A laser radar cooling system

Abstract

The present invention discloses a laser radar heat dissipation system in the field of radar heat dissipation technology, including: a dust acquisition module determines whether to adjust the initial fan speed of the radiator according to dust influence parameters; a first adjustment module compares the dust influence parameters with historical data, determines the speed contrast according to the comparison result, and when the speed contrast is greater than or equal to the speed contrast threshold, determines the speed adjustment factor according to the historical data; when the speed contrast is less than the speed ratio threshold, determines the speed adjustment factor through the speed prediction model; a temperature processing module determines the interference mark of each real-time temperature value on the temperature interference data chain based on the standard heat dissipation value of the radiator; a second adjustment module determines the heat dissipation output value of the laser radar surface based on the heat dissipation type and heat dissipation area, and determines the target fan speed of the radiator according to the adjustment coefficient. The present invention effectively adjusts the initial fan speed of the radiator by processing dust and ambient temperature.

Description

Technical Field

[0001] The present invention relates to the field of radar heat dissipation technology, and in particular to a laser radar heat dissipation system. Background Art

[0002] As a high-precision sensor, LiDAR is widely used in autonomous driving, robot navigation, environmental monitoring and other fields. Its working principle is based on the emission and reflection of laser beams. It receives the reflected laser beams to build a high-precision three-dimensional image of the surrounding environment. However, during the long-term operation of LiDAR, a certain amount of heat will be generated during the laser emission and reception process. If the heat cannot be effectively dissipated, the heat will cause the temperature of the LiDAR surface to rise, which will affect the performance and accuracy of the LiDAR and even cause failure.

[0003] At present, the heat dissipation of LiDAR relies on the speed of the fan in the radiator to drive the air flow to reduce the temperature of the LiDAR surface. However, traditional heat dissipation has certain limitations. LiDARs are mostly in outdoor environments. During long-term use, dust, dirt and other impurities are easily accumulated on the surface of the LiDAR. These impurities will affect the fan's heat dissipation efficiency for the LiDAR, resulting in a decrease in heat dissipation performance. Moreover, in a high temperature environment, the increase in ambient temperature will also affect the fan's heat dissipation efficiency for the LiDAR, thereby exacerbating the problem of poor heat dissipation of the LiDAR. In response to the influence of impurities such as dust and high temperature environment, the existing heat dissipation cannot dynamically adjust the speed of the fan in the radiator, resulting in insufficient heat dissipation of the LiDAR surface, thereby affecting the stability and accuracy of the LiDAR.

[0004] Therefore, how to provide a laser radar heat dissipation system is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0005] In view of this, the present invention proposes a laser radar heat dissipation system, which aims to solve the problem of how to deal with the influence of impurities such as dust and high temperature environment, and the inability to dynamically adjust the speed of the fan in the radiator, resulting in insufficient heat dissipation on the laser radar surface.

[0006] The present invention provides a laser radar heat dissipation system, comprising:

[0007] a dust acquisition module configured to collect image data of the laser radar surface, analyze the image data, obtain dust impact parameters of the laser radar surface, and determine whether to adjust the initial fan speed of the radiator according to the dust impact parameters;

[0008] A first adjustment module is configured to compare the dust impact parameter with the historical data when determining to adjust the initial fan speed of the radiator, determine a speed contrast according to the comparison result, determine a speed adjustment factor according to the historical data when the speed contrast is greater than or equal to a speed contrast threshold, determine a speed adjustment factor through a speed prediction model when the speed contrast is less than a speed ratio threshold, and determine the secondary fan speed according to the speed adjustment factor;

[0009] a temperature processing module configured to determine a heat dissipation area on the surface of the laser radar, establish a plurality of temperature collection points in the heat dissipation area, obtain a real-time temperature value of each temperature collection point, generate a temperature interference data link according to all the real-time temperature values, determine an interference mark of each real-time temperature value on the temperature interference data link based on a standard heat dissipation value of the radiator, adjust the temperature interference data link according to the interference mark, and determine a heat dissipation temperature data link;

[0010] The second adjustment module is configured to parse the real-time temperature value on the heat dissipation temperature data link, determine the heat dissipation type of the laser radar surface according to the analysis result, the heat dissipation type includes local heat dissipation and overall heat dissipation, determine the heat dissipation output value of the laser radar surface based on the heat dissipation type and the heat dissipation area, determine the adjustment coefficient of the secondary fan speed according to the heat dissipation output value, and determine the target fan speed of the radiator according to the adjustment coefficient.

[0011] Further, when determining whether to adjust the initial fan speed of the radiator according to the dust impact parameter, it includes:

[0012] The dust acquisition module compares the dust impact parameter with a preset standard dust impact parameter, and determines whether to adjust the initial fan speed of the radiator according to the comparison result;

[0013] When the dust impact parameter is greater than the standard dust impact parameter, determining to adjust the initial fan speed of the radiator;

[0014] When the dust impact parameter is less than or equal to the standard dust impact parameter, it is determined that the initial fan speed of the radiator is not adjusted, and the initial fan speed of the radiator is determined as the secondary fan speed.

[0015] Further, when determining to adjust the initial fan speed of the radiator, a comparison is performed based on the dust impact parameter and the historical data, and a speed contrast is determined based on the comparison result, including:

[0016] The first adjustment module acquires the historical data, wherein the historical data includes a historical dust impact parameter, a historical speed adjustment factor, and a historical initial fan speed, and the historical dust impact parameter, the historical speed adjustment factor, and the historical initial fan speed correspond to each other;

[0017] When the first adjustment module determines the rotation speed contrast according to the comparison result, the rotation speed contrast is obtained by the following formula:

[0018] ;

[0019] in, Indicates the speed contrast, Indicates dust impact parameters, represents the historical dust impact parameter, Indicates the initial fan speed. Indicates the initial fan speed in history. and represents the weight coefficient, and .

[0020] Further, when the speed contrast is greater than or equal to the speed contrast threshold, determining the speed adjustment factor according to the historical data includes:

[0021] The first adjustment module compares the rotation speed contrast with a preset rotation speed contrast threshold, and determines a rotation speed adjustment factor according to the comparison result;

[0022] When there is data in the historical data whose rotation speed contrast is greater than or equal to the rotation speed contrast threshold, determining the rotation speed adjustment factor according to the historical data;

[0023] When there is only one data in the historical data whose rotation speed contrast is greater than or equal to the rotation speed contrast threshold, the first adjustment module uses the historical rotation speed adjustment factor corresponding to the data as the rotation speed adjustment factor;

[0024] When there is not only one data in the historical data whose rotation speed contrast is greater than the rotation speed contrast threshold, the first adjustment module uses the historical rotation speed adjustment factor corresponding to the rotation speed contrast with the largest rotation speed contrast determined by the data as the rotation speed adjustment factor;

[0025] When there is not only one piece of data in the historical data whose rotation speed contrast is equal to the rotation speed contrast threshold, the first adjustment module uses the average of the historical rotation speed adjustment factors corresponding to each piece of data as the rotation speed adjustment factor.

[0026] Furthermore, when the speed adjustment factor is determined by the speed prediction model and the speed of the secondary fan is determined according to the speed adjustment factor, it includes:

[0027] The first adjustment module takes all historical initial fan speeds, all historical speed adjustment factors, and all historical dust impact parameters as a model data set, divides the model data set into a training set and a test set, and selects a random forest model as an initial model;

[0028] The first adjustment module iteratively trains the initial model according to the training set, and verifies the iteratively trained initial model according to the test set to obtain a prediction accuracy rate;

[0029] If the prediction accuracy of the initial model after the current iterative training is less than the prediction accuracy of the initial model after the previous iterative training, the absolute values ​​of the model parameters of the initial model after the current iterative training are summed through L1 regularization, and the squares of the model parameters of the initial model after the current iterative training are summed through L2 regularization. The summed results of L1 regularization and L2 regularization are added as penalty terms to the loss function of the initial model after the current iterative training, and iterative training is continued until the preset number of iterations is reached;

[0030] If the prediction accuracy of the initial model after the current iterative training is greater than or equal to the prediction accuracy of the initial model after the previous iterative training, the iterative training is stopped, the initial model after the current iterative training is determined as the speed prediction model, the dust impact parameter and the initial fan speed of the radiator are input into the speed prediction model, and the speed adjustment factor is determined;

[0031] The secondary fan speed is a product of the speed adjustment factor and the initial fan speed of the radiator.

[0032] Further, when determining the interference mark of each real-time temperature value on the temperature interference data link based on the standard heat dissipation value of the radiator, it includes:

[0033] The temperature processing module determines the temperature difference between each real-time temperature value and the ambient temperature value, and determines the interference mark of each real-time temperature value according to the temperature difference and the standard heat dissipation value of the radiator, wherein the interference mark includes a matching mark and a difference mark;

[0034] When the temperature difference is equal to the standard heat dissipation value, the temperature processing module generates the matching mark for the corresponding real-time temperature value;

[0035] When the temperature difference is not equal to the standard heat dissipation value, the temperature processing module generates the difference mark for the corresponding real-time temperature value.

[0036] Further, when the temperature interference data link is adjusted according to the interference mark to determine the heat dissipation temperature data link, it includes:

[0037] When the matching mark does not exist on the temperature interference data link, the temperature processing module determines the temperature interference data link as the heat dissipation temperature data link;

[0038] When the matching mark exists on the temperature interference data link, the temperature processing module removes the real-time temperature value corresponding to the matching mark to obtain the heat dissipation temperature data link.

[0039] Furthermore, when the real-time temperature value on the heat dissipation temperature data link is parsed and the heat dissipation type of the laser radar surface is determined according to the parsing result, it includes:

[0040] The second regulating module analyzes the real-time temperature values ​​corresponding to the difference marks on the heat dissipation temperature data chain, determines the temperature average of all real-time temperature values, counts the first difference number of the difference marks whose real-time temperature values ​​are greater than the temperature average, and counts the second difference number of the difference marks whose real-time temperature values ​​are less than the temperature average;

[0041] When the first difference number is greater than or equal to the second difference number, it is determined that the heat dissipation type of the laser radar surface is overall heat dissipation;

[0042] When the first difference number is less than the second difference number, it is determined that the heat dissipation type of the laser radar surface is local heat dissipation.

[0043] Further, when determining the heat dissipation output value of the laser radar surface based on the heat dissipation type and the heat dissipation area, it includes:

[0044] The second adjustment module obtains the heat dissipation output value according to the following formula:

[0045] ;

[0046] in, Indicates the heat dissipation output value, Represents the area of ​​the heat dissipation region, represents the density of the lidar material, and Represents the weight coefficient. When the heat dissipation type of the laser radar surface is determined to be overall heat dissipation, The value is [0.5, 1], The value is (0, 0.5). When the heat dissipation type of the laser radar surface is determined to be local heat dissipation, The value is (0, 0.5), The value is [0.5, 1], represents the specific heat capacity of the lidar material, Represents the thermal conductivity of the lidar material.

[0047] Further, when determining the adjustment coefficient of the secondary fan speed according to the heat dissipation output value, and determining the target fan speed of the radiator according to the adjustment coefficient, it includes:

[0048] The second adjustment module pre-sets a first preset heat dissipation output value and a second preset heat dissipation output value, and pre-sets a first preset adjustment coefficient, a second preset adjustment coefficient, and a third preset adjustment coefficient;

[0049] When the heat dissipation output value is greater than the first preset heat dissipation output value, determining the first preset adjustment coefficient as the adjustment coefficient of the secondary fan speed;

[0050] When the heat dissipation output value is less than or equal to the first preset heat dissipation output value and greater than the second preset heat dissipation output value, determining the second preset adjustment coefficient as the adjustment coefficient of the secondary fan speed;

[0051] When the heat dissipation output value is less than or equal to the second preset heat dissipation output value, determining the third preset adjustment coefficient as the adjustment coefficient of the secondary fan speed;

[0052] The target fan speed is a product value of the secondary fan speed and an adjustment coefficient of the secondary fan speed.

[0053] Compared with the prior art, the beneficial effects of the present invention are: according to the dust condition on the surface of the laser radar, the dust impact parameters are obtained by analyzing the image data, and it is determined whether the initial fan speed of the radiator needs to be adjusted, thereby avoiding the accumulation of dust affecting the heat dissipation effect of the fan on the surface of the laser radar, thereby ensuring the stability and reliability of the laser radar in a dusty environment, and comparing the dust impact parameters with historical data to determine the speed contrast, and based on the speed contrast, the speed adjustment factor is determined according to the speed prediction model or historical data, so as to adjust the initial fan speed, thereby improving the flexibility and adjustment accuracy of the system and reducing the risk that the initial fan speed of the radiator cannot meet the heat dissipation requirements of the laser radar surface. Multiple temperature collection points are set in the heat dissipation area, and the temperature changes of each collection point are monitored in real time. A temperature interference data chain is generated, and the interference mark is determined according to the standard heat dissipation value, which effectively identifies the temperature fluctuations in the high-temperature environment, ensures the accuracy of the temperature data, and improves the system's adaptability to high-temperature environments. In addition, by analyzing the heat dissipation temperature data chain, the heat dissipation type of the lidar surface is judged, and the heat dissipation output value is calculated according to different heat dissipation types and heat dissipation areas, ensuring that the system can determine the adjustment coefficient of the secondary fan speed according to different situations, thereby improving the heat dissipation efficiency of the lidar surface and the system's adaptability, effectively avoiding performance degradation or equipment damage of the lidar due to overheating, and enabling the lidar to maintain a good working condition in a high-temperature environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0055] Figure 1 A functional block diagram of a laser radar heat dissipation system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0056] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0057] In some embodiments of the present application, see Figure 1As shown, a laser radar heat dissipation system includes:

[0058] A dust acquisition module is configured to collect image data of the laser radar surface, analyze the image data, obtain dust impact parameters of the laser radar surface, and determine whether to adjust the initial fan speed of the radiator according to the dust impact parameters;

[0059] The first adjustment module is configured to compare the dust impact parameter with the historical data when determining to adjust the initial fan speed of the radiator, determine the speed contrast according to the comparison result, determine the speed adjustment factor according to the historical data when the speed contrast is greater than or equal to the speed contrast threshold, determine the speed adjustment factor through the speed prediction model when the speed contrast is less than the speed ratio threshold, and determine the secondary fan speed according to the speed adjustment factor;

[0060] The temperature processing module is configured to determine the heat dissipation area of ​​the laser radar surface, establish a plurality of temperature collection points in the heat dissipation area, obtain the real-time temperature value of each temperature collection point, generate a temperature interference data link according to all the real-time temperature values, determine the interference mark of each real-time temperature value on the temperature interference data link based on the standard heat dissipation value of the radiator, adjust the temperature interference data link according to the interference mark, and determine the heat dissipation temperature data link;

[0061] The second adjustment module is configured to parse the real-time temperature value on the heat dissipation temperature data link, determine the heat dissipation type of the lidar surface according to the analysis result, the heat dissipation type includes local heat dissipation and overall heat dissipation, determine the heat dissipation output value of the lidar surface based on the heat dissipation type and the heat dissipation area, determine the adjustment coefficient of the secondary fan speed according to the heat dissipation output value, and determine the target fan speed of the radiator according to the adjustment coefficient.

[0062] Specifically, the dust acquisition module uses image acquisition devices such as cameras to acquire image data on the laser radar surface, and analyzes the image data through image processing algorithms to extract key features of the laser radar surface, and then obtain the dust impact parameters of the laser radar surface. For the surface of the laser radar, the accumulation of dust will affect the heat dissipation effect. The dust impact parameters provide a basis for the subsequent adjustment of the initial fan speed of the radiator, reducing the risk of dust causing poor heat dissipation on the laser radar surface. The first adjustment module compares the current dust impact parameters with historical data. The system can evaluate the speed contrast between the dust condition on the current laser radar surface and the historical condition. If the speed contrast is high, the historical data can be used directly to determine the speed adjustment factor, and then determine the secondary fan speed to meet the laser radar surface's demand for heat dissipation. If the speed contrast is low, it indicates that the current dust condition is different from the historical condition. The speed adjustment factor is determined by the speed prediction model to obtain the secondary fan speed. The initial fan speed of the radiator can be dynamically adjusted according to the different dust conditions on the laser radar surface, which improves the adaptability and flexibility of the system. The temperature processing module is responsible for determining the heat dissipation area of ​​the laser radar surface. It also uses image acquisition devices such as cameras to determine its heat dissipation area according to the length, width and other dimensions of the laser radar surface, and Moreover, the temperature processing module sets several temperature collection points in the heat dissipation area, preferably 20 temperature collection points, which can be adjusted according to actual needs. These temperature collection points obtain the real-time temperature value in the heat dissipation area in real time through temperature devices such as temperature sensors. The temperature processing module generates a temperature interference data chain for all real-time temperature values ​​to avoid a single real-time temperature value from interfering with the data and affecting the determination of the subsequent adjustment coefficient. The interference mark of each real-time temperature value on the temperature interference data chain is determined based on the standard heat dissipation value of the radiator. According to the relationship between the real-time temperature value, the ambient temperature value, and the standard heat dissipation value, the interference mark is dynamically determined, thereby adjusting the temperature interference data chain, effectively avoiding the influence of the ambient temperature value on the radiator, thereby obtaining an accurate adjustment coefficient, improving the heat dissipation capacity of the laser radar surface, and optimizing the heat dissipation effect. The second adjustment module can determine whether the heat dissipation type of the laser radar surface is local heat dissipation or overall heat dissipation by parsing the temperature data chain, and adopts different heat dissipation strategies for different heat dissipation types, thereby dynamically determining the target fan speed of the radiator, ensuring that the laser radar can be effectively cooled in different ambient temperatures, and ensuring that the laser radar is maintained at the optimal operating temperature.

[0063] It can be understood that through the precise temperature collection, dust detection, historical data comparison and adjustment of fan speed by the dust acquisition module, the first adjustment module, the temperature processing module and the second adjustment module, the working state of the laser radar in a high temperature environment is ensured, thereby ensuring the reliability of the laser radar.

[0064] In some embodiments of the present application, when determining whether to adjust the initial fan speed of the radiator according to the dust impact parameter, the method includes:

[0065] The dust acquisition module compares the dust impact parameter with the preset standard dust impact parameter, and determines whether to adjust the initial fan speed of the radiator according to the comparison result;

[0066] When the dust impact parameter is greater than the standard dust impact parameter, it is determined to adjust the initial fan speed of the radiator;

[0067] When the dust impact parameter is less than or equal to the standard dust impact parameter, it is determined that the initial fan speed of the radiator is not adjusted, and the initial fan speed of the radiator is determined as the secondary fan speed.

[0068] In some embodiments of the present application, when determining to adjust the initial fan speed of the radiator, a comparison is performed based on the dust impact parameter and the historical data, and a speed contrast is determined based on the comparison result, including:

[0069] The first adjustment module obtains historical data, the historical data includes historical dust impact parameters, historical speed adjustment factors and historical initial fan speeds, and the historical dust impact parameters, historical speed adjustment factors and historical initial fan speeds correspond to each other;

[0070] When the first adjustment module determines the speed contrast ratio according to the comparison result, the speed contrast ratio is obtained by the following formula:

[0071] ;

[0072] in, Indicates the speed contrast, Indicates dust impact parameters, represents the historical dust impact parameter, Indicates the initial fan speed. Indicates the initial fan speed in history. and represents the weight coefficient, and .

[0073] Specifically, a camera or other image acquisition device is used to photograph or scan the surface of the laser radar, and image processing technology is used to extract features in the image, such as color, texture, brightness and other features, to reflect the dust impact parameters of the laser radar surface. The dust impact parameters of the surface are analyzed according to the extracted features, and whether to adjust the initial fan speed of the radiator is determined according to the analysis results. If the dust impact parameter is less than or equal to the standard dust impact parameter, it means that the dust condition on the laser radar surface meets the requirements of the initial fan speed for heat dissipation. If the dust impact parameter is greater than the standard dust impact parameter, it means that the dust condition on the laser radar surface meets the requirements of the initial fan speed for heat dissipation. This means that there is a lot of dust on the laser radar surface, which has affected the heat dissipation effect and the initial fan speed needs to be adjusted. When it is determined that adjustment is needed, the speed contrast is used to find data similar to the current adjustment in the historical data, so as to adjust the initial fan speed, thereby improving the accuracy of the system in adjusting the initial fan speed and ensuring the heat dissipation requirements of the laser radar.

[0074] In some embodiments of the present application, when the speed contrast is greater than or equal to the speed contrast threshold, determining the speed adjustment factor according to historical data includes:

[0075] The first adjustment module compares the speed contrast with a preset speed contrast threshold, and determines a speed adjustment factor according to the comparison result;

[0076] When there is data in the historical data whose rotation speed contrast is greater than or equal to the rotation speed contrast threshold, the rotation speed adjustment factor is determined according to the historical data;

[0077] When there is only one data in the historical data whose speed contrast is greater than or equal to the speed contrast threshold, the first adjustment module uses the historical speed adjustment factor corresponding to the data as the speed adjustment factor;

[0078] When the data in the historical data with a rotation speed contrast greater than the rotation speed contrast threshold is not unique, the first adjustment module uses the historical rotation speed adjustment factor corresponding to the maximum rotation speed contrast determined by the data as the rotation speed adjustment factor;

[0079] When there is not only one data in the historical data whose rotation speed contrast is equal to the rotation speed contrast threshold, the first adjustment module uses the average of the historical rotation speed adjustment factors corresponding to each data as the rotation speed adjustment factor.

[0080] Specifically, when there is a speed contrast greater than or equal to the speed contrast threshold in the historical data, the speed adjustment factor is determined according to the historical data, and the speed contrast threshold is used to judge the degree of matching between the current adjustment situation and the historical adjustment situation. When historical data with a higher speed contrast is found, these data can be directly used to determine the speed adjustment factor to ensure the reliability and consistency of adjusting the initial fan speed. Moreover, when there are multiple data greater than the speed contrast threshold, the historical speed adjustment factor corresponding to the largest speed contrast is preferentially selected as the speed adjustment factor, and the data closest to the adjustment situation and the historical adjustment situation is found to determine the speed adjustment factor, thereby improving the degree of automation of system adjustment and the accuracy and efficiency of adjusting the initial fan speed.

[0081] In some embodiments of the present application, when the speed adjustment factor is determined by the speed prediction model, and the speed of the secondary fan is determined according to the speed adjustment factor, the following steps are included:

[0082] The first adjustment module takes all historical initial fan speeds, all historical speed adjustment factors, and all historical dust impact parameters as a model data set, divides the model data set into a training set and a test set, and selects a random forest model as an initial model;

[0083] The first adjustment module iteratively trains the initial model according to the training set, and verifies the iteratively trained initial model according to the test set to obtain the prediction accuracy;

[0084] If the prediction accuracy of the initial model after the current iterative training is less than the prediction accuracy of the initial model after the previous iterative training, the absolute values ​​of the model parameters of the initial model after the current iterative training are summed through L1 regularization, and the squares of the model parameters of the initial model after the current iterative training are summed through L2 regularization. The summed results of L1 regularization and L2 regularization are added as penalty terms to the loss function of the initial model after the current iterative training, and iterative training is continued until the preset number of iterations is reached;

[0085] If the prediction accuracy of the initial model after the current iterative training is greater than or equal to the prediction accuracy of the initial model after the previous iterative training, the iterative training is stopped, the initial model after the current iterative training is determined as the speed prediction model, the dust impact parameter and the initial fan speed of the radiator are input into the speed prediction model, and the speed adjustment factor is determined;

[0086] The secondary fan speed is the product of the speed adjustment factor and the initial fan speed of the heat sink.

[0087] Specifically, all historical initial fan speeds, all historical speed adjustment factors, and all historical dust impact parameters are used as model data sets. The model data set contains various situations of initial fan speed adjustment and dust impact parameters in different periods. The model data set is divided into training set and test set. 50%-70% of the data is used as the training set, and the rest is used as the test set. It is ensured that both the training set and the test set contain data of various adjustment situations, thereby improving the generalization ability of the model. The random forest model is selected as the initial model. The random forest model contains multiple decision trees to capture the complex relationships in the data. L1 regularization and L2 regularization are used to prevent the initial model from overfitting during the training process, and to avoid the problem that the initial model relies too much on training data and leads to poor prediction on new data. L1 regularization penalizes model parameters by summing the absolute values ​​of the initial model parameters, which helps Feature selection and dimensionality reduction, L2 regularization penalizes model parameters by summing the squares of model parameters, which helps to reduce the complexity of the model and improve generalization ability. During the iterative training of the initial model, when the prediction accuracy is less than the prediction accuracy of the initial model after the previous iterative training, the regularization terms (L1 and L2) are added to the loss function for optimization to avoid the risk of overfitting of the initialized model for iterative training, thereby improving the stability and accuracy of the model. If the current prediction accuracy is higher or equal to the prediction accuracy after the last iterative training, the training is stopped and the initial model after the current iterative training is selected as the speed prediction model. The dust impact parameter and the initial fan speed of the radiator are input into the speed prediction model to obtain the speed adjustment factor and determine the secondary fan speed. As the speed adjustment factor increases, the secondary fan speed increases, thereby improving the heat dissipation effect of the system on the lidar.

[0088] It is understandable that the speed adjustment factor can be accurately obtained through the speed prediction model. In each iterative training, the model will try to learn the patterns and relationships in the data to improve its prediction or classification capabilities, so as to find out the impact of dust impact parameters on the initial fan speed at different times, and then determine the speed adjustment factor. In addition, in the iterative training, the model performance during iterative training is measured according to the prediction accuracy, including the loss function value, recall rate, etc., and then iterative training is carried out, which helps the model to more stably approach the global optimal solution. By continuously optimizing the parameters of the initial model, the speed prediction model is determined so that it can accurately output the speed adjustment factor, thereby improving the accuracy of system regulation.

[0089] In some embodiments of the present application, when determining the interference mark of each real-time temperature value on the temperature interference data link based on the standard heat dissipation value of the radiator, it includes:

[0090] The temperature processing module determines the temperature difference between each real-time temperature value and the ambient temperature value, and determines the interference mark of each real-time temperature value according to the temperature difference and the standard heat dissipation value of the radiator, wherein the interference mark includes a coincidence mark and a difference mark;

[0091] When the temperature difference is equal to the standard heat dissipation value, the temperature processing module generates a matching mark for the corresponding real-time temperature value;

[0092] When the temperature difference is not equal to the standard heat dissipation value, the temperature processing module generates a difference mark for the corresponding real-time temperature value.

[0093] In some embodiments of the present application, when adjusting the temperature interference data link according to the interference mark to determine the heat dissipation temperature data link, it includes:

[0094] When there is no matching mark on the temperature interference data link, the temperature processing module determines the temperature interference data link as a heat dissipation temperature data link;

[0095] When there is a matching mark on the temperature interference data link, the temperature processing module removes the real-time temperature value corresponding to the matching mark to obtain the heat dissipation temperature data link.

[0096] Specifically, by accurately tracking the difference between the real-time temperature value and the ambient temperature value, it can be adjusted adaptively according to different environmental conditions. When the temperature difference is not equal to the standard heat dissipation value, it indicates that the current ambient temperature affects the heat dissipation of the radiator to the laser radar. The standard heat dissipation value of the radiator indicates that in a suitable ambient temperature, the laser radar can be effectively cooled to achieve the ideal working temperature of the laser radar. The temperature difference is not equal to the standard heat dissipation value, indicating that the current ambient temperature fluctuates. If the secondary fan speed is not adjusted, the heat dissipation effect of the laser radar will be affected. According to the comparison between the temperature difference and the standard heat dissipation value, an interference mark is dynamically generated, and the heat dissipation temperature data chain is determined according to the interference mark, laying a data foundation for subsequent adjustments. When there is a matching mark, it means that the temperature difference at this time is equal to the standard heat dissipation value, which meets the heat dissipation requirements of the laser radar. If the temperature interference data chain is all matching marks, it means that there is no need to adjust the secondary fan speed. Otherwise, the matching mark needs to be removed, and the heat dissipation temperature data chain is determined according to the difference mark, which improves the system's adaptability and compatibility to different ambient temperatures, thereby optimizing the efficiency of heat dissipation adjustment.

[0097] In some embodiments of the present application, when parsing the real-time temperature value on the heat dissipation temperature data link and determining the heat dissipation type of the laser radar surface according to the parsing result, it includes:

[0098] The second adjustment module analyzes the real-time temperature values ​​corresponding to the difference marks on the heat dissipation temperature data chain, determines the temperature average of all real-time temperature values, counts the first difference number of difference marks whose real-time temperature values ​​are greater than the temperature average, and counts the second difference number of difference marks whose real-time temperature values ​​are less than the temperature average;

[0099] When the first difference number is greater than or equal to the second difference number, it is determined that the heat dissipation type of the laser radar surface is overall heat dissipation;

[0100] When the first difference amount is less than the second difference amount, it is determined that the heat dissipation type of the laser radar surface is local heat dissipation.

[0101] In some embodiments of the present application, when determining the heat dissipation output value of the laser radar surface based on the heat dissipation type and the heat dissipation area, it includes:

[0102] The second adjustment module obtains the heat dissipation output value according to the following formula:

[0103] ;

[0104] in, Indicates the heat dissipation output value, Represents the area of ​​the heat dissipation region, represents the density of the lidar material, and Represents the weight coefficient. When the heat dissipation type of the laser radar surface is determined to be overall heat dissipation, The value is [0.5, 1], The value is (0, 0.5). When the heat dissipation type of the laser radar surface is determined to be local heat dissipation, The value is (0, 0.5), The value is [0.5, 1], represents the specific heat capacity of the lidar material, Represents the thermal conductivity of the lidar material.

[0105] Specifically, by analyzing the real-time temperature values ​​of the difference marks, we can count the number of difference marks whose real-time temperature values ​​are greater than and less than the temperature mean, so as to determine the heat dissipation type of the laser radar surface. The temperature mean represents the comprehensive level of the real-time temperature values ​​corresponding to the difference marks. When the first difference number is greater than or equal to the second difference number, it indicates that there are more difference marks greater than the temperature mean, and the corresponding real-time temperature values ​​that exceed the temperature mean are also greater. This indicates that the heat dissipation type of the laser radar surface is overall heat dissipation, which requires comprehensive heat dissipation. Otherwise, it means that local heat dissipation is required to ensure that the radiator fan can make appropriate adjustments according to different heat dissipation types. Determine the heat dissipation output value of the laser radar surface based on the heat dissipation type and heat dissipation area. and As a weight coefficient, when the heat dissipation type is overall heat dissipation, the heat dissipation output value obtained is slightly higher than the heat dissipation output value obtained when the heat dissipation type is local heat dissipation. In addition, the heat dissipation output value is calculated based on the properties of the lidar material: thermal conductivity, specific heat capacity and density, as well as the area of ​​the heat dissipation area. The properties of the lidar material are determined according to the proportion of the lidar material. For example, when the iron material proportion of the lidar is 70% and the aluminum material proportion is 30, the lidar material is determined to be iron, and its thermal conductivity, specific heat capacity and density are determined according to the physical properties of iron. The heat dissipation output value lays a data foundation for the subsequent adjustment of the fan speed.

[0106] In some embodiments of the present application, when determining the adjustment coefficient of the secondary fan speed according to the heat dissipation output value and determining the target fan speed of the radiator according to the adjustment coefficient, it includes:

[0107] The second adjustment module pre-sets a first preset heat dissipation output value and a second preset heat dissipation output value, and pre-sets a first preset adjustment coefficient, a second preset adjustment coefficient, and a third preset adjustment coefficient;

[0108] When the heat dissipation output value is greater than the first preset heat dissipation output value, determining the first preset adjustment coefficient as the adjustment coefficient of the secondary fan speed;

[0109] When the heat dissipation output value is less than or equal to the first preset heat dissipation output value and greater than the second preset heat dissipation output value, determining the second preset adjustment coefficient as the adjustment coefficient of the secondary fan speed;

[0110] When the heat dissipation output value is less than or equal to the second preset heat dissipation output value, determining the third preset adjustment coefficient as the adjustment coefficient of the secondary fan speed;

[0111] The target fan speed is a product value of the secondary fan speed and the adjustment coefficient of the secondary fan speed.

[0112] Specifically, the first preset heat dissipation output value is preferably 5, the second preset heat dissipation output value is preferably 3, the first preset adjustment coefficient is preferably 1.7, the second preset adjustment coefficient is preferably 1.5, and the third preset adjustment coefficient is preferably 1.3. For example, when the material of the laser radar is iron, 80W / (m·K), is 450J / (kg·K), 7874kg / m 3 , 0.01m 2 , the heat dissipation type of the laser radar surface is overall heat dissipation, Select 1, Choose 0.4, then When the value is equal to 5.47, the first preset adjustment coefficient is selected as the adjustment coefficient of the secondary fan speed. When the heat dissipation output value is high, it indicates that the laser radar has a high heat dissipation demand and needs to improve the heat dissipation effect of the radiator. By dynamically selecting the adjustment coefficient of the secondary fan speed, the target fan speed is obtained, which improves the accuracy and reliability of the system's adjustment of the secondary fan speed, thereby improving the heat dissipation effect on the surface of the laser radar.

[0113] In summary, the beneficial effects of the present invention are as follows: according to the dust condition on the surface of the laser radar, the dust impact parameters are obtained by analyzing the image data, and it is determined whether the initial fan speed of the radiator needs to be adjusted, thereby avoiding the accumulation of dust affecting the heat dissipation effect of the fan on the surface of the laser radar, thereby ensuring the stability and reliability of the laser radar in a dusty environment, and comparing the dust impact parameters with historical data to determine the speed contrast, and based on the speed contrast, the speed adjustment factor is determined according to the speed prediction model or historical data, so as to adjust the initial fan speed, thereby improving the flexibility and adjustment accuracy of the system and reducing the risk that the initial fan speed of the radiator cannot meet the heat dissipation requirements of the laser radar surface. Multiple temperature collection points are set in the heat dissipation area, and the temperature changes of each collection point are monitored in real time. A temperature interference data chain is generated, and the interference mark is determined according to the standard heat dissipation value, which effectively identifies the temperature fluctuations in the high-temperature environment, ensures the accuracy of the temperature data, and improves the system's adaptability to high-temperature environments. In addition, by analyzing the heat dissipation temperature data chain, the heat dissipation type of the lidar surface is judged, and the heat dissipation output value is calculated according to different heat dissipation types and heat dissipation areas, ensuring that the system can determine the adjustment coefficient of the secondary fan speed according to different situations, thereby improving the heat dissipation efficiency of the lidar surface and the system's adaptability, effectively avoiding performance degradation or equipment damage of the lidar due to overheating, and enabling the lidar to maintain a good working condition in a high-temperature environment.

[0114] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0115] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0116] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0117] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A laser radar heat dissipation system, characterized in that: include: a dust acquisition module configured to collect image data of the laser radar surface, analyze the image data, obtain dust impact parameters of the laser radar surface, and determine whether to adjust the initial fan speed of the radiator according to the dust impact parameters; A first adjustment module is configured to compare the dust impact parameter with the historical data when determining to adjust the initial fan speed of the radiator, determine a speed contrast according to the comparison result, determine a speed adjustment factor according to the historical data when the speed contrast is greater than or equal to a speed contrast threshold, determine a speed adjustment factor through a speed prediction model when the speed contrast is less than the speed contrast threshold, and determine the secondary fan speed according to the speed adjustment factor; a temperature processing module configured to determine a heat dissipation area on the surface of the laser radar, establish a plurality of temperature collection points in the heat dissipation area, obtain a real-time temperature value of each temperature collection point, generate a temperature interference data link according to all the real-time temperature values, determine an interference mark of each real-time temperature value on the temperature interference data link based on a standard heat dissipation value of the radiator, adjust the temperature interference data link according to the interference mark, and determine a heat dissipation temperature data link; The second adjustment module is configured to parse the real-time temperature value on the heat dissipation temperature data link, determine the heat dissipation type of the laser radar surface according to the analysis result, the heat dissipation type includes local heat dissipation and overall heat dissipation, determine the heat dissipation output value of the laser radar surface based on the heat dissipation type and the heat dissipation area, determine the adjustment coefficient of the secondary fan speed according to the heat dissipation output value, and determine the target fan speed of the radiator according to the adjustment coefficient.

2. The laser radar heat dissipation system according to claim 1, characterized in that: When determining whether to adjust the initial fan speed of the radiator according to the dust impact parameter, the method includes: The dust acquisition module compares the dust impact parameter with a preset standard dust impact parameter, and determines whether to adjust the initial fan speed of the radiator according to the comparison result; When the dust impact parameter is greater than the standard dust impact parameter, determining to adjust the initial fan speed of the radiator; When the dust impact parameter is less than or equal to the standard dust impact parameter, it is determined that the initial fan speed of the radiator is not adjusted, and the initial fan speed of the radiator is determined as the secondary fan speed.

3. The laser radar heat dissipation system according to claim 2, characterized in that: When determining to adjust the initial fan speed of the radiator, comparing the dust impact parameter with the historical data, and determining the speed contrast ratio according to the comparison result, the method includes: The first adjustment module acquires the historical data, wherein the historical data includes a historical dust impact parameter, a historical speed adjustment factor, and a historical initial fan speed, and the historical dust impact parameter, the historical speed adjustment factor, and the historical initial fan speed correspond to each other; When the first adjustment module determines the rotation speed contrast according to the comparison result, the rotation speed contrast is obtained by the following formula: ； in, Indicates the speed contrast, Indicates dust impact parameters, represents the historical dust impact parameter, Indicates the initial fan speed. Indicates the initial fan speed in history. and represents the weight coefficient, and .

4. The laser radar heat dissipation system according to claim 3, characterized in that: When the speed contrast is greater than or equal to the speed contrast threshold, determining a speed adjustment factor according to the historical data includes: The first adjustment module compares the rotation speed contrast with a preset rotation speed contrast threshold, and determines a rotation speed adjustment factor according to the comparison result; When there is data in the historical data whose rotation speed contrast is greater than or equal to the rotation speed contrast threshold, determining the rotation speed adjustment factor according to the historical data; When there is only one data in the historical data whose rotation speed contrast is greater than or equal to the rotation speed contrast threshold, the first adjustment module uses the historical rotation speed adjustment factor corresponding to the data as the rotation speed adjustment factor; When there is not only one data in the historical data whose rotation speed contrast is greater than the rotation speed contrast threshold, the first adjustment module uses the historical rotation speed adjustment factor corresponding to the rotation speed contrast with the largest rotation speed contrast determined by the data as the rotation speed adjustment factor; When there is not only one piece of data in the historical data whose rotation speed contrast is equal to the rotation speed contrast threshold, the first adjustment module uses the average of the historical rotation speed adjustment factors corresponding to each piece of data as the rotation speed adjustment factor.

5. The laser radar heat dissipation system according to claim 4, characterized in that: When the speed adjustment factor is determined by the speed prediction model and the speed of the secondary fan is determined according to the speed adjustment factor, the method includes: The first adjustment module takes all historical initial fan speeds, all historical speed adjustment factors, and all historical dust impact parameters as a model data set, divides the model data set into a training set and a test set, and selects a random forest model as an initial model; The first adjustment module iteratively trains the initial model according to the training set, and verifies the iteratively trained initial model according to the test set to obtain a prediction accuracy rate; If the prediction accuracy of the initial model after the current iterative training is less than the prediction accuracy of the initial model after the previous iterative training, the absolute values ​​of the model parameters of the initial model after the current iterative training are summed through L1 regularization, and the squares of the model parameters of the initial model after the current iterative training are summed through L2 regularization. The summed results of L1 regularization and L2 regularization are added as penalty terms to the loss function of the initial model after the current iterative training, and iterative training is continued until the preset number of iterations is reached; If the prediction accuracy of the initial model after the current iterative training is greater than or equal to the prediction accuracy of the initial model after the previous iterative training, the iterative training is stopped, the initial model after the current iterative training is determined as the speed prediction model, the dust impact parameter and the initial fan speed of the radiator are input into the speed prediction model, and the speed adjustment factor is determined; The secondary fan speed is a product of the speed adjustment factor and the initial fan speed of the radiator.

6. The laser radar heat dissipation system according to claim 5, characterized in that: When determining the interference mark of each real-time temperature value on the temperature interference data link based on the standard heat dissipation value of the radiator, it includes: The temperature processing module determines the temperature difference between each real-time temperature value and the ambient temperature value, and determines the interference mark of each real-time temperature value according to the temperature difference and the standard heat dissipation value of the radiator, wherein the interference mark includes a matching mark and a difference mark; When the temperature difference is equal to the standard heat dissipation value, the temperature processing module generates the matching mark for the corresponding real-time temperature value; When the temperature difference is not equal to the standard heat dissipation value, the temperature processing module generates the difference mark for the corresponding real-time temperature value.

7. The laser radar heat dissipation system according to claim 6, characterized in that: When the temperature interference data link is adjusted according to the interference mark to determine the heat dissipation temperature data link, it includes: When the matching mark does not exist on the temperature interference data link, the temperature processing module determines the temperature interference data link as the heat dissipation temperature data link; When the matching mark exists on the temperature interference data link, the temperature processing module removes the real-time temperature value corresponding to the matching mark to obtain the heat dissipation temperature data link.

8. The laser radar heat dissipation system according to claim 7, characterized in that: When the real-time temperature value on the heat dissipation temperature data link is parsed and the heat dissipation type of the laser radar surface is determined according to the parsing result, it includes: The second regulating module analyzes the real-time temperature values ​​corresponding to the difference marks on the heat dissipation temperature data chain, determines the temperature average of all real-time temperature values, counts the first difference number of the difference marks whose real-time temperature values ​​are greater than the temperature average, and counts the second difference number of the difference marks whose real-time temperature values ​​are less than the temperature average; When the first difference number is greater than or equal to the second difference number, it is determined that the heat dissipation type of the laser radar surface is overall heat dissipation; When the first difference number is less than the second difference number, it is determined that the heat dissipation type of the laser radar surface is local heat dissipation.

9. The laser radar heat dissipation system according to claim 8, characterized in that: When determining the heat dissipation output value of the laser radar surface based on the heat dissipation type and the heat dissipation area, it includes: The second adjustment module obtains the heat dissipation output value according to the following formula: ； in, Indicates the heat dissipation output value, Represents the area of ​​the heat dissipation region, represents the density of the lidar material, and Represents the weight coefficient. When the heat dissipation type of the laser radar surface is determined to be overall heat dissipation, The value is [0.5, 1], The value is (0, 0.5). When the heat dissipation type of the laser radar surface is determined to be local heat dissipation, The value is (0, 0.5), The value is [0.5, 1], represents the specific heat capacity of the lidar material, Represents the thermal conductivity of the lidar material.

10. The laser radar heat dissipation system according to claim 9, characterized in that: When determining the adjustment coefficient of the secondary fan speed according to the heat dissipation output value, and determining the target fan speed of the radiator according to the adjustment coefficient, the method includes: The second adjustment module pre-sets a first preset heat dissipation output value and a second preset heat dissipation output value, and pre-sets a first preset adjustment coefficient, a second preset adjustment coefficient, and a third preset adjustment coefficient; When the heat dissipation output value is greater than the first preset heat dissipation output value, determining the first preset adjustment coefficient as the adjustment coefficient of the secondary fan speed; When the heat dissipation output value is less than or equal to the first preset heat dissipation output value and greater than the second preset heat dissipation output value, determining the second preset adjustment coefficient as the adjustment coefficient of the secondary fan speed; When the heat dissipation output value is less than or equal to the second preset heat dissipation output value, determining the third preset adjustment coefficient as the adjustment coefficient of the secondary fan speed; The target fan speed is a product value of the secondary fan speed and an adjustment coefficient of the secondary fan speed.

Images