A Dynamic Control Method and System for Vehicle Lights Based on Graph Neural Networks and State Machines

Abstract

This disclosure provides a method and system for dynamic control of vehicle lights based on graph neural networks and state machines, relating to the field of vehicle lighting control technology. The method includes acquiring vehicle driving state, vehicle tilt angle changes, and external environment data, converting all data into graph structure data, and modeling the state machine in the adaptive lighting control algorithm as a directed graph, where the vehicle state is used as nodes and external environment data is used as edges, forming a graph structure of vehicle-environment interaction. This graph structure is then input into a graph neural network model to learn the dependencies between nodes in the state machine graph and the state representation of each node. The influence of these dependencies on the rotation angle of the vehicle motor and the headlight adjustment is obtained, dynamically adjusting the C, V, and E modes of the adaptive lighting control algorithm. Based on the classification results, the current headlight illumination angle and motor rotation angle are controlled to ensure optimal matching with the current environment.

Description

Technical Field

[0001] This disclosure relates to the field of vehicle lighting control technology, specifically to a method and system for dynamic control of vehicle lights based on graph neural networks and state machines. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] Traditional ALS algorithms for vehicle lighting play a crucial role in vehicle lighting control, but their decisions are primarily based on preset rules and patterns, lacking the sensitivity and adaptability to real-time environmental changes. With the development of vehicle intelligence, higher demands are being placed on vehicle lighting systems, including the need for precise perception of complex road environments, dynamic adjustment of lighting patterns, and improved energy efficiency and safety.

[0004] There are relatively few artificial intelligence algorithms in the field of vehicle lighting algorithms, and graph neural networks are even less used in vehicle lighting algorithms. The traditional ALS algorithm is generally implemented by accepting CAN signals, establishing a corresponding state machine system, and establishing rules to output the rotation angle of the headlight motor. Furthermore, the traditional static tilt adjustment function of vehicle headlights refers to the change in the angle and position of the headlights caused by the change in the vehicle body tilt angle when the vehicle is stationary (vehicle speed = 0 km / h) due to heavy load and changes in the load position. The vehicle lighting algorithm system should make the correct vertical tilt adjustment of the headlights.

[0005] Therefore, traditional automotive headlight ALS algorithm systems are rule-based systems and still have many shortcomings and problems:

[0006] Traditional ALS (Advanced Driver Assistance Systems) rely primarily on pre-defined rules, often based on assumptions about specific scenarios, such as adjusting headlight angles when a vehicle turns. However, these pre-defined rules cannot respond in real-time to complex and ever-changing driving environments, such as the sudden appearance of pedestrians, animals, or non-standard road signs. This lack of sensitivity and adaptability to real-time environmental changes prevents the system from reacting promptly and appropriately to unexpected situations. Rule-based ALS systems are designed with a fixed set of rules, which are difficult to modify or expand once set. However, real-world driving environments are dynamic, including weather, lighting conditions, and traffic conditions. Fixed rule sets struggle to cope with environmental diversity, especially in extreme weather or complex traffic conditions, where the system may fail to provide optimal lighting strategies, impacting driving safety and energy efficiency. Because rule-based ALS systems rely on pre-defined general rules, they cannot be personalized based on individual driver preferences or driving styles. For example, some drivers may prefer stronger lighting at night, while others may prioritize energy conservation. Rule-based systems struggle to meet these personalized needs, resulting in a limited user experience. Summary of the Invention

[0007] To address the aforementioned issues, this disclosure proposes a dynamic control method and system for vehicle lights based on graph neural networks and state machines. The method constructs a graph structure comprising the state machine, vehicle, road, and environmental conditions. A graph neural network is used to learn the state representation of each node in the graph, providing a decision-making basis for the state transition of the state machine. This achieves a shift from traditional rule-driven to data-driven control, ensuring optimal matching between the headlight motor rotation angle and the headlight illumination angle with the current environmental conditions. Furthermore, the representation of node states is optimized to achieve refined and intelligent control of the vehicle light system.

[0008] According to some embodiments, the present disclosure adopts the following technical solutions:

[0009] A dynamic control method for vehicle lights based on graph neural networks and state machines includes:

[0010] Acquire vehicle driving status, vehicle tilt angle changes, and external environment data, and convert all data into graph structure data;

[0011] Based on graph structure data, an adaptive vehicle lighting control algorithm is introduced. The state machine in the adaptive vehicle lighting control algorithm is modeled as a directed graph, where the vehicle state is used as a node and the external environment data is used as an edge, forming a graph structure for vehicle-environment interaction.

[0012] The graph structure of vehicle-environment interaction is input into the graph neural network model to learn the dependencies between nodes in the state machine graph and the state representation of each node. The influence of the dependencies between nodes on the rotation angle of the vehicle motor and the headlight adjustment is obtained. The adaptive headlight control algorithm is dynamically adjusted to three modes: C, V, and E, and the classification results of the three modes are output. Based on the classification results, the current headlight illumination angle and motor rotation angle are controlled to ensure the best matching state with the current environment.

[0013] According to some embodiments, the present disclosure adopts the following technical solutions:

[0014] A vehicle lighting dynamic control system based on graph neural networks and state machines includes:

[0015] The data acquisition module is used to acquire vehicle driving status, vehicle tilt angle changes, and external environment data, and convert all data into graph structure data.

[0016] The graph structure modeling module is used to introduce an adaptive vehicle lighting control algorithm based on graph structure data. It models the state machine in the adaptive vehicle lighting control algorithm as a directed graph, where the vehicle state is used as a node and the external environment data is used as an edge, forming a graph structure for vehicle-environment interaction.

[0017] The adaptive control module is used to input the graph structure of vehicle-environment interaction into the graph neural network model, learn the dependencies between nodes in the state machine graph and the state representation of each node, obtain the influence of the dependencies between nodes on the rotation angle of the vehicle motor and the headlight adjustment, dynamically adjust the three modes of adaptive headlight control algorithm (C, V, E), and output the classification results of the three modes. Based on the classification results, the current headlight illumination angle and motor rotation angle are controlled to ensure the best matching state with the current environment.

[0018] According to some embodiments, the present disclosure adopts the following technical solutions:

[0019] A computer program product includes a computer program that, when executed by a processor, implements the aforementioned dynamic control method for vehicle lights based on graph neural networks and state machines.

[0020] According to some embodiments, the present disclosure adopts the following technical solutions:

[0021] A non-transitory computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the aforementioned dynamic control method for vehicle lights based on graph neural networks and state machines.

[0022] According to some embodiments, the present disclosure adopts the following technical solutions:

[0023] An electronic device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the vehicle lighting dynamic control method based on graph neural networks and state machines.

[0024] Compared with the prior art, the beneficial effects of this disclosure are as follows:

[0025] This disclosure presents a dynamic control method for vehicle lights based on graph neural networks and state machines. It constructs graph data structures to train a Dynamic State Aware Graph Neural Network (DSGNN). The DSGNN handles complex graph connections and learns node representations through information propagation, enabling dynamic adjustment of the motor rotation angle. The DSGNN learns dependencies between nodes through message passing, intelligently adjusting the motor rotation angle and vehicle light illumination mode. The training results from the DSGNN are used to dynamically adjust the C, V, and E modes of the ALS (Autonomous Lighting System) to ensure optimal matching of the illumination angle with current environmental conditions. The classification results output by the DSGNN determine which of the three modes (C, V, and E) the ALS belongs to, thus outputting the required motor adjustment angle.

[0026] This disclosure presents a dynamic control method for vehicle lights based on graph neural networks and state machines. By introducing a Dynamic State Aware Graph Neural Network (DSGNN), the vehicle lighting system can analyze environmental changes in real time, such as road type, weather conditions, and light levels, based on data collected by sensors during vehicle operation. This allows for intelligent adjustment of the lighting mode, significantly outperforming traditional rule-driven ALS algorithms. The dynamic graph data processing capability of the graph neural network ensures the system can quickly respond to changes in vehicle state, such as rapid acceleration, deceleration, and hill driving, automatically adjusting the headlight angle to improve driving safety and lighting efficiency. Furthermore, to address new scenarios, the system can continuously update the graph database based on the new scenario to optimize the model trained by the graph neural network. Through continuous learning and optimization, the vehicle lighting system can continuously adjust the motor rotation angle and headlight illumination angle based on accumulated real-time data, achieving optimal matching with environmental conditions.

[0027] This disclosure presents a dynamic control method for vehicle lights based on graph neural networks and state machines. Unlike traditional ALS systems that require customized rules based on different vehicle models and styles, this disclosure transforms the state machine into a graph structure. The graph neural network can learn the complex relationships between nodes and edges, providing data support for intelligent mode switching. Through comparison, it was found that this disclosure can reduce power consumption. By intelligently adjusting the motor rotation angle, the system can effectively reduce unnecessary light adjustments, reduce energy consumption, and improve overall energy efficiency. Attached Figure Description

[0028] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0029] Figure 1 This is a schematic diagram of the traditional ALS algorithm implementation process. Detailed Implementation

[0030] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0031] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0032] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this disclosure. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0033] Example 1

[0034] One embodiment of this disclosure provides a dynamic control method for vehicle lights based on graph neural networks and state machines, the steps of which include:

[0035] Step 1: Acquire vehicle driving status, vehicle tilt angle changes, and external environment data, and convert all data into graph structure data;

[0036] Step 2: Based on graph structure data, an adaptive headlight control algorithm is introduced. The state machine in the adaptive headlight control algorithm is modeled as a directed graph, where the vehicle state is the node and the external environment data is the edge, forming a graph structure for vehicle-environment interaction.

[0037] Step 3: Input the graph structure of vehicle-environment interaction into the graph neural network model, learn the dependencies between nodes in the state machine graph and the state representation of each node, obtain the influence of the dependencies between nodes on the rotation angle of the vehicle motor and the headlight adjustment, dynamically adjust the three modes of adaptive headlight control algorithm (C, V, E), and output the classification results of the three modes. Based on the classification results, control the current headlight illumination angle and motor rotation angle to ensure the best matching state with the current environment.

[0038] As one embodiment, this disclosure discloses a dynamic control method for vehicle lights based on graph neural networks and state machines. Graph neural networks utilize a graph neural network structure, which can process data with complex graph connection relationships. The basic idea of ​​graph neural networks is to learn node representations in a graph by propagating information through relationships between nodes. The neural network learns the representation of each node, which includes the node's structural and feature information. Similarly, it learns the representation of each edge, reflecting the relationship between the two nodes connected by the edge. Pooling operations are used to obtain the representation of the entire graph, which can be used for graph classification tasks. Different automakers have different state machine representations for adaptive headlights. This disclosure constructs a directed graph from the large number of state machine graphs accumulated by automakers and uses it as input to the graph neural network for learning.

[0039] Traditional ALS (vehicle control algorithm) algorithms generally accept CAN signals, establish a corresponding state machine system, and establish rules to output the rotation angle of the headlight motor.

[0040] Traditional vehicle headlight static tilt adjustment refers to the adjustment of the headlights' angle and position when the vehicle is stationary (vehicle speed = 0 km / h). Changes in vehicle load and load position cause variations in the vehicle's tilt angle, resulting in changes in the headlights' illumination angle and position. The headlight algorithm system should then correctly adjust the headlights' tilt angle. Static dimming should be applicable to the following vehicle conditions:

[0041] 1. Driver and passengers getting on and off the vehicle

[0042] 2. Movement of occupants within the vehicle

[0043] 3. Changes in luggage load

[0044] 4. Fuel loss

[0045] When the vehicle is stationary, the headlight algorithm constantly monitors changes in the vehicle's tilt angle, maintaining a constant number of samples and actively filtering noise. Within a time T after a change in the tilt angle, the ECU drives the dimming motor to adjust the headlights. To avoid excessively frequent headlight adjustments that could affect the driver's perception, very small changes in the tilt angle will not trigger the headlight algorithm to adjust the headlight angle.

[0046] If the headlights turn off during static pitch adjustment, the adjustment will immediately disengage and the headlights will return to their initial position. If the headlights turn back on and the conditions for starting static pitch adjustment are met, the headlights will then adjust to the correct position.

[0047] Considering various operating conditions during vehicle operation (rapid acceleration and deceleration, inclines, bumpy roads), the dynamic tilt adjustment function of the vehicle's headlights needs to be divided into the following modes:

[0048] 1. Dynamic pitch adjustment function for rapid acceleration and deceleration

[0049] 2. Ramp dynamic pitch adjustment function

[0050] 3. Bumpy road surface recognition and dynamic pitch adjustment function

[0051] During rapid acceleration, the vehicle's upward tilt causes the headlights to rise higher. The headlight algorithm system should adjust the headlights to lower the beam angle to prevent glare for oncoming drivers. During rapid deceleration, the vehicle's downward tilt lowers the headlights, resulting in a rapid decrease in illumination distance. The headlight algorithm system should adjust the headlights to raise the beam angle to maintain a reasonable illumination distance on the road and ensure driving safety. During acceleration and deceleration, the vehicle's pitch angle changes very rapidly, and the response time of the headlight algorithm system's dimming system should be less than a certain time.

[0052] When the vehicle speed increases beyond a certain speed, it is considered to have entered rapid acceleration mode. Once in rapid acceleration mode, the headlights immediately adjust to lower the beam angle within a very short time. When the vehicle speed increases less than a certain speed, it is considered to have exited rapid acceleration mode. Once out of rapid acceleration mode, the headlights immediately adjust to return to the normal beam angle within a very short time. When the vehicle speed decreases beyond a certain speed, it is considered to have entered rapid deceleration mode. Once in rapid deceleration mode, the headlights immediately adjust to raise the beam angle within a very short time. When the vehicle speed decreases less than a certain speed, it is considered to have exited rapid deceleration mode. Once out of deceleration mode, the headlights immediately adjust to return to the normal beam angle within a very short time.

[0053] If the headlights are off during the rapid acceleration / deceleration pitch adjustment function, the function will immediately disengage and the headlights will return to their initial position. If the headlights are turned on again and the conditions for activating the rapid acceleration / deceleration pitch adjustment function are met, the headlights will immediately adjust to the correct position. This can be optimized based on real-vehicle testing.

[0054] When a vehicle travels on a slope for more than T hours, the vehicle's center of gravity shifts due to the slope's inclination, causing a change in the vehicle's tilt angle. At this point, the headlight algorithm system should detect the vehicle's tilt using the height sensor and adjust the headlight angle to compensate for the vehicle's pitch caused by the slope, providing an appropriate lighting distance on the slope.

[0055] If the headlights are off during the ramp dimming and tilt adjustment function, the function will immediately disengage and the headlights will return to their initial position. If the headlights are turned on again and the conditions for starting the ramp dimming and tilt adjustment function are met, the headlights will immediately adjust to the correct position.

[0056] Therefore, in view of the problems existing in the above-mentioned prior art, this disclosure provides a dynamic control method for vehicle lights based on graph neural networks and state machines, and the specific implementation process is as follows:

[0057] Step 1: Acquire vehicle driving status, vehicle tilt angle changes, and external environment data, and convert all data into graph structure data;

[0058] Specifically, sensors are used to collect vehicle driving status, vehicle tilt angle changes, and external environment data in real time, and all data are converted into graph structure data. The vehicle driving status includes vehicle speed, vehicle position, and vehicle driving direction. The external environment data includes road type, weather data, and light level. The attributes of the edges are the switching conditions and migration conditions between various vehicle states.

[0059] Step 2: Based on graph structure data, an adaptive headlight control algorithm is introduced. The state machine in the adaptive headlight control algorithm is modeled as a directed graph, where the vehicle state is the node and the external environment data is the edge, forming a graph structure for vehicle-environment interaction.

[0060] An adaptive headlight control algorithm is introduced, and the state machine of the adaptive headlight control is represented as a directed graph, in which the vehicle state and environmental conditions are modeled as attributes of nodes and edges, forming a graph representation of vehicle-environment interaction.

[0061] Specifically, the first step is to construct a graph structure data. Based on Chery's extensive data on the ALS algorithm for vehicle lights in real-world testing scenarios, and various previously accumulated state machine structure data, a graph data structure for training the graph neural network is constructed. Node features include, but are not limited to, vehicle state, position, speed, and direction; edge attributes include road type, weather conditions, and light level.

[0062] Step 3: Input the graph structure of vehicle-environment interaction into the graph neural network model, learn the dependencies between nodes in the state machine graph and the state representation of each node, obtain the influence of the dependencies between nodes on the rotation angle of the vehicle motor and the headlight adjustment, dynamically adjust the C, V, and E modes of the adaptive headlight control algorithm, and output the classification results of the three modes. Based on the classification results, control the current headlight illumination angle and motor rotation angle to ensure the best matching state with the current environment.

[0063] The attributes of the edges are the switching and transition conditions between various states. Graph neural networks are used to handle complex graph connections and learn node representations through information propagation.

[0064] This disclosure proposes a Dynamic State-Aware Graph Neural Network (DSGNN) for real-time adaptive vehicle lighting control. This network achieves data-driven C / V / E pattern decision-making through a dynamic vehicle-environment interaction graph (nodes = vehicle state, edges = environmental conditions and transition rules), and comprises four main modules:

[0065] 1. Dynamic Graph Builder: Node Feature h iv =[velocity, Δθ, GPS, a] x ,a y ], edge feature h ije =[Road type, light intensity, weather], depending on Data updates the graph structure G_t;

[0066] 2. Multi-head edge-aware convolutional layer: Message generation m ij =MLP(h iv ,h ije Attention weights Node update

[0067] 3. Hierarchical graph pooling: Graph-level embedding h_G = connectivity (mean pooling, max pooling) is generated by TOP-K clustering of nodes based on L2 norm.

[0068] 4. Adaptive classifier: Outputs pattern probability The control command θ_ is adjusted based on a probability threshold (e.g., -5°±α when p_C>0.7). Experiments show that state graph-based training achieves 93.7% classification accuracy and 18ms latency, reducing invalid motor adjustments by 41% compared to traditional ALS. Core innovations include: edge feature-driven state transition modeling, real-time node clustering and pooling, and CAN feedback for online parameter optimization (W_a, W_g).

[0069] Where Δθ is the tilt angle change, a_x / y = acceleration, α / β is the environmental compensation coefficient, SUM is the summation, and σ is the activation function.

[0070] Dynamic State Aware Graph Neural Network (DSGNN) can dynamically adjust the motor rotation angle based on a graph system (including factors such as rapid acceleration / deceleration, slopes, bumpy roads, road type, weather conditions, light levels, and C, V, E modes under different conditions). The graph neural network learns the dependencies between nodes through a message passing mechanism, intelligently adjusting the motor rotation angle and the headlight illumination mode. The results trained using the graph neural network dynamically adjust the C, V, and E modes of the ALS to ensure the optimal match between the illumination angle and the current environmental conditions. The classification results output by the graph neural network determine which of the three modes (C, V, E) belongs to the ALS, thus outputting the required motor adjustment angle.

[0071] Specifically, Dynamic State Aware Graph Neural Networks (DSGNNs) process complex graph-structured data. Through message passing between nodes, they learn the state representation of each node, thereby optimizing motor control and headlight adjustment in the ALS algorithm. For different vehicle states and environmental conditions, the graph neural network can dynamically adjust the control strategy based on the input state machine graph to ensure the optimal lighting mode. The final output of the graph neural network is pre-classified category information. By using this category information, the computational load is reduced, enabling more precise adjustment of the motor angle.

[0072] This disclosure enables the automatic learning of dependencies between nodes and how these relationships affect headlight adjustment during model training through a message passing mechanism. This dynamic graph data processing capability allows the graph neural network to update node states in real time. Using the state machine in the headlight ALS algorithm as input, the graph neural network determines state transitions, achieving intelligent switching from one state to another. This closed-loop mechanism allows the headlight system to continuously learn and optimize during operation. The graph neural network can handle complex graph structure data, learning the state representation of each node in the graph. These states can be high-dimensional embeddings of node features, providing a decision-making basis for state machine transitions and realizing a shift from traditional rule-driven to data-driven approaches. This ensures optimal matching between the headlight motor rotation angle and the headlight illumination angle with current environmental conditions. Further optimization of node state representation enables refined and intelligent control of the headlight system.

[0073] This disclosure verifies the system's response speed and adjustment accuracy under conditions such as rapid acceleration, rapid deceleration, and hill driving through real-vehicle testing. Furthermore, by collecting and analyzing a large amount of driving data, the parameters of the graph neural network model are continuously optimized to improve the system's adaptability and stability under various environmental conditions. Experimental results show that in scenarios such as rapid acceleration and deceleration and hill driving, the ALS system fused with the graph neural network can significantly reduce glare, maintain a reasonable lighting distance, and improve driving safety. It also shows a significant improvement in energy efficiency and a reduction in energy consumption.

[0074] Example 2

[0075] One embodiment of this disclosure provides a vehicle lighting dynamic control system based on graph neural networks and state machines, including:

[0076] The data acquisition module is used to acquire vehicle driving status, vehicle tilt angle changes, and external environment data, and convert all data into graph structure data.

[0077] The graph structure modeling module is used to introduce an adaptive vehicle lighting control algorithm based on graph structure data. It models the state machine in the adaptive vehicle lighting control algorithm as a directed graph, where the vehicle state is used as a node and the external environment data is used as an edge, forming a graph structure for vehicle-environment interaction.

[0078] The adaptive control module is used to input the graph structure of vehicle-environment interaction into the graph neural network model, learn the dependencies between nodes in the state machine graph and the state representation of each node, obtain the influence of the dependencies between nodes on the rotation angle of the vehicle motor and the headlight adjustment, dynamically adjust the three modes of adaptive headlight control algorithm (C, V, E), and output the classification results of the three modes. Based on the classification results, the current headlight illumination angle and motor rotation angle are controlled to ensure the best matching state with the current environment.

[0079] Example 3

[0080] One embodiment of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned vehicle lighting dynamic control method based on graph neural networks and state machines.

[0081] Example 4

[0082] One embodiment of this disclosure provides a non-transitory computer-readable storage medium for storing computer instructions. When these computer instructions are executed by a processor, they implement the aforementioned vehicle lighting dynamic control method based on graph neural networks and state machines.

[0083] Example 5

[0084] One embodiment of this disclosure provides an electronic device, including a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the vehicle lighting dynamic control method based on graph neural networks and state machines.

[0085] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0086] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0087] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.

Claims

1. A dynamic control method for vehicle lights based on graph neural networks and state machines, characterized in that, include: Acquire vehicle driving status, vehicle tilt angle changes, and external environment data, and convert all data into graph structure data; Based on graph structure data, an adaptive vehicle lighting control algorithm is introduced. The state machine in the adaptive vehicle lighting control algorithm is modeled as a directed graph, where the vehicle state is used as a node and the external environment data is used as an edge, forming a graph structure for vehicle-environment interaction. The graph structure of vehicle-environment interaction is input into the graph neural network model to learn the dependencies between nodes in the state machine graph and the state representation of each node. The influence of the dependencies between nodes on the rotation angle of the vehicle motor and the headlight adjustment is obtained. The adaptive headlight control algorithm is dynamically adjusted to three modes: C, V, and E, and the classification results of the three modes are output. Based on the classification results, the current headlight illumination angle and motor rotation angle are controlled to ensure the best matching state with the current environment.

2. The vehicle lighting dynamic control method based on graph neural networks and state machines as described in claim 1, characterized in that, The vehicle driving status includes vehicle speed, vehicle position, and vehicle direction. The external environment data includes road type, weather data, and light level. The edge attributes are the switching and migration conditions between various vehicle states.

3. The vehicle lighting dynamic control method based on graph neural networks and state machines as described in claim 1, characterized in that, In the graph neural network model, the dependencies between nodes in the state machine graph structure and the impact of these dependencies on headlight adjustment are automatically learned through message passing and node feature updates. The structural features of each node and each edge are extracted, the node states are updated through the message passing mechanism, the graph data is dynamically learned, and the optimized node state representation is output.

4. The vehicle lighting dynamic control method based on graph neural networks and state machines as described in claim 3, characterized in that, The optimized node state representation of the output is used to make state transition decisions for the state machine. The node state representation is used to determine the state transition, realizing intelligent switching from one state to another. The C, V, and E modes of the adaptive headlight control algorithm are dynamically adjusted to ensure that the rotation angle of the headlight motor and the illumination angle of the headlight are optimally matched with the current environmental conditions.

5. The vehicle lighting dynamic control method based on graph neural networks and state machines as described in claim 1, characterized in that, When the vehicle is stationary, the ECU continuously monitors changes in the vehicle's state. Within a time T after the vehicle's tilt angle changes, the ECU drives the dimming motor to complete the headlight adjustment. During static tilt adjustment, if the headlight signal turns off, the ECU immediately disengages from static tilt adjustment and adjusts the headlights back to their initial position. When the headlight signal turns back on and the static tilt adjustment start conditions are met, the headlights are adjusted to the correct static tilt adjustment position.

6. The vehicle lighting dynamic control method based on graph neural networks and state machines as described in claim 1, characterized in that, When the vehicle speed increases beyond a certain speed, it is determined that it has entered the rapid acceleration mode, and the headlights will adjust to lower the beam angle within a very short time. When the vehicle speed increases less than a certain speed, it is determined that it has exited the rapid acceleration mode, and the headlights will adjust to return to the normal beam angle within a very short time. When the vehicle speed decreases beyond a certain speed, it is determined that it has entered the rapid deceleration mode, and the headlights will adjust to raise the beam angle within a very short time. When the vehicle speed decreases less than a certain speed, it is determined that it has exited the rapid deceleration mode, and the headlights will adjust to return to the normal beam angle within a very short time.

7. A vehicle lighting dynamic control system based on graph neural networks and state machines, characterized in that, include: The data acquisition module is used to acquire vehicle driving status, vehicle tilt angle changes, and external environment data, and convert all data into graph structure data. The graph structure modeling module is used to introduce an adaptive vehicle lighting control algorithm based on graph structure data. It models the state machine in the adaptive vehicle lighting control algorithm as a directed graph, where the vehicle state is used as a node and the external environment data is used as an edge, forming a graph structure for vehicle-environment interaction. The adaptive control module is used to input the graph structure of vehicle-environment interaction into the graph neural network model, learn the dependencies between nodes in the state machine graph and the state representation of each node, obtain the influence of the dependencies between nodes on the rotation angle of the vehicle motor and the headlight adjustment, dynamically adjust the three modes of adaptive headlight control algorithm (C, V, E), and output the classification results of the three modes. Based on the classification results, the current headlight illumination angle and motor rotation angle are controlled to ensure the best matching state with the current environment.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the vehicle lighting dynamic control method based on graph neural networks and state machines as described in any one of claims 1-6.

9. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the vehicle lighting dynamic control method based on graph neural networks and state machines as described in any one of claims 1-6.

10. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the vehicle lighting dynamic control method based on graph neural networks and state machines as described in any one of claims 1-6.

Images