signal information, speed limit information, etc.

[0183] In one embodiment, for segmentation according to time periods, each segmented data packet should include the following parts:

[0184] Segment ID: Uniquely identifies the data packet of this segment.

[0185] Start time and end time: mark the time range of this segment.

[0186] Track point collection: contains detailed information of all track points in the segment.

[0187] In one embodiment, a separate data packet is created for each segment, including all track point information of the segment.

[0188] The data packet includes segment identifier, start time, end time and trajectory point set.

[0189] The data is sent in segments to the vehicle-side control module through the on-board communication module (such as 4G / 5G module).

[0190] Each packet contains a unique segment identifier to ensure the order and integrity of the packets.

[0191] In a certain embodiment, assuming that the transmission is divided into segments every 1 minute according to a time period, the specific process is as follows:

[0192] The first segment of data contains the trajectory points from 12:00 to 12:01.

[0193] The data packet includes the segment identifier, start time, end time and information of all trajectory points within the time period.

[0194] Sent to the vehicle-side control module via the network.

[0195] The second segment of data contains the track points from 12:01 to 12:02.

[0196] The data packet structure is similar, including the new segment identifier, start time, end time, and trajectory point information within the time period.

[0197] Continue to send it to the vehicle-side control module through the network.

[0198] Repeat the above process until all trajectory data are sent.

[0199] In one embodiment, for the segmentation strategy based on data packet size, the global optimal trajectory is segmented according to the maximum data packet size limit. The specific steps are as follows:

[0200] Starting from the first trajectory point, trajectory points are added to the current data packet one by one until the data packet approaches the maximum size limit.

[0201] After adding each track point, check the size of the current data packet. If the data packet size exceeds the limit after adding the track point, stop adding and the current data packet forms a segment.

[0202] Start a new data packet and continue adding track points starting from the last track point that exceeded the limit.

[0203] The above process is repeated until all trajectory points are added to the corresponding data packets.

[0204] In one embodiment, each data packet should contain a unique segment identifier and all track point information of the segment. Then it is sent to the vehicle-side control module through a suitable network transmission protocol. The following are the specific steps:

[0205] Package the segment data, including segment identifier and trajectory point set.

[0206] Use a reliable transport protocol such as TCP to send data packets, ensuring the sequential arrival and integrity of the data.

[0207] After receiving the data packet, the vehicle-side control module sends a confirmation message back to the sender to ensure the integrity of the data packet.

[0208] In one embodiment, the process of segmented transmission based on the size of the data packet assumes that the maximum size of each data packet is 100KB, and the specific process is as follows:

[0209] Start adding from the first trace point until the packet is close to 100KB.

[0210] The current data packet contains multiple track point information, forming the first segment.

[0211] The data packet includes a segment identifier and a set of included trajectory points.

[0212] Sent to the vehicle-side control module via the network.

[0213] Start a new data packet and continue adding from the last trace point that exceeded the limit until the packet approaches 100KB.

[0214] The current packet forms the second fragment.

[0215] The data packet structure is similar, containing a new segment identifier and a set of trajectory points.

[0216] Continue to send it to the vehicle-side control module through the network.

[0217] Repeat the above process until all trajectory data are sent.

[0218] In one embodiment, for the segmentation strategy based on distance, first, the size of each distance segment needs to be determined. For example, assume that each distance segment is 1 kilometer.

[0219] According to the preset distance segments, the global optimal trajectory is segmented. The specific steps are as follows:

[0220] Starting from the first track point, track points are added to the current data packet one by one, and the cumulative distance from the starting point is calculated at the same time.

[0221] After each track point is added, the cumulative distance from the starting point to the current track point is checked. If the cumulative distance exceeds the preset distance segment size (such as 1 km), the addition is stopped and the current data packet forms a segment.

[0222] Start a new data packet and continue adding track points starting from the last track point that exceeds the range segment size.

[0223] The above process is repeated until all trajectory points are added to the corresponding data packets.

[0224] In one embodiment, each data packet should contain a unique segment identifier, the start and end locations, and all track point information of the segment. Then it is sent to the vehicle-side control module through a suitable network transmission protocol. The following are the specific steps:

[0225] Package the segment data, including the segment identifier, start and end locations, and trajectory point collection.

[0226] Use a reliable transport protocol such as TCP to send data packets, ensuring the sequential arrival and integrity of the data.

[0227] In one embodiment, after receiving the data packet, the vehicle-side control module sends a confirmation message back to the sending end to ensure the integrity of the data packet. If the receiving end fails to successfully receive a data packet, the data packet needs to be retransmitted.

[0228] In one embodiment, the network communication module of the vehicle-side control module is responsible for receiving the trajectory data packet from the cloud, while ensuring the integrity and order of the data packet to ensure the accurate transmission of the trajectory information. After receiving the data packet, the data parsing module will parse it and extract key information such as the segment identification, starting and ending positions, and trajectory point set. This allows the vehicle-side control module to effectively prepare for subsequent trajectory tracking tasks.

[0229] In one embodiment, in order to accurately understand the vehicle status and surrounding environment, the vehicle-side control module also needs to obtain information in real time through multiple sensors. For monitoring the vehicle status, the speed sensor obtains the vehicle's driving speed in real time, the accelerometer measures the vehicle's longitudinal and lateral acceleration, the gyroscope provides the vehicle's yaw angular velocity, and the steering wheel angle sensor obtains the vehicle's steering angle.

[0230] In one embodiment, the vehicle is equipped with a variety of sensors to obtain information about the surrounding environment. In addition, other environmental sensors such as laser radar, camera, ultrasonic sensor, etc. can detect surrounding obstacles and lane information, providing comprehensive environmental perception capabilities.

[0231] In one embodiment, the road adhesion coefficient sensor is used to measure the friction between the vehicle and the road surface, and the road curvature sensor obtains the curvature information of the road. These data may be provided by a lidar for data reporting.

[0232] In one embodiment, the trajectory tracking module in the vehicle-side control module will combine the fused vehicle status information and the trajectory data sent from the cloud to generate real-time control instructions for the vehicle, so that the vehicle can dynamically adjust its own driving path according to actual conditions.

[0233] In one embodiment, the vehicle-mounted control module coordinates the steering, braking and acceleration systems of the vehicle to ensure that the vehicle can maintain stability when performing steering operations. Through the feedback control mechanism, the vehicle-side control module dynamically adjusts the vehicle's operating parameters, such as adjusting the steering wheel angle, braking force and accelerator pedal position to cope with real-time changes in road conditions and vehicle status, so that the vehicle can travel smoothly along the planned trajectory. At the same time, the vehicle-side control module will continue to monitor the stability of the vehicle, such as detecting whether there are signs of skidding or loss of control, and take corresponding corrective measures, such as increasing braking force or adjusting the steering angle to restore the stability of the vehicle. This process ensures that the vehicle can always maintain good steering stability and driving safety under various complex road conditions and driving conditions.

[0234] In one embodiment, the vehicle-side MPC optimization process includes:

[0235] Receive and parse the optimal trajectory data from the cloud, including timestamp, position, speed, acceleration, heading angle, yaw rate, steering angle, curvature, road adhesion coefficient and environmental information.

[0236] Construct a linearized state space model, including using the current state and reference trajectory points, linearizing the vehicle dynamics model, and discretizing it into a state transfer matrix and a control input matrix.

[0237] Based on the linearized model and reference trajectory, the cost function and constraints of the optimization problem are constructed, and the control input sequence is solved through the optimization algorithm.

[0238] Apply the resulting control input to the vehicle, update the vehicle state and repeat the above steps.

[0239] In one embodiment, the vehicle is controlled based on the trajectory and stability constraints sent from the cloud and the environment is perceived in real time. The vehicle-side control module receives the optimal trajectory data transmitted from the cloud, including the following parameters:

[0240] Timestamp:

[0241] Location:

[0242] speed:

[0243] Acceleration: , usually only includes the longitudinal acceleration , Normally 0.

[0244] Heading angle:

[0245] Yaw rate:

[0246] Steering angle:

[0247] Curvature:

[0248] Road adhesion coefficient:

[0249] Environmental Information:

[0250] Stability control is performed based on the vehicle-side control module, including local linearization and discretization based on the received trajectory point data starting from the current state.

[0251] Assuming that the sampling time interval is T, the vehicle state space model is a linearized state space model. In order to reduce the calculation amount of the vehicle main control chip and improve the control stability, the linearized and discretized state space model is used in the vehicle MPC optimization. The state space model is specifically: .

[0252] The state vector and control vector .

[0253] The state vector is represented as: .

[0254] State variables describe the state of the vehicle at a certain moment, including information such as position, speed, and direction.

[0255] : The lateral position of the vehicle at time k (usually corresponding to the x coordinate of the world coordinate system or the local coordinate system).

[0256] : The longitudinal position of the vehicle at time k (usually corresponding to the y coordinate of the world coordinate system or the local coordinate system).

[0257] : The heading angle of the vehicle at time k, that is, the angle between the vehicle's driving direction and a reference direction (such as the north or the x-axis direction of the local coordinate system).

[0258] : The speed of the vehicle at time k (usually the longitudinal speed of the vehicle).

[0259] The control vector is expressed as: .

[0260] Control variables Describes the control input of the vehicle at a certain moment, including information such as acceleration and steering angle.

[0261] : The acceleration of the vehicle at time k (usually the longitudinal acceleration).

[0262] : The steering angle of the vehicle at time k, that is, the steering angle of the front wheels.

[0263] System Matrix A k and B k Used to describe the linearized vehicle dynamics model.

[0264] A k : state transfer matrix, describing the state variable x k How it changes over time.

[0265] B k : Control input matrix, describing the control input u k How to affect the state variable x k .

[0266] Specifically, the linearization matrix is:

[0267]

[0268]

[0269] in:

[0270] T: Sampling time interval. The sampling time interval T is used to discretize the time step of the continuous-time dynamics model.

[0271] L: Vehicle wheelbase, i.e. the distance between the front and rear axles. The vehicle wheelbase L is the distance between the front and rear axles. It is a key parameter in the vehicle dynamics model and affects the steering characteristics of the vehicle.

[0272] : Steering angle The tangent of .

[0273] Optionally, use the reference trajectory and linearized model provided by the cloud to construct the optimization problem:

[0274]

[0275] Among them, the cost function weight matrix (Q, R, ) middle:

[0276] Q: State error weight matrix, used to adjust the impact of state deviation in the optimization problem.

[0277] R : Control input weight matrix, used to adjust the influence of control input in the optimization problem.

[0278] : Terminal state error weight matrix, used to adjust the impact of terminal state deviation in the optimization problem.

[0279] The state error weight matrix Q is a diagonal matrix, and each diagonal element corresponds to the weight of the state variable. It determines the degree of penalty for different state errors during the optimization process:

[0280] ;

[0281] in, They are the corresponding state variables The weight of .

[0282] Control input weight matrix Is a diagonal matrix, each diagonal element corresponds to the weight of the control input. It determines the degree of penalty for different control input changes during the optimization process:

[0283] ;

[0284] in, and They are the corresponding control inputs and The weight of .

[0285] Terminal state error weight matrix Is a diagonal matrix used to penalize the weight of the state error at the end of the prediction time domain, which is usually set high to ensure that the terminal state error is small:

[0286] ;

[0287] in, They are terminal state variables The weight of .

[0288] in is the reference trajectory point provided by the cloud. It is obtained from the parameter values ​​of the corresponding trajectory points in the optimal trajectory data transmitted from the cloud.

[0289] Reference trajectory ( ) is the optimal trajectory data provided by the cloud, which contains the state that the vehicle should reach at each moment under ideal circumstances.

[0290] : The state vector of the reference trajectory point at time, including the lateral position of the vehicle at the reference trajectory point at time , vertical position , heading angle and speed , the reference state vector is expressed as: , the reference point It is obtained from the parameter values ​​of the corresponding trajectory points in the optimal trajectory data transmitted from the cloud.

[0291] Directly use the trajectory points provided by the cloud as reference points , the reference point The optimal trajectory data is obtained from the corresponding trajectory point parameter values ​​in the cloud, avoiding complex path planning and trajectory generation calculations on the vehicle side.

[0292] Optionally, a shorter prediction time domain N is selected to reduce the number of optimization variables and the computational complexity. Specifically, the prediction time domain N may be shorter than the driving time corresponding to the data packet sent from the cloud.

[0293] Optionally, a fixed sampling time T may be used to avoid the computational complexity caused by changing the sampling time at any time. Optionally, the fixed sampling time T may be equal to the driving time corresponding to the minimum data packet sent from the cloud.

[0294] In one embodiment, in each control cycle, the following steps are performed:

[0295] Using the linearized model and the current state , predict the state sequence of the next N steps, using the reference trajectory points provided by the cloud as the target point for prediction.

[0296] Based on the predicted state sequence and reference trajectory, the cost function and constraints of the optimization problem are constructed. The cost function mainly includes the cost of state deviation and control input:

[0297] in, and is a weight matrix used to adjust the weights of state deviations and control inputs.

[0298] Solve the above optimization problem through optimization algorithms (such as linear quadratic programming or gradient descent method) to obtain the control input sequence .

[0299] The first control input obtained by optimization Applied to the system to update the vehicle's control status.

[0300] Update vehicle status based on actual vehicle response and sensor feedback , and serves as the initial state of the next cycle.

[0301] Repeat the above steps to continuously perform MPC optimization and control input application to ensure that the vehicle follows the optimal trajectory.

[0302] The present invention utilizes a linearized state space model and reference trajectory points on the vehicle side, constructs a cost function and constraints of the optimization problem, solves the control input sequence and applies it to the vehicle, thereby enhancing the robustness and stability of the control system.

[0303] The present invention sends the optimal trajectory including timestamp, position, speed, acceleration, heading angle, yaw angular velocity, steering angle, curvature, road adhesion coefficient and environmental information in real time through the cloud. The vehicle side can adjust the control strategy in real time, thereby improving the real-time and accuracy of steering control.

[0304] The present invention determines all road sections through the cloud through the path planning algorithm, calculates the actual adhesion coefficient, and combines the nonlinear MPC trajectory optimization model to generate the global optimal steering stability control trajectory to ensure the optimal control effect of the vehicle in the global range. By minimizing the lateral acceleration change rate and the yaw rate change rate, the control input is optimized to further improve the stability and comfort of vehicle steering.

[0305] The present invention uses the MPC model to perform trajectory optimization and real-time control through the collaborative work of the cloud and the vehicle, effectively solving the problems of real-time performance, poor environmental adaptability and lack of global optimization in the prior art, and significantly improving the performance of vehicle steering stability control.

[0306] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, 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 may 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 or 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 disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. Computer readable signal media may also be any computer readable medium other than computer readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0307] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0308] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0309] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the 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 implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square 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 square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart 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.

[0310] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0311] The above introduces the preferred embodiments of the present invention, which is intended to make the spirit of the present invention clearer and easier to understand, but is not intended to limit the present invention. All modifications, substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection outlined by the claims attached to the present invention.

Claims

1. A method for controlling vehicle steering stability, the method comprising: The cloud collects and processes data, optimizes the trajectory based on the MPC model, and sends the optimal trajectory with stability constraints to the vehicle. The vehicle side perceives the environment in real time and performs vehicle steering stability control based on the trajectory sent from the cloud; Perform nonlinear MPC trajectory optimization in the cloud; In the cloud-based MPC trajectory planning model, based on vehicle kinematic and dynamic constraints, the maximum lateral acceleration limit, the maximum yaw rate limit and the minimum curvature radius limit stability-related constraints are introduced into the model; in the objective function, the evaluation index related to steering stability is added to minimize the lateral acceleration change rate and the yaw rate change rate to obtain the control input; After obtaining the actual adhesion coefficients of all road segments, the cloud uses nonlinear MPC model predictive control MPC for trajectory optimization; Among them, all parameters defined in the optimization process of the nonlinear MPC deployed in the cloud include: : The vehicle is at time The longitudinal position of : The vehicle is at time lateral position; : The vehicle is at time The heading angle; : The vehicle is at time speed; : The vehicle is at time acceleration; : The vehicle is at time Steering angle; :At the moment The longitudinal position of the reference trajectory; :At the moment The lateral position of the reference trajectory; : reference speed; : Weight coefficient of each item in the objective function; : Vehicle wheelbase; : sampling time step; g: acceleration due to gravity; : actual adhesion coefficient; : speed limit; : acceleration limit; : Steering angle limit; : lateral acceleration of the vehicle at time t; : radius of curvature; The dynamic model of the vehicle used in the nonlinear MPC deployed in the cloud is expressed by the following difference equation: ； ； ； ； in is the sampling time step; Objective function used in nonlinear MPC deployed in the cloud Aims to minimize the following: The error between the longitudinal position of the vehicle and the reference trajectory; : The error between the vehicle's lateral position and the reference trajectory; : The error between vehicle speed and reference speed; : The square of the acceleration, representing energy consumption or smooth driving; : The square of the steering angle, representing the steering smoothness; The objective function used in the nonlinear MPC deployed in the cloud is of the following form: ； in: : Weight coefficient, used to balance various costs; The actual constraints used in nonlinear MPC deployed in the cloud include physical constraints: Speed ​​Limit: ; Acceleration limit: ; Steering angle limit: ; Maximum lateral acceleration constraint, subject to road adhesion coefficient Impact: ;in is the radius of curvature, g is the acceleration due to gravity; The cloud sends the generated optimal trajectory to the vehicle-side control module; the vehicle-side control module receives and analyzes the optimal trajectory data from the cloud; The vehicle-side MPC optimization process includes: Construct a linearized state space model, including using the vehicle's current state and reference trajectory points, linearizing the vehicle dynamics model, and discretizing it into a state transfer matrix and the control input matrix ; Based on the linearized model and reference trajectory, the cost function and constraints of the optimization problem are constructed, and the control input sequence is solved through the optimization algorithm; applying the resulting control input to the vehicle; Among them, the vehicle is controlled based on the trajectory and stability constraints sent from the cloud and the real-time perception of the environment. The vehicle-side control module receives the optimal trajectory data transmitted from the cloud, including the following parameters: Timestamp: ; Location: ; speed: ; Acceleration: , including only the longitudinal acceleration , is 0; Heading angle: ; Yaw rate: ; Steering angle: ; Curvature: ; Road adhesion coefficient: ; Environmental Information: ; Stability control based on the vehicle-side control module, including local linearization and discretization based on the received reference trajectory points starting from the current state of the vehicle; The sampling time interval is T, the vehicle state space model is a linearized state space model, and the linearized and discretized state space model is used in the vehicle-mounted MPC optimization. The state space model is specifically: ; The state vector of the vehicle's current state at time k is and control vector ; The state vector of the vehicle's current state at time k is expressed as: ; The state variables describe the state of the vehicle at a certain moment, including position, speed, and direction information; : The lateral position of the vehicle at time k; : The longitudinal position of the vehicle at time k; : The heading angle of the vehicle at time k, that is, the angle between the vehicle's driving direction and the reference direction; : The speed of the vehicle at time k; The control vector is expressed as: ; : The acceleration of the vehicle at time k; : The steering angle of the vehicle at time k, that is, the steering angle of the front wheels; : state transfer matrix, describing the state variables how it changes over time; : Control input matrix, describing the control input How to affect state variables ; Specifically, the linearization matrix is: in: T: sampling time interval, which is used to discretize the time step of the continuous-time dynamics model; :Steering angle The tangent of Using the reference trajectory and linearized model provided by the cloud, we construct the optimization problem: ， Among them, the cost function weight matrix Q, R, middle: Q: state error weight matrix, used to adjust the impact of state deviation in the optimization problem; R: control input weight matrix, used to adjust the influence of control input in the optimization problem; : Terminal state error weight matrix, used to adjust the impact of terminal state deviation in the optimization problem; vehicle current state , the reference trajectory points provided by the cloud ; The state error weight matrix Q is a diagonal matrix, each diagonal element corresponds to the weight of the state variable; it determines the degree of penalty for different state errors during the optimization process: ； in, They are the corresponding state variables The weight of The control input weight matrix R is a diagonal matrix, each diagonal element corresponds to the weight of the control input; it determines the degree of penalty for different control input changes during the optimization process: ； in, and They are the corresponding control inputs and The weight of Terminal state error weight matrix : ； in, They are terminal state variables The weight of in is the reference trajectory point provided by the cloud, which is obtained from the parameter value of the corresponding reference trajectory point in the optimal trajectory data transmitted by the cloud; Reference track points It is the optimal trajectory data provided by the cloud, which contains the state that the vehicle should ideally reach at each moment; : The state vector of the reference trajectory point at time k, including the lateral position of the vehicle at the reference trajectory point at time k , vertical position , heading angle and speed , the reference state vector is expressed as: , the reference trajectory point Obtained from the parameter values ​​of the corresponding reference trajectory points in the optimal trajectory data transmitted from the cloud; In each control cycle, the following steps are performed: Using the linearized model and the current state of the vehicle , predicting the future The state sequence of the steps, in which the reference trajectory points provided by the cloud are used As the target point for prediction; Based on the predicted state sequence and reference trajectory, the cost function and constraints of the optimization problem are constructed; the cost function includes the cost of state deviation and control input: in, and is a weight matrix used to adjust the weights of state deviation and control input; the current state of the vehicle , the reference trajectory points provided by the cloud ; The above process means solving the above optimization problem, that is, minimizing the above cost function to obtain the corresponding control input sequence ; The first control input obtained by optimization Applied to the system to update the vehicle's control status; Update the current state of the vehicle at the next moment based on the actual vehicle response and sensor feedback , and used as the initial state of the next cycle; Repeat the above steps to continuously perform MPC optimization and control input application to ensure that the vehicle follows the optimal trajectory; Among them, the trajectory optimization based on the MPC model is used to send the optimal trajectory with stability constraints. In the sending strategy, the corresponding segmentation strategy is selected according to the network mode supported by the vehicle, user settings or real-time network conditions.

2. The vehicle steering stability control method according to claim 1, characterized in that: The cloud collects data, and the data types include destination, estimated arrival time, weather, road type and road adhesion coefficient.

3. The vehicle steering stability control method according to claim 1, characterized in that: The cloud has a unified database structure in which each road segment information corresponds to a location in a high-precision map.

4. The vehicle steering stability control method according to claim 1, characterized in that: The navigation trajectory is determined in the cloud, including determining all road segments through a path planning algorithm according to the destination and arrival time, and querying the corresponding adhesion coefficient correction value from the database according to the road segment ID and weather in the driving path.

5. The vehicle steering stability control method according to claim 4, characterized in that: Calculating the actual adhesion coefficient in the cloud also includes: For each road segment, the weather correction value is applied to calculate the actual adhesion coefficient; the formula is as follows: ; in: : Actual adhesion coefficient; : Adhesion coefficient under standard conditions; : Adhesion coefficient correction value under weather conditions.

6. A vehicle steering stability control system, the system comprising a cloud side and a vehicle side, the cloud side and the vehicle side both comprising a memory and a processor, the memory storing a computer program, the processor being used to execute the computer program to implement any one of the methods of claims 1-5.

Images