A vehicle lamp multi-objective optimization method, device and equipment based on a particle swarm algorithm

Abstract

The application discloses a vehicle lamp multi-objective optimization method, device and equipment based on a particle swarm algorithm, and relates to the field of optoelectronic technology. The method comprises the following steps: detecting the communication distance and the ambient light intensity of a target object in front in real time, and obtaining the minimum exit light flux threshold under the current vehicle driving mode; taking the communication distance and the ambient light intensity as environmental variables, taking the laser power, the beam divergence angle and the modulation depth as optimization variables, and under the forced constraint of meeting the minimum exit light flux threshold, iteratively calculating a comprehensive evaluation function by using the particle swarm algorithm, and outputting a globally optimal control parameter combination; generating a high-frequency driving signal according to the optimal modulation depth, driving a blue laser diode to output a blue laser beam carrying a communication code at an optimal laser power, and dynamically adjusting a projection optical module according to the optimal beam divergence angle to change the divergence angle of the exit white light illumination beam. The method significantly improves the adaptive ability and multi-objective comprehensive performance of the intelligent vehicle lamp in a dynamic traffic environment.

Description

Technical Field

[0001] This invention relates to the field of optoelectronics technology, and more specifically to a method, apparatus, and device for multi-objective optimization of vehicle lights based on particle swarm optimization algorithm. Background Technology

[0002] With the rapid development of intelligent transportation and vehicle-to-everything (V2X) technologies, automotive headlights are transforming from simple nighttime illumination tools into multifunctional intelligent nodes integrating lighting, environmental perception, and visible light communication. Systems based on blue laser-excited yellow phosphors, due to their high brightness, long illumination distance, and the high-frequency modulation characteristics inherent in laser diodes, have become an ideal hardware platform for realizing long-distance inter-vehicle / vehicle-to-infrastructure optical communication.

[0003] However, existing vehicle lighting design methods face the following technical bottlenecks: First, in vehicle lighting communication systems that use blue light to excite phosphors, there is an inherent contradiction between lighting, communication, and biosafety. To obtain high luminous flux and high color rendering, it is usually necessary to increase the blue light component, but this can easily cause blue light hazard to the retinas of pedestrians or drivers ahead. At the same time, because the afterglow effect of phosphors filters out high-frequency modulation signals, the high-speed communication bandwidth of the system is highly dependent on the residual direct blue light that is not absorbed by the phosphors and is directly transmitted. Therefore, increasing communication bandwidth (requiring an increase in residual blue light) and ensuring human photobiological safety (requiring a reduction in blue light) constitute a strong nonlinear physical conflict. Second, in real traffic scenarios, drastic fluctuations in ambient illuminance can lead to a sharp deterioration in the communication signal-to-noise ratio, and changes in target distance can cause exponential changes in blue light radiation density. Traditional static design or conventional linear weighted optimization methods cannot adaptively optimize according to real-time changes in detection distance and ambient illuminance, making it difficult to simultaneously meet basic lighting compliance, blue light safety limits, and visible light communication bandwidth requirements during dynamic driving.

[0004] Therefore, there is an urgent need for a multi-objective optimization method for vehicle lights based on particle swarm optimization that can overcome the above-mentioned shortcomings. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-objective optimization method, device, and equipment for vehicle lights based on particle swarm optimization (PSO). By real-time detection of communication distance and ambient illuminance, and using laser power, beam divergence angle, and modulation depth as joint optimization variables, the PSO algorithm dynamically outputs the globally optimal combination of control parameters while satisfying the mandatory constraints of basic lighting. This achieves adaptive and collaborative optimization of vehicle lighting, blue light safety, and visible light communication performance. Compared with traditional static design or linear weighted methods, this method can adjust the beam divergence angle and modulation depth in real time according to the distance to the target ahead and drastic changes in ambient illuminance, effectively balancing blue light biosafety in close-range scenarios and communication bandwidth requirements in long-range scenarios, while ensuring compliance with basic lighting regulations. Therefore, this method significantly improves the adaptive capability and multi-objective comprehensive performance of intelligent vehicle lights in dynamic traffic environments, breaking down the physical performance barriers between lighting, communication, and safety.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a multi-objective optimization method for vehicle lights based on particle swarm optimization, the method comprising:

[0008] Real-time detection of the communication distance to the target object in front of the vehicle and the ambient light level of the current environment, and acquisition of the minimum outgoing light flux threshold in the current vehicle driving mode;

[0009] Using communication distance and ambient illuminance as environmental variables, and laser power, beam divergence angle, and modulation depth as optimization variables, a particle swarm optimization algorithm is used to iteratively calculate the comprehensive evaluation function under the mandatory constraint of meeting the minimum emitted light flux threshold, and output the globally optimal combination of control parameters. The globally optimal combination of control parameters includes the optimal laser power, the optimal beam divergence angle, and the optimal modulation depth.

[0010] A high-frequency driving signal is generated based on the optimal modulation depth to drive the blue laser diode to output a blue laser beam carrying communication codes with optimal laser power;

[0011] The projection optics module is dynamically adjusted according to the optimal beam divergence angle to change the divergence angle of the emitted white light illumination beam.

[0012] In some embodiments, communication distance and ambient illuminance are used as environmental variables, and laser power, beam divergence angle, and modulation depth are used as optimization variables. Under the mandatory constraint of satisfying the minimum emitted light flux threshold, a particle swarm optimization algorithm is used to iteratively calculate the comprehensive evaluation function and output the globally optimal combination of control parameters, including:

[0013] For each particle in the particle swarm, the combination of its carried parameters is substituted into the preset physical model, and combined with the real-time detected communication distance and ambient illuminance, the blue light hazard assessment value, communication signal-to-noise ratio and total emitted light flux are calculated respectively.

[0014] A penalty function is constructed based on the blue light hazard assessment value and the communication signal-to-noise ratio, and the safety weight and communication bandwidth weight are dynamically calculated based on the current communication distance.

[0015] A fitness objective function is constructed based on security weights, communication bandwidth weights, and a penalty function. The fitness objective function is then used as a comprehensive evaluation function for particle swarm optimization, outputting the optimal laser power, optimal beam divergence angle, and optimal modulation depth.

[0016] In some embodiments, calculating the blue light hazard assessment value includes:

[0017] Extract the residual high-frequency blue light power transmitted through the phosphor layer, and calculate the irradiated area of ​​the blue light beam at the target distance based on the beam divergence angle and communication distance.

[0018] The blue light radiation power density at different test distances was converted to a unified benchmark distance using a squared scaling factor, and combined with the environmental limit for the no-hazard exemption level, the blue light hazard assessment value was calculated.

[0019] The signal-to-noise ratio (SNR) of communication is calculated in the following way:

[0020] The signal-to-noise ratio (SNR) is calculated based on the effective blue light signal power at the communication distance, background shot noise caused by ambient light intensity, and thermal noise.

[0021] In some embodiments, the security weight and communication bandwidth weight are dynamically calculated based on the current communication distance, including:

[0022] , ;

[0023] Where L is the current communication distance. The preset critical handover distance threshold is β, where β is the steepness coefficient; when the communication distance is less than At that time, safety weight Dominant; when the communication distance is greater than At that time, communication bandwidth weight Dominant.

[0024] In some embodiments, the penalty function includes a blue light safety penalty term and a communication signal-to-noise ratio penalty term:

[0025] The signal-to-noise ratio threshold for the forward error correction communication baseline is preset;

[0026] The first penalty item is constructed based on the blue light hazard assessment value;

[0027] A second penalty term is constructed based on the communication signal-to-noise ratio and the signal-to-noise ratio threshold;

[0028] The fitness objective function is expressed as:

[0029] ;

[0030] in, For safety utility function, For bandwidth utility function, This is a combined penalty term consisting of the first penalty term and the second penalty term. This is a penalty term for luminous flux constraints.

[0031] In some embodiments, the iterative optimization process of the particle swarm optimization algorithm includes:

[0032] Initialize the particle swarm and set the three-dimensional optimization variable set; the three-dimensional optimization variable set includes laser power, beam divergence angle and modulation depth.

[0033] Calculate the fitness value of each particle in the current particle swarm, and update the individual's historical best position and global best position;

[0034] The particle swarm is driven to move in a multidimensional solution space, and the iteration is repeated until the maximum number of iterations or the convergence condition is met.

[0035] Extract the global optimal solution combination, output the optimal laser power and optimal modulation depth to the communication modulation module, and output the optimal beam divergence angle to the projection optics module.

[0036] Secondly, the present invention also provides a multi-objective optimization device for vehicle lights based on particle swarm optimization algorithm, the device comprising:

[0037] The information detection module is used to detect the communication distance of the target object in front of the vehicle and the ambient light intensity of the current environment in real time, and to obtain the minimum outgoing light flux threshold under the current vehicle driving mode.

[0038] The parameter solving module is used to iteratively calculate the comprehensive evaluation function using the communication distance and ambient illuminance as environmental variables, and the laser power, beam divergence angle and modulation depth as optimization variables, under the mandatory constraint of meeting the minimum emitted light flux threshold, and output the globally optimal combination of control parameters. The globally optimal combination of control parameters includes the optimal laser power, the optimal beam divergence angle and the optimal modulation depth.

[0039] The beam output module is used to generate a high-frequency drive signal based on the optimal modulation depth, which drives the blue laser diode to output a blue laser beam carrying communication codes at the optimal laser power.

[0040] An angle adjustment module is used to dynamically adjust the projection optics module according to the optimal beam divergence angle, thereby changing the divergence angle of the emitted white light illumination beam.

[0041] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the multi-objective optimization method for vehicle lights based on the particle swarm optimization algorithm provided in the first aspect.

[0042] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the multi-objective optimization method for vehicle lights based on the particle swarm optimization algorithm provided in the first aspect.

[0043] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the multi-objective optimization method for vehicle lights based on the particle swarm optimization algorithm provided in the first aspect.

[0044] The beneficial effects of this invention are as follows: By real-time detection of communication distance and ambient illuminance, and using laser power, beam divergence angle, and modulation depth as joint optimization variables, this invention dynamically outputs the globally optimal combination of control parameters using a particle swarm optimization algorithm, while satisfying the mandatory constraints of basic lighting. This achieves adaptive and collaborative optimization of vehicle lighting, blue light safety, and visible light communication performance. Compared with traditional static design or linear weighted methods, this method can adjust the beam divergence angle and modulation depth in real time according to the distance to the target ahead and drastic changes in ambient illuminance, effectively balancing blue light biosafety in close-range scenarios and communication bandwidth requirements in long-range scenarios, while ensuring compliance with basic lighting regulations. Therefore, this method significantly improves the adaptive capability and multi-objective comprehensive performance of intelligent vehicle lights in dynamic traffic environments, breaking down the physical performance barriers between lighting, communication, and safety.

[0045] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0046] Figure 1 This is a flowchart illustrating a multi-objective optimization method for vehicle lights based on particle swarm optimization algorithm according to an embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram of the convergence curve of a nighttime urban scene according to an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of the convergence curve of a high-speed daytime scene according to an embodiment of the present invention;

[0049] Figure 4 This is a schematic diagram of a multi-objective optimization device for vehicle lights based on particle swarm optimization algorithm according to an embodiment of the present invention;

[0050] Figure 5 This is a schematic diagram of an electronic device structure provided in an embodiment of this application. Detailed Implementation

[0051] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] It should be noted that references to "an embodiment," "embodiment," "example embodiment," etc., in this specification refer to the described embodiment including specific features, structures, or characteristics; however, not every embodiment must include these specific features, structures, or characteristics. Furthermore, such expressions do not refer to the same embodiment. Moreover, when describing specific features, structures, or characteristics in conjunction with embodiments, whether or not explicitly described, it is indicated that incorporating such features, structures, or characteristics into other embodiments is within the knowledge of those skilled in the art.

[0053] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0054] In some embodiments, such as Figure 1 As shown, a multi-objective optimization method for vehicle lights based on particle swarm optimization algorithm is provided. The specific method includes:

[0055] S101, detects the communication distance of the target object in front of the vehicle and the ambient light intensity of the current environment in real time, and obtains the minimum outgoing light flux threshold under the current vehicle driving mode.

[0056] Specifically, the system needs to acquire key information about the current driving environment in real time. An environmental perception module installed at the front of the vehicle (including millimeter-wave radar or lidar, and an ambient light sensor) continuously detects the communication distance to targets ahead of the vehicle. This communication distance refers to the straight-line distance between the vehicle and the vehicle or pedestrian ahead, ranging from approximately 2 meters in urban areas to 200 meters on highways. Simultaneously, the ambient light sensor measures the ambient illuminance, i.e., the light intensity of the surrounding traffic environment, which can be as low as 10 lux at night and as high as 15,000 lux under strong sunlight during the day. Furthermore, the onboard central control module sets a basic lighting constraint based on the current vehicle driving mode (day or night), namely a minimum emitted luminous flux threshold. This threshold represents the minimum luminous flux that the headlights must output to meet safety lighting requirements; for example, a strict setting of 800 lumens for nighttime mode and a relaxed setting of 400 lumens for daytime driving mode. This real-time detection and threshold acquisition form the input basis for adaptive optimization. By sensing the distance to the target ahead and the ambient light level in real time, and defining the baseline for basic lighting, this method provides an accurate physical basis for subsequent dynamic optimization, avoiding insufficient lighting or safety risks caused by sudden changes in the environment of traditional vehicle lights.

[0057] S102 uses communication distance and ambient illuminance as environmental variables, and laser power, beam divergence angle and modulation depth as optimization variables. Under the mandatory constraint of meeting the minimum emitted light flux threshold, it uses particle swarm optimization algorithm to iteratively calculate the comprehensive evaluation function and output the globally optimal combination of control parameters.

[0058] The globally optimal combination of control parameters includes the optimal laser power, the optimal beam divergence angle, and the optimal modulation depth.

[0059] Specifically, the laser power here refers to the power of the output light wave of the blue laser diode, measured in watts. It directly affects the total brightness and blue light component of the headlights. The beam divergence angle refers to the angle at which the emitted white light illumination beam gradually spreads during propagation, measured in degrees. It determines the coverage area of ​​the light spot at the target distance. The modulation depth refers to the degree of amplitude variation in the high-frequency drive signal used to carry communication codes, with a value between 0 and 1. It directly affects the energy proportion of the communication signal. The parameter optimization module has a built-in multi-objective evaluation algorithm based on particle swarm optimization: First, each particle in the particle swarm (i.e., a set of candidate parameter combinations) is substituted into a preset physical model to calculate the blue light hazard assessment value, communication signal-to-noise ratio, and total emitted light flux. Then, the safety weight and communication bandwidth weight are dynamically adjusted according to the communication distance (for example, the safety weight increases significantly at close range, while the communication weight dominates at long range), and a comprehensive evaluation function including a blue light safety penalty term and a signal-to-noise ratio penalty term is constructed. Finally, the individual optimal and global optimal positions of the particles are updated through multiple iterations until the optimal laser power, optimal beam divergence angle, and optimal modulation depth are obtained. For example, in a nighttime urban close-range (5 meters, ambient illuminance 500 lux) scenario, the system outputs an optimal laser power of 5.59 watts, an optimal divergence angle of 34.05 degrees, and an optimal modulation depth of 0.668. However, in a daytime high-speed long-range (100 meters, ambient illuminance 8000 lux) scenario, the output becomes 13.65 watts, 8.6 degrees, and 0.889. Figure 2 This demonstrates how the various modules of the intelligent vehicle lighting system work together to complete this optimization process. Figure 3 This provides the complete execution cycle of the adaptive optimization. By using laser power, divergence angle, and modulation depth as joint optimization variables, and employing the particle swarm optimization algorithm to search for the global optimum in a nonlinear multi-constraint space, this method can coordinate the physical conflicts between blue light safety, communication bandwidth, and basic lighting in real time, overcoming the shortcomings of traditional linear weighted methods that are prone to getting trapped in local optima or failing to adapt to environmental changes.

[0060] Optionally, using communication distance and ambient illuminance as environmental variables, and laser power, beam divergence angle, and modulation depth as optimization variables, under the mandatory constraint of meeting the minimum emitted light flux threshold, a particle swarm optimization algorithm is used to iteratively calculate the comprehensive evaluation function and output the globally optimal combination of control parameters. This includes: for each particle in the particle swarm, substituting its parameter combination into a preset physical model, and combining the real-time detected communication distance and ambient illuminance to calculate the blue light hazard assessment value, communication signal-to-noise ratio, and total emitted light flux; constructing a penalty function based on the blue light hazard assessment value and communication signal-to-noise ratio, and dynamically calculating the safety weight and communication bandwidth weight based on the current communication distance; constructing a fitness objective function based on the safety weight, communication bandwidth weight, and penalty function, and using the fitness objective function as the comprehensive evaluation function for particle swarm iterative optimization to output the optimal laser power, optimal beam divergence angle, and optimal modulation depth.

[0061] The calculation of blue light hazard assessment value includes: extracting the residual high-frequency blue light power transmitted through the phosphor layer; calculating the irradiated area of ​​the blue light beam at the target distance based on the beam divergence angle and communication distance; converting the blue light radiation power density at different test distances to a unified reference distance using a squared scaling factor; and combining this with the environmental limit for the no-hazard exemption level to obtain the blue light hazard assessment value.

[0062] The signal-to-noise ratio (SNR) of communication is calculated as follows: based on the effective blue light signal power at the communication distance, the background shot noise caused by ambient light intensity, and the thermal noise, the SNR is calculated.

[0063] Security weights and communication bandwidth weights are dynamically calculated based on the current communication distance, including:

[0064] , ;

[0065] Where L is the current communication distance. The preset critical handover distance threshold is β, where β is the steepness coefficient; when the communication distance is less than At that time, safety weight Dominant; when the communication distance is greater than At that time, communication bandwidth weight Dominant.

[0066] The penalty function includes a blue light safety penalty term and a communication signal-to-noise ratio (SNR) penalty term. A fitness objective function is constructed based on safety weights, communication bandwidth weights, and the penalty function, including: a preset SNR threshold for the forward error correction communication baseline; a first penalty term constructed based on the blue light hazard assessment value; and a second penalty term constructed based on the communication SNR and the SNR threshold. The fitness objective function is expressed as:

[0067] ;

[0068] in, For safety utility function, For bandwidth utility function, This is a combined penalty term consisting of the first penalty term and the second penalty term. This is a penalty term for luminous flux constraints.

[0069] For example, within the parameter optimization module, before the particle swarm optimization algorithm begins iteration, a physical model evaluation is performed on each particle (i.e., each candidate combination of laser power, beam divergence angle, and modulation depth). Specifically, the parameter optimization module uses the real-time detected communication distance and ambient illuminance as fixed environmental variables, and substitutes the laser power, beam divergence angle, and modulation depth carried by each particle into a preset multidimensional physical model to calculate three key indicators: blue light hazard assessment value, communication signal-to-noise ratio, and total emitted light flux. The blue light hazard assessment value quantifies the safety of blue light radiation at the target distance to the human eye; the communication signal-to-noise ratio reflects the power ratio of the effective signal at the receiving end to various noises, directly determining the reliability of the communication link; and the total emitted light flux is used to verify whether the basic lighting constraints are met. By mapping each set of candidate parameters to physically measurable safety, communication, and lighting indicators, the particle swarm optimization algorithm obtains a quantitative basis for evaluating the merits of the parameters, thus enabling it to search for the optimal solution in a directional manner in subsequent iterations, avoiding blind random searches.

[0070] For the specific calculation of the blue light hazard assessment value, the system first extracts the residual high-frequency blue light power transmitted through the phosphor layer. It's important to note that the phosphor layer absorbs and converts the blue laser beam; some blue light is absorbed by the phosphor and excited to produce yellow light, while the unabsorbed blue light (typically with nanosecond-level response speeds, suitable for high-speed communication) is directly transmitted. The power of this residual blue light is determined by the incident blue light power, the phosphor thickness, and the doping concentration. Then, based on the beam divergence angle and communication distance, the system calculates the irradiated area of ​​the blue light beam at the target distance. Next, to standardize the safety assessment criteria across different communication distances, the system converts the actual measured blue light radiation power density to a benchmark test distance (e.g., the reference distance specified in the IEC 62471 standard) using a squared scaling factor. The conversion is based on the inverse proportionality between optical power density and the square of the distance. Finally, combining this with the no-hazard exemption level (RG0) environmental limit specified in the IEC 62471 standard—that is, the maximum permissible blue light radiation power per unit area—the system calculates the blue light hazard assessment value. :

[0071] ;

[0072] In the formula, The blue light transmittance of the fluorescent layer; The environmental limits corresponding to RG0 (no hazard exemption level) as specified in IEC 62471 standard shall be adopted. When the distance is close and there are pedestrians (such as...) When the area decreases sharply, the system needs to increase the divergence angle. Or reduce power Only then can it be maintained Safety boundaries.

[0073] For example, in a close-range (5-meter) scenario in a nighttime urban area, the system automatically increases the divergence angle or decreases the laser power to keep the evaluated value within a safe limit of no less than 0. For the communication signal-to-noise ratio (SNR), the system calculates it based on the effective blue light signal power at the communication distance, background shot noise caused by ambient light intensity, and the inherent thermal noise of the receiver. The effective blue light signal power depends on the residual blue light power, modulation depth, the ratio of the receiver lens area to the irradiated surface area, and the responsivity of the photodetector; background shot noise is proportional to ambient light intensity, with higher noise levels in brighter environments; thermal noise is a constant inherent to the receiver circuitry. Dividing the signal power by the total noise power, taking the logarithm to base 10, and then multiplying by 10 yields the communication SNR in decibels.

[0074] ;

[0075] In the formula, For the lens area, Photoelectric responsivity, Thermal noise, Indicates illuminance Drastically varying background shot noise sets the signal-to-noise ratio threshold for the system's forward error correction (FEC) communication baseline. Normalized communication bandwidth Based on Shannon's theorem, the total emitted luminous flux of the vehicle lighting system was calculated. .

[0076] By introducing a distance squared ratio conversion, the consistency and comparability of blue light safety assessments at different distances are ensured. At the same time, the impact of ambient light intensity on shot noise is incorporated into the signal-to-noise ratio calculation, enabling the model to accurately reflect changes in communication performance under strong sunlight during the day or dark conditions at night, providing accurate physical input for subsequent dynamic weight adjustments.

[0077] After obtaining the blue light hazard assessment value and communication signal-to-noise ratio for each particle, the system needs to dynamically calculate the safety weight and communication bandwidth weight based on the current communication distance. Here, the safety weight reflects the priority of protecting human eyes from blue light in the current scenario, while the communication bandwidth weight reflects the priority of ensuring high-speed data transmission. The two are smoothly transitioned using a sigmoid function: Safety Weight Communication bandwidth weight Where L is the current real-time detected communication distance. This is the system's preset critical switching distance threshold, which can be set to, for example, 20 meters or 30 meters. β is the steepness coefficient, used to control the rate of change of the weights near the threshold. When the communication distance is less than... At that time, the exponent term The weights are much less than 1, with the security weight approaching 1 and the communication bandwidth weight approaching 0, indicating that the system should prioritize blue light safety and strictly limit residual blue light output; when the communication distance is greater than 1... At this point, the exponential term is much greater than 1, the security weight approaches 0, and the communication bandwidth weight approaches 1, indicating that the natural attenuation of blue light is sufficient, and the system can prioritize high communication bandwidth. Nearby, the weights change continuously, enabling a smooth switch from a safety-dominated mode to a communication-dominated mode. For example, at a close distance of 5 meters, the safety weight is close to 1; at a long distance of 100 meters, the communication weight is close to 1. By dynamically adjusting the weights using the Sigmoid function, the parameter abrupt changes and instabilities caused by hard threshold switching in traditional methods are avoided. This allows the vehicle lighting system to continuously and smoothly change its optimization focus according to the actual distance, ensuring absolute safety at close range without sacrificing long-range communication performance.

[0078] Next, the system constructs a penalty function to handle extreme cases that violate security or communication thresholds. The penalty function includes a blue light safety penalty term and a communication signal-to-noise ratio (SNR) penalty term. First, the system presets a SNR threshold for forward error correction (FEC) communication, typically set at 10 dB. Below this threshold, the bit error rate rises sharply, making reliable communication impossible. Then, a first penalty term is constructed based on the blue light hazard assessment value: if the assessment value is below the safety limit (e.g., less than 0), a large positive penalty is applied, forcing the particle swarm optimization algorithm away from these dangerous parameter combinations; if the assessment value is within the safety range, the penalty term approaches 0. Similarly, a second penalty term is constructed based on the difference between the communication SNR and the preset threshold: when the SNR is below the threshold, the penalty term grows exponentially; when the SNR is above the threshold, the penalty term rapidly decays to 0. In practical implementations, an exponential penalty function can be used, for example... ,in , k1 and k2 are positive coefficients and are large constants. In addition, a luminous flux constraint penalty term is introduced. This is used to ensure that the total emitted light flux is not lower than the minimum emitted light flux threshold; if it is lower than the threshold, a penalty is applied. The final fitness objective function is: .in This is a safety utility function, and its value increases as the blue light hazard assessment value increases; Let be the bandwidth utility function, whose value increases with the communication signal-to-noise ratio. The negative sign indicates that the algorithm needs to maximize the weighted sum of security and communication utility while minimizing the penalty term. Through the double exponential penalty term, the hard constraints of security and communication are transformed into continuously differentiable soft barriers, guiding the particle swarm optimization algorithm to find the optimal compromise solution while satisfying the bottom lines of security and communication. Simultaneously, the negative sign structure means that a smaller fitness function value indicates better overall performance, facilitating the particle swarm optimization algorithm to perform a minimization search and thus efficiently converge to the globally optimal combination of control parameters. Figure 2 and Figure 3 Convergence curves for nighttime urban scenarios and daytime highway scenarios are presented respectively, verifying the effectiveness of the fitness function.

[0079] Optionally, the iterative optimization process of the particle swarm optimization algorithm includes: initializing the particle swarm and setting a three-dimensional optimization variable set; the three-dimensional optimization variable set includes laser power, beam divergence angle, and modulation depth; calculating the fitness value of each particle in the current particle swarm, updating the individual's historical best position and the global best position; driving the particle swarm to move in the multi-dimensional solution space, repeating the iteration until the maximum number of iterations or the convergence condition is met; extracting the global optimal solution combination, outputting the optimal laser power and optimal modulation depth to the communication modulation module, and outputting the optimal beam divergence angle to the projection optics module.

[0080] Specifically, the algorithm randomly generates a certain number of particles to form an initial particle swarm based on a pre-defined search space. Each particle represents a set of system hardware control strategies to be optimized, containing three dimensions of variables: the laser power P of the blue laser diode, which is typically set between the minimum power P_min (e.g., 1 watt) and the maximum power P_max (e.g., 15 watts); the divergence angle α of the emitted beam, which ranges from the minimum divergence angle α_min (e.g., 5 degrees) to the maximum divergence angle α_max (e.g., 40 degrees); and the communication modulation depth m, which ranges from 0 to 1. The number of particles is an important parameter of the algorithm, and in this embodiment, it is set to 60 particles. In addition to carrying the current position coordinates (i.e., a specific set of P, α, and m values), each particle also maintains two historical optimal positions: the individual historical optimal position, which is the best combination of control parameters with the best fitness found by the particle so far; and the global optimal position, which is the best among all the individual historical optimal positions in the entire particle swarm. By reasonably setting the boundary range of the three-dimensional optimization variables and the number of particles, we can ensure sufficient coverage of the search space and control the computational complexity, thus laying a good foundation for subsequent efficient iterations.

[0081] After initialization, the system enters an iterative loop. In each iteration, the parameter optimization module first calculates the fitness value of each particle in the current particle swarm. The fitness value here refers to the numerical value of the fitness objective function F(x), which comprehensively reflects the overall performance of the laser power, divergence angle, and modulation depth represented by the particle in three dimensions: security, communication, and illumination. The smaller the value, the better the overall performance. After calculating the fitness value of each particle, the algorithm compares it with the fitness value corresponding to the particle's historical best position: if the current particle's fitness value is smaller, the historical best position is replaced with the current position. Subsequently, the algorithm collects the updated historical best positions of all particles and selects the one with the smallest fitness value as the global best position for the current iteration step. Through particle-by-particle fitness evaluation and a real-time update mechanism of dual optima (individual best and global best), the particle swarm optimization algorithm can continuously record the best solutions encountered during the search process, ensuring that excellent parameter combinations are not lost due to subsequent random movements, thus guaranteeing the convergence of the algorithm.

[0082] Next, the algorithm drives the particle swarm to move in the multidimensional solution space, converging towards a better solution region. The motion of each particle is influenced by three directions: inertial direction (the tendency for the particle to maintain its previous motion trend), individual cognitive direction (the tendency for the particle to be attracted to its own historical best position), and social cognitive direction (the tendency for the particle to be attracted to the global best position of the entire swarm). Specifically, the velocity vector of each particle is updated based on the weighted sum of these three directions, and then the particle's position vector (i.e., the current laser power, divergence angle, and modulation depth) is obtained by adding the updated velocity to the original position. By adjusting the inertial weight, individual learning factor, and social learning factor, the balance between global exploration and local development of the particle swarm can be controlled. This series of operations—fitness calculation, individual and global best updates, and velocity and position updates—is repeated until a preset termination condition is met. In this embodiment, the termination condition is reaching the maximum number of iterations, 120. Figure 2 The convergence curve of the nighttime urban scene shown indicates that the fitness value drops rapidly in the initial stage (first 5 iterations), and the algorithm quickly locks in a better search region; a plateau occurs from generation 7 to generation 16, during which the particle swarm is conducting a local search and trying to escape the local optimum; the curve flattens out completely after about generation 23, indicating that the algorithm has found the global optimum. Figure 3The convergence curve for the high-speed daytime scene shown exhibits a step-like descent, briefly plateauing between generations 8 and 13 before breaking through again in generation 14, further demonstrating the algorithm's ability to escape local optima and obtain globally optimal control strategies. Through the particle swarm optimization's velocity-position update mechanism, the algorithm can efficiently search in high-dimensional nonlinear, multi-constraint solution spaces, avoiding the pitfalls of exhaustive search or gradient descent methods that easily get trapped in local optima. Furthermore, compared to evolutionary algorithms such as genetic algorithms, the particle swarm optimization algorithm has a faster convergence speed and requires less parameter tuning.

[0083] After the iteration terminates, the system extracts the global optimal solution combination. Specifically, the parameter optimization module records the global optimal position obtained from the last iteration update, which includes specific values ​​in three dimensions: optimal laser power P, optimal beam divergence angle α, and optimal modulation depth m. Subsequently, the module sends these optimal values ​​to the corresponding execution units: the optimal laser power P and optimal modulation depth m are output to the communication modulation module to generate a high-frequency electrical signal to drive the blue laser diode; the optimal beam divergence angle α is output to the projection optics module to drive a stepper motor or piezoelectric actuator to change the relative position of the lens group. For example, in a nighttime urban close-range scenario, the output optimal laser power is 5.59 watts, the optimal divergence angle is 34.05 degrees, and the optimal modulation depth is 0.668. In a daytime high-speed long-range scenario, the output optimal laser power is 13.65 watts, the optimal divergence angle is 8.6 degrees, and the optimal modulation depth is 0.889. At this point, a complete adaptive optimization cycle is completed, and the vehicle lighting system completes the entire closed-loop control from perception and decision-making to execution based on real-time environmental changes. By precisely decomposing and allocating the globally optimal solution obtained by the algorithm convergence to the communication modulation module and the projection optics module, the optimization results are seamlessly converted into physical execution instructions, ensuring that the theoretically optimal parameters can be accurately implemented in actual hardware, thereby truly breaking down the performance barriers between vehicle lights in terms of lighting, communication and safety.

[0084] S103 generates a high-frequency driving signal based on the optimal modulation depth, driving the blue laser diode to output a blue laser beam carrying communication codes at the optimal laser power.

[0085] Specifically, the communication modulation module receives the optimal modulation depth from the parameter optimization module and the communication data to be transmitted from the vehicle central control module. It uses this modulation depth to amplitude modulate the high-frequency carrier, forming an electrical high-frequency drive signal with a specific duty cycle and level drop. The blue laser diode in the light source module responds to this drive signal, emitting a blue laser beam carrying communication encoding information at optimal laser power (e.g., 5.59 watts or 13.65 watts). This blue laser beam then illuminates a phosphor coating with a preset thickness d and a doping concentration of C. A portion of the blue light is absorbed by the phosphor and excited to emit yellow light, while the unabsorbed residual blue light (with a nanosecond-level response speed suitable for high-frequency communication) is directly transmitted, mixing with the yellow light to form a white light beam for illumination. By utilizing the optimal modulation depth and optimal laser power to drive the light source, this method ensures maximum energy efficiency of the communication signal while precisely controlling the transmission amount of residual blue light using a particle swarm optimization algorithm, thus guaranteeing the reliability of the visible light communication link without sacrificing illumination quality.

[0086] S104: Dynamically adjust the projection optics module according to the optimal beam divergence angle to change the divergence angle of the emitted white light illumination beam.

[0087] Specifically, the projection optics module includes a variable focal length lens group or a collimating lens mechanism driven by a micro-motor. After receiving the optimal divergence angle command from the parameter optimization module, it drives an internal stepper motor or piezoelectric actuator to precisely change the relative position between the lens groups, thereby adjusting the diffusion of the emitted beam. For example, in a close-range nighttime scenario, the optimal divergence angle is 34.05 degrees. At this time, the lens group will move to a position that broadly diffuses the beam, increasing the spot area and reducing the blue light radiation density per unit area, thus protecting the eyes of pedestrians. In a high-speed, long-range daytime scenario, the optimal divergence angle is 8.6 degrees. The lens group is then adjusted to a collimated state, making the beam energy highly concentrated to compensate for the optical power attenuation after long-distance transmission, ensuring that the signal-to-noise ratio of long-distance communication still exceeds the forward error correction threshold of 10 dB. Figure 2 and Figure 3 The convergence curves of the particle swarm optimization algorithm are shown in nighttime urban scenarios and daytime highway scenarios, demonstrating that the algorithm can quickly and stably find the optimal divergence angle. By dynamically adjusting the beam divergence angle, this method achieves adaptive control of the spatial distribution of light energy—actively diffusing at close range to reduce blue light hazards and narrowing at long range to increase communication distance. This directly supports the synergistic optimization of safety and communication performance at the physical optics level, further enhancing the environmental adaptability of the vehicle lighting system.

[0088] Based on the same inventive concept, this application also provides a particle swarm optimization device for implementing the aforementioned multi-objective optimization method for vehicle lights based on particle swarm optimization. The solution provided by this device is similar to the implementation described in the above method. Therefore, the specific limitations of the one or more embodiments of the multi-objective optimization device for vehicle lights based on particle swarm optimization provided below can be found in the limitations of the multi-objective optimization method for vehicle lights based on particle swarm optimization above, and will not be repeated here.

[0089] In one embodiment, such as Figure 4 As shown, a multi-objective optimization device for vehicle lights based on particle swarm optimization algorithm is provided. The device includes:

[0090] The information detection module 30 is used to detect the communication distance of the target object in front of the vehicle and the ambient light intensity of the current environment in real time, and to obtain the minimum outgoing light flux threshold in the current vehicle driving mode.

[0091] The parameter solving module 31 is used to iteratively calculate the comprehensive evaluation function using the communication distance and ambient illuminance as environmental variables, and the laser power, beam divergence angle and modulation depth as optimization variables, under the mandatory constraint of satisfying the minimum emitted light flux threshold, and output the globally optimal combination of control parameters; the globally optimal combination of control parameters includes the optimal laser power, the optimal beam divergence angle and the optimal modulation depth.

[0092] The beam output module 32 is used to generate a high-frequency driving signal according to the optimal modulation depth, and drive the blue laser diode to output a blue laser beam carrying communication code with optimal laser power.

[0093] Angle adjustment module 33 is used to dynamically adjust the projection optics module according to the optimal beam divergence angle, thereby changing the divergence angle of the emitted white light illumination beam.

[0094] This application also provides an electronic device, in some embodiments, referring to... Figure 5 As shown, the electronic device 700 includes an input unit 710, a memory 720, a processor 730, and an output unit 740. The memory 720 stores program instructions that can be executed on the processor 730. The processor 730 can execute the multi-objective optimization method and / or technical solution for vehicle lights based on the particle swarm optimization algorithm in the foregoing embodiments by calling the program instructions. The electronic device 700 can be a mobile terminal device such as a mobile phone or a computer.

[0095] Furthermore, embodiments of this application also provide a computer-readable storage medium for storing a computer program that executes a multi-objective optimization method for vehicle lights based on a particle swarm optimization algorithm. For example, computer program instructions, when executed by a computer, can invoke or provide the methods and / or technical solutions according to this application through the operation of the computer. The program instructions that invoke the methods of this application may be stored in a fixed or removable storage medium, and / or transmitted via data streams in broadcast or other signal carrying media, and / or stored in a storage medium that operates according to the program instructions.

[0096] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0097] The technical features of the above embodiments can be arbitrarily integrated. For the sake of brevity, not all possible integrations of the technical features in the above embodiments are described. However, as long as the integration of these technical features does not contradict each other, they should be considered to be within the scope of this specification.

[0098] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A multi-objective optimization method for vehicle headlights based on particle swarm optimization, characterized in that, Applied to intelligent vehicle lighting systems, the method includes: Real-time detection of the communication distance to the target object in front of the vehicle and the ambient light level of the current environment, and acquisition of the minimum outgoing light flux threshold in the current vehicle driving mode; Using the communication distance and ambient illuminance as environmental variables, and laser power, beam divergence angle, and modulation depth as optimization variables, under the mandatory constraint of satisfying the minimum emitted light flux threshold, a particle swarm optimization algorithm is used to iteratively calculate the comprehensive evaluation function and output the globally optimal combination of control parameters; the globally optimal combination of control parameters includes the optimal laser power, the optimal beam divergence angle, and the optimal modulation depth. Using the communication distance and ambient illuminance as environmental variables, and laser power, beam divergence angle, and modulation depth as optimization variables, under the mandatory constraint of satisfying the minimum emitted light flux threshold, a particle swarm optimization algorithm is used to iteratively calculate the comprehensive evaluation function and output the globally optimal combination of control parameters, including: For each particle in the particle swarm, its parameter combination is substituted into a preset physical model. Combined with the real-time detected communication distance and ambient illuminance, the blue light hazard assessment value, communication signal-to-noise ratio, and total emitted light flux are calculated respectively. A penalty function is constructed based on the blue light hazard assessment value and the communication signal-to-noise ratio, and the safety weight and communication bandwidth weight are dynamically calculated based on the current communication distance. A fitness objective function is constructed based on the safety weight, the communication bandwidth weight, and the penalty function. The fitness objective function is used as a comprehensive evaluation function to perform iterative optimization of the particle swarm, and the optimal laser power, optimal beam divergence angle, and optimal modulation depth are output. The penalty function includes a blue light safety penalty term and a communication signal-to-noise ratio (SNR) penalty term. A fitness objective function is constructed based on the safety weight, the communication bandwidth weight, and the penalty function, including: setting a preset SNR threshold for the forward error correction communication baseline; constructing a first penalty term based on the blue light hazard assessment value; and constructing a second penalty term based on the communication SNR and the SNR threshold. The fitness objective function is expressed as: ； in, For safety utility function, For bandwidth utility function, This is a combined penalty term consisting of the first penalty term and the second penalty term. This is a penalty term for luminous flux constraints; A high-frequency driving signal is generated based on the optimal modulation depth to drive the blue laser diode to output a blue laser beam carrying communication codes at the optimal laser power. The projection optics module is dynamically adjusted according to the optimal beam divergence angle to change the divergence angle of the emitted white light illumination beam.

2. The multi-objective optimization method for vehicle lights based on particle swarm optimization as described in claim 1, characterized in that, Calculating the blue light hazard assessment value includes: Extract the residual high-frequency blue light power transmitted through the phosphor layer, and calculate the irradiated area of ​​the blue light beam at the target distance based on the beam divergence angle and the communication distance. The blue light radiation power density at different test distances is converted to a unified reference distance using a squared scaling factor, and combined with the environmental limit for the no-hazard exemption level, the blue light hazard assessment value is calculated. The communication signal-to-noise ratio is calculated in the following way: The communication signal-to-noise ratio is calculated based on the effective blue light signal power at the communication distance, the background shot noise caused by the ambient light intensity, and the thermal noise.

3. The multi-objective optimization method for vehicle lights based on particle swarm optimization as described in claim 1, characterized in that, Security weights and communication bandwidth weights are dynamically calculated based on the current communication distance, including: ， ； Where L is the current communication distance. The preset critical switching distance threshold is β, where β is the steepness coefficient; when the communication distance is less than At that time, the security weight Dominant; when the communication distance is greater than At that time, the communication bandwidth weight Dominant.

4. The multi-objective optimization method for vehicle lights based on particle swarm optimization as described in claim 1, characterized in that, The iterative optimization process of the particle swarm optimization algorithm includes: Initialize the particle swarm and set a three-dimensional optimization variable set; the three-dimensional optimization variable set includes laser power, beam divergence angle and modulation depth. Calculate the fitness value of each particle in the current particle swarm, and update the individual's historical best position and global best position; The particle swarm is driven to move in a multidimensional solution space, and the iteration is repeated until the maximum number of iterations or the convergence condition is met. Extract the global optimal solution combination, output the optimal laser power and the optimal modulation depth to the communication modulation module, and output the optimal beam divergence angle to the projection optics module.

5. A multi-objective optimization device for vehicle lights based on particle swarm optimization algorithm, characterized in that, The device includes: The information detection module is used to detect the communication distance of the target object in front of the vehicle and the ambient light intensity of the current environment in real time, and to obtain the minimum outgoing light flux threshold under the current vehicle driving mode. The parameter solving module is used to iteratively calculate a comprehensive evaluation function using a particle swarm optimization algorithm, with the communication distance and ambient illuminance as environmental variables, and laser power, beam divergence angle, and modulation depth as optimization variables, under the mandatory constraint of satisfying the minimum emitted light flux threshold. The module outputs a globally optimal combination of control parameters, including optimal laser power, optimal beam divergence angle, and optimal modulation depth. For each particle in the particle swarm, its parameter combination is substituted into a preset physical model. Combined with the real-time detected communication distance and ambient illuminance, the blue light hazard assessment value, communication signal-to-noise ratio, and total emitted light flux are calculated respectively. A penalty function is constructed based on the blue light hazard assessment value and the communication signal-to-noise ratio, and the safety weight and communication bandwidth weight are dynamically calculated based on the current communication distance. A fitness objective function is constructed based on the safety weight, the communication bandwidth weight, and the penalty function. The fitness objective function is used as a comprehensive evaluation function to perform iterative optimization of the particle swarm, and the optimal laser power, optimal beam divergence angle, and optimal modulation depth are output. The penalty function includes a blue light safety penalty term and a communication signal-to-noise ratio (SNR) penalty term. A fitness objective function is constructed based on the safety weight, the communication bandwidth weight, and the penalty function, including: setting a preset SNR threshold for the forward error correction communication baseline; constructing a first penalty term based on the blue light hazard assessment value; and constructing a second penalty term based on the communication SNR and the SNR threshold. The fitness objective function is expressed as: ； in, For safety utility function, For bandwidth utility function, This is a combined penalty term consisting of the first penalty term and the second penalty term. This is a penalty term for luminous flux constraints; The beam output module is used to generate a high-frequency driving signal based on the optimal modulation depth, and drive the blue laser diode to output a blue laser beam carrying communication codes with optimal laser power. An angle adjustment module is used to dynamically adjust the projection optics module according to the optimal beam divergence angle, thereby changing the divergence angle of the emitted white light illumination beam.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the multi-objective optimization method for vehicle lights based on the particle swarm optimization algorithm as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the multi-objective optimization method for vehicle lights based on the particle swarm optimization algorithm as described in any one of claims 1 to 4.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the multi-objective optimization method for vehicle lights based on the particle swarm optimization algorithm as described in any one of claims 1 to 4.

Images