Intelligent street lamp management system and method based on artificial intelligence data analysis
By installing an image acquisition unit and an artificial intelligence data analysis module at the street light, identifying pedestrians and dynamically controlling the on-stage state of the street light, the light and dark contrast problem caused by untimely opening of street lights in the existing technology is solved, and better lighting effects and power savings are achieved. .
Patent Information
- Application Number
- CN202411363744.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-28
- Publication Date
- 2025-02-18
AI Technical Summary
The existing smart street light management system only turns on the street lights when pedestrians approach the street lights, resulting in a strong contrast between the light and dark areas, making it difficult for pedestrians to see objects in the dark and find specific information about the road ahead.
By installing an image acquisition unit at the street light, identifying pedestrians and uploading location information, using artificial intelligence data analysis to generate pedestrian movement paths and movement probability, dynamically control the opening and closing of street lights, and ensure that the roads within the pedestrian's field of vision are properly illuminated.
Improve the street light control effect, help pedestrians better obtain specific information on the road ahead, reduce the impact of the contrast effect of the light and dark area alternating areas on pedestrians, and save electricity resources.
Smart Images
Figure CN119478800A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of street lamp management, and in particular to an intelligent street lamp management system and method based on artificial intelligence data analysis. Background Art
[0002] The smart street light management system is a high-tech street light management tool that integrates artificial intelligence data analysis. It realizes intelligent management of street lights by installing various sensors and smart devices on street lights, and also provides a variety of service functions for pedestrians. In the current management of street lights, in order to save electricity resources, street lights are usually turned on to illuminate pedestrians when pedestrians arrive near street lights. When the street lights are turned on, there may be a strong contrast between the brightly illuminated area and the unilluminated area, which makes it difficult for pedestrians to see objects in the dark and pedestrians cannot obtain specific information on the road they are going to next. Therefore, how to improve the control effect of street lights and help pedestrians better obtain specific information on the road ahead has become an urgent problem to be solved. Summary of the invention
[0003] The purpose of the present invention is to provide a smart street lamp management system and method based on artificial intelligence data analysis to solve the problems raised in the prior art.
[0004] To achieve the above objectives, the present invention provides the following technical solutions: a smart street lamp management method based on artificial intelligence data analysis, comprising:
[0005] S11, the image acquisition unit at the street lamp acquires a scene image of the road at the street lamp, performs recognition analysis on the acquired scene image, recognizes pedestrians in the scene image, and if there are pedestrians in the scene image, uploads the location information of the pedestrian and the current street lamp;
[0006] S12, generating a moving path of the pedestrian and the probability of the pedestrian moving along each moving path according to the current position of the pedestrian and the position information that the pedestrian has passed; turning on and off the street light based on the moving path of the pedestrian and the probability of the pedestrian moving along each moving path;
[0007] S13, when the image acquisition unit at the next street light determines that the pedestrian has arrived at the next street light, it updates the pedestrian's moving path and the probability of the pedestrian moving along each moving path, and turns the street light on and off based on the updated pedestrian's moving path and the probability of the pedestrian moving along each moving path, and repeats step S13 until the pedestrian arrives at the destination.
[0008] In step S12, the step of generating the pedestrian's moving path and the probability of the pedestrian moving along each moving path according to the pedestrian's current location and the location information that the pedestrian has passed through further includes the following steps:
[0009] Obtain the historical movement path data of pedestrians in the park, and add the street lamp positions that the current pedestrian has passed to the set U in the order of movement according to the street lamp positions that the current pedestrian has passed; let the elements in the set U be relevant elements, and filter out the movement paths that contain all relevant elements and whose arrangement order is the same as that of the set U from the historical movement path data, and calculate the probability of the selected movement path appearing: path i =n i / n, where path i represents the probability of the i-th selected moving path appearing, n i represents the number of times the i-th filtered moving path appears, and n represents the total number of times all moving paths appear.
[0010] In order to determine the on / off status of street lights, it is necessary to determine the path chosen by pedestrians when they move forward and provide lighting effects for the path ahead within the pedestrian's field of vision so that pedestrians can obtain specific information on the path ahead. As the number of street lights that pedestrians have passed through increases within the park, the number of movement paths that match the pedestrian's behavior gradually decreases until the pedestrian's movement path is completely determined.
[0011] In step S12, the method of turning on and off the street lamp based on the moving paths of pedestrians and the probability of pedestrians moving along each moving path also includes the following steps:
[0012] S31, generating a pedestrian's field of vision according to the pedestrian's position and moving direction, and adding roads included in the moving path whose probability of appearing in the pedestrian's field of vision is not 0 to the demand domain;
[0013] S32, testing the lighting function of the street lamp. When the street lamp is turned on and the brightness of the road reaches a specified value under the lighting of the street lamp, the road is added to the influence domain of the street lamp to obtain the influence domain of all street lamps in the park.
[0014] S33, if the influence domain of the street lamp overlaps with the demand domain, the street lamp is added to the candidate set V, and the probability of a pedestrian passing each street lamp in the candidate set V is calculated: P j =∑path j , where P j represents the probability of a pedestrian passing the jth street light in the set V, path j represents the probability of the moving path containing the jth street light appearing.
[0015] Since pedestrians are captured by the image capture unit at the street lamp, the position of the street lamp is the position of the pedestrian. When the position of the street lamp where the pedestrian is located changes, the pedestrian's demand domain is updated; at the same time, the position of the street lamp where the pedestrian passes reflects the pedestrian's movement path. When the image capture unit of a street lamp detects a pedestrian, the image capture units of other street lamps also detect the pedestrian, and it is judged that the pedestrian has passed the previous street lamp; the pedestrian's field of vision can be refreshed according to the position of the street lamp to determine the pedestrian's field of vision at different street lamp positions. In order to reduce the loss of power resources caused by street lamps, the opening and closing of street lamps are controlled, and only the street lamps that illuminate the road ahead of pedestrians are turned on. In order to reduce the switching loss of street lamps, under the condition of satisfying the condition of illuminating the road ahead of pedestrians, street lamps with a higher probability of pedestrians passing are preferentially turned on; for example, there are street lamps A and B, street lamp A is on the road ahead of pedestrians, and the probability of street lamp A being on the road ahead of pedestrians is greater than street lamp B, then pedestrians will have a higher demand for street lamp A, because street lamp A can provide more lighting for pedestrians.
[0016] In step S12, the method of turning on and off the street lamp based on the moving paths of pedestrians and the probability of pedestrians moving along each moving path also includes the following steps:
[0017] S41, binary-encode the streetlights in the candidate set V to obtain individuals, where each binary bit of the individual corresponds to a streetlight in the candidate set V; if the streetlight needs to be turned on, the encoding value of the bit corresponding to the streetlight is 1, and if the streetlight does not need to be turned on, the encoding value of the bit corresponding to the streetlight is 0;
[0018] S42, obtaining the probability of a pedestrian passing each street lamp in the candidate set V, creating an initial population that meets the constraint condition according to the probability of a pedestrian passing each street lamp in the candidate set V, and setting the number of iterations to 0; the constraint condition is that after the influence domains of all street lamps that need to be turned on are added together, the demand domain can be completely covered;
[0019] S43, calculate the fitness value of each individual in the population, fit = k1(1 / x + k2×y), where fit represents the fitness value of the individual, x represents the number of street lamps that need to be turned on corresponding to the individual, let the street lamps that need to be turned on corresponding to the individual be the relevant street lamps, y represents the average probability of pedestrians passing the relevant street lamps, let P u represents the probability of a pedestrian passing the uth relevant street light, then y = ∑P u / n u , n urepresents the number of relevant street lamps; k1 is the scaling factor, which is set according to the needs; k2 is the restriction coefficient, which is used to compare the pros and cons of individuals who turn on the same number of street lamps. The number of turned-on street lamps is used as the first judgment criterion. The fewer the number of turned-on street lamps, the higher the fitness of the individual; when there are individuals with the same number of turned-on street lamps, the average probability of pedestrians passing the relevant street lamps is used as the second judgment criterion. The higher the average probability of pedestrians passing the relevant street lamps, the higher the fitness of the individual. k2 is set to a constant less than 1 / N(N-1), where N is the number of street lamps in the alternative set V;
[0020] S44, select individuals in the population according to their fitness values and pass the selected individuals on to the next generation;
[0021] S45, performing a crossover operation on the individuals selected in step S44, exchanging chromosomes to generate new individuals, and determining whether the addition of the influence domains of all street lamps that need to be turned on corresponding to the new individuals can completely cover the demand domain. If not, the new individuals are excluded from the population. If they can completely cover the demand domain, the new individuals are retained and become valid solutions.
[0022] S46, mutate the individuals with a certain probability to create new genetic genes; determine whether the influence domains of all street lamps that need to be turned on corresponding to the new individuals obtained after the mutation can completely cover the demand domain after adding up; if not, exclude the new individuals from the population; if it can completely cover the demand domain, retain the new individuals and make them valid solutions; use all valid solutions as the new generation population;
[0023] S47, determine whether the termination condition is met. If so, stop the algorithm, find the individual with the highest fitness value, and determine the on and off status of the street light according to the binary code of the individual with the highest fitness value; if not, add 1 to the number of iterations and return to step S43; the termination condition refers to the number of iterations reaching the set value.
[0024] In order to reduce the loss of electric power resources caused by street lamps, the fewer the number of street lamps turned on, the higher the lighting efficiency. Therefore, street lamps are the first judgment criterion for the fitness value. The optimal individual may not be determined by the number of street lamps turned on. The number of street lamps that need to be turned on may be the same. The second judgment criterion is added again. The second judgment criterion affects the number of times the street lamps are turned on and off. For example, two street lamps on different roads after a fork in the road can both provide lighting for the fork in the road. When the fork in the road is in the demand domain, street lamps with a higher probability of pedestrians passing through are turned on first. In order to prevent the second judgment criterion from affecting the first judgment criterion, the amplitude of the second judgment criterion is limited.
[0025] In step S12, the following steps are also included:
[0026] When the number of pedestrians on the road is not one, each pedestrian corresponds to a demand domain, and the demand domains of all pedestrians are added together. The overall alternative set is obtained by taking the union of the alternative sets of all pedestrians. Each pedestrian corresponds to the average probability of passing the relevant street lamp. The average probability of all pedestrians passing the relevant street lamp is added together to obtain the overall average probability of passing the relevant street lamp. Repeat steps S41 and S45 to determine the on and off status of the street lamp, wherein the fitness value is determined according to the number of street lamps that need to be turned on and the overall average probability of passing the relevant street lamp.
[0027] When the number of pedestrians on the road is not uniform, the demand domain of each pedestrian is first analyzed separately, and then all demand domains are added together to obtain the overall demand domain;
[0028] In step S12, the following steps are also included:
[0029] The historical movement path data of pedestrians in the park is obtained, and the scenes of pedestrians arriving at each street lamp on different movement paths are pre-stored. The on and off states of the street lamps in each scene are pre-determined and stored in the database. When the image acquisition unit at the street lamp recognizes the pedestrian, the position information of the street lamp passed by the pedestrian is obtained, the position information of the pedestrian has passed is obtained, and the on and off states of the street lamps corresponding to the current position information of the pedestrian have passed are obtained from the database to control the street lamps.
[0030] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a smart street light management system based on artificial intelligence data analysis, comprising: a data analysis module, a control module, a data storage module and a street light module; the output end of the street light module is connected to the input end of the data storage module and the control module, so as to identify pedestrians on the road and provide road lighting for pedestrians through street lights; the output end of the data storage module is connected to the input end of the data analysis module, so as to store the historical movement data of pedestrians in the park; the output end of the data analysis module is connected to the input end of the control module, so as to determine the street lights that need to be turned on and off based on the location of the pedestrians and the location information that the pedestrians have passed; the output end of the control module is connected to the input end of the street light module, so as to generate control signals for turning on and off the street lights.
[0031] The street lamp module also includes an image capture unit, a first communication unit and a lighting unit; the image capture unit is used to capture images on the road at the street lamp and identify pedestrians on the road; the communication unit is used to upload the location information of the street lamp and the pedestrian information to the control module and the data storage module; the lighting unit is used to provide lighting for pedestrians on the road. The data analysis module also includes a database, a path analysis unit, an artificial intelligence unit and a second communication unit; the path analysis unit generates the pedestrian's moving path and the probability of the pedestrian moving along each moving path according to the pedestrian's current location and the location information that has been passed; the artificial intelligence unit determines the on and off state of the street lamp through a genetic algorithm, and provides the pedestrian with the required lighting service by turning on the least number of street lamps. When the number of turned-on street lamps is the same, the street lamp is turned on according to the average probability of pedestrians passing the street lamp to provide the pedestrian with the required lighting service; the database is used to store the iterative results of the genetic algorithm; the second communication unit is used to exchange data with the control module. The control module also includes a third communication unit and a control unit; the third communication unit is used to obtain the pedestrian's current position information and passed position information from the street light module, and obtain the on and off status of the street light from the database based on the pedestrian's current position information and passed position information; the control unit is used to generate an on and off control signal for the street light.
[0032] Compared with the prior art, the beneficial effects of the present invention are: analyzing the movement path of pedestrians, providing lighting for the roads within the pedestrian's field of vision, helping pedestrians to better obtain specific information on the road ahead, and reducing the impact of the contrast effect caused by the lack of lighting on the road ahead on pedestrians; by controlling the on and off status of street lamps, the effect of saving electricity resources is achieved. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 This is a schematic diagram of the structure of the smart street light management system based on artificial intelligence data analysis of the present invention. DETAILED DESCRIPTION
[0034] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0035] Example: Figure 1As shown, the present invention provides a technical solution, a smart street light management system based on artificial intelligence data analysis, including a data analysis module, a control module, a data storage module and a street light module; the output end of the street light module is connected to the input end of the data storage module and the control module, so as to identify pedestrians on the road and provide road lighting for pedestrians through street lights; the output end of the data storage module is connected to the input end of the data analysis module, so as to store the historical movement data of pedestrians in the park; the output end of the data analysis module is connected to the input end of the control module, and the street lights that need to be turned on and off are determined based on the location of the pedestrians and the location information that the pedestrians have passed; the output end of the control module is connected to the input end of the street light module, so as to generate a control signal for turning on and off the street lights.
[0036] The street lamp module also includes an image capture unit, a first communication unit and a lighting unit; the image capture unit is used to capture images on the road at the street lamp and identify pedestrians on the road; the communication unit is used to upload the location information of the street lamp and the pedestrian information to the control module and the data storage module; the lighting unit is used to provide lighting for pedestrians on the road. The data analysis module also includes a database, a path analysis unit, an artificial intelligence unit and a second communication unit; the path analysis unit generates the pedestrian's moving path and the probability of the pedestrian moving along each moving path according to the pedestrian's current location and the location information that has been passed; the artificial intelligence unit determines the on and off state of the street lamp through a genetic algorithm, and provides the pedestrian with the required lighting service by turning on the least number of street lamps. When the number of turned-on street lamps is the same, the street lamp is turned on according to the average probability of pedestrians passing the street lamp to provide the pedestrian with the required lighting service; the database is used to store the iterative results of the genetic algorithm; the second communication unit is used to exchange data with the control module. The control module also includes a third communication unit and a control unit; the third communication unit is used to obtain the pedestrian's current position information and passed position information from the street light module, and obtain the on and off status of the street light from the database based on the pedestrian's current position information and passed position information; the control unit is used to generate an on and off control signal for the street light.
[0037] Embodiment: The present invention provides a technical solution, a smart street lamp management method based on artificial intelligence data analysis, comprising:
[0038] S11, the image acquisition unit at the street lamp acquires a scene image of the road at the street lamp, performs recognition analysis on the acquired scene image, recognizes pedestrians in the scene image, and if there are pedestrians in the scene image, uploads the location information of the pedestrian and the current street lamp;
[0039] S12, obtaining historical movement path data of pedestrians in the park, pre-storing scenes of pedestrians arriving at various street lamps along different movement paths, pre-determining the on and off states of the street lamps in each scene and storing them in a database, when the image acquisition unit at the street lamp recognizes a pedestrian, obtaining the position information of the street lamp that the pedestrian has passed, obtaining the position information that the pedestrian has passed, obtaining the on and off states of the street lamp corresponding to the current position information that the pedestrian has passed from the database, and controlling the street lamp;
[0040] Obtain the historical movement path data of pedestrians in the park, and add the street lamp positions that the current pedestrian has passed to the set U in the order of movement according to the street lamp positions that the current pedestrian has passed; let the elements in the set U be relevant elements, and filter out the movement paths that contain all relevant elements and whose arrangement order is the same as that of the set U from the historical movement path data, and calculate the probability of the selected movement path appearing: path i =n i / n, where path i represents the probability of the i-th selected moving path appearing, n i represents the number of times the i-th selected moving path appears, and n represents the total number of times all moving paths appear;
[0041] S31, generating a pedestrian's field of vision according to the pedestrian's position and moving direction, and adding roads included in the moving path whose probability of appearing in the pedestrian's field of vision is not 0 to the demand domain;
[0042] S32, testing the lighting function of the street lamp. When the street lamp is turned on and the brightness of the road reaches a specified value under the lighting of the street lamp, the road is added to the influence domain of the street lamp to obtain the influence domain of all street lamps in the park.
[0043] S33, if the influence domain of the street lamp overlaps with the demand domain, the street lamp is added to the candidate set V, and the probability of a pedestrian passing each street lamp in the candidate set V is calculated: P j =∑path j , where P j represents the probability of a pedestrian passing the jth street light in the set V, path j represents the probability of the moving path containing the jth street light appearing.
[0044] When 4 moving paths are screened out that contain all relevant elements and the order of the relevant elements is the same as that of set U, the probability of each moving path appearing is one-fourth, and only 2 moving paths contain the j-th street lamp, then the probability of a pedestrian passing the j-th street lamp in set V is 1 / 4+1 / 4=1 / 2.
[0045] Turn street lights on and off using genetic algorithms:
[0046] S41, binary-encode the streetlights in the candidate set V to obtain individuals, where each binary bit of the individual corresponds to a streetlight in the candidate set V; if the streetlight needs to be turned on, the encoding value of the bit corresponding to the streetlight is 1, and if the streetlight does not need to be turned on, the encoding value of the bit corresponding to the streetlight is 0;
[0047] S42, obtaining the probability of a pedestrian passing each street lamp in the candidate set V, creating an initial population that meets the constraint condition according to the probability of a pedestrian passing each street lamp in the candidate set V, and setting the number of iterations to 0; the constraint condition is that after the influence domains of all street lamps that need to be turned on are added together, the demand domain can be completely covered;
[0048] S43, calculate the fitness value of each individual in the population, fit = k1(1 / x + k2×y), where fit represents the fitness value of the individual, x represents the number of street lamps that need to be turned on corresponding to the individual, let the street lamps that need to be turned on corresponding to the individual be the relevant street lamps, y represents the average probability of pedestrians passing the relevant street lamps, let P u represents the probability of a pedestrian passing the uth relevant street light, then y = ∑P u / n u , n u represents the number of relevant street lamps; k1 is the scaling factor, which is set according to the needs; k2 is the restriction coefficient, which is used to compare the pros and cons of individuals who turn on the same number of street lamps. The number of turned-on street lamps is used as the first judgment criterion. The fewer the number of turned-on street lamps, the higher the fitness of the individual; when there are individuals with the same number of turned-on street lamps, the average probability of pedestrians passing the relevant street lamps is used as the second judgment criterion. The higher the average probability of pedestrians passing the relevant street lamps, the higher the fitness of the individual. k2 is set to a constant less than 1 / N(N-1), where N is the number of street lamps in the alternative set V;
[0049] S44, select individuals in the population according to their fitness values and pass the selected individuals on to the next generation;
[0050] S45, performing a crossover operation on the individuals selected in step S44, exchanging chromosomes to generate new individuals, and determining whether the addition of the influence domains of all street lamps that need to be turned on corresponding to the new individuals can completely cover the demand domain. If not, the new individuals are excluded from the population. If they can completely cover the demand domain, the new individuals are retained and become valid solutions.
[0051] S46, mutate the individuals with a certain probability to create new genetic genes; determine whether the influence domains of all street lamps that need to be turned on corresponding to the new individuals obtained after the mutation can completely cover the demand domain after adding up; if not, exclude the new individuals from the population; if it can completely cover the demand domain, retain the new individuals and make them valid solutions; use all valid solutions as the new generation population;
[0052] S47, determine whether the termination condition is met. If so, stop the algorithm, find the individual with the highest fitness value, and determine the on and off status of the street light according to the binary code of the individual with the highest fitness value; if not, add 1 to the number of iterations and return to step S43; the termination condition refers to the number of iterations reaching the set value.
[0053] When the number of pedestrians on the road is not one, each pedestrian corresponds to a demand domain, and the demand domains of all pedestrians are added together. The overall alternative set is obtained by taking the union of the alternative sets of all pedestrians. Each pedestrian corresponds to the average probability of passing the relevant street lamp. The average probability of all pedestrians passing the relevant street lamp is added together to obtain the overall average probability of passing the relevant street lamp. Repeat steps S41 and S45 to determine the on and off status of the street lamp, wherein the fitness value is determined according to the number of street lamps that need to be turned on and the overall average probability of passing the relevant street lamp.
[0054] When the number of street lights in the alternative set V is 5, k2 is set to 1 / 25. In this way, the first judgment criterion for turning on 5 street lights is 1 / 5, and the first judgment criterion for turning on 4 street lights is 1 / 4. When the second judgment criterion for turning on 5 street lights reaches the maximum and the second judgment criterion for turning on 4 street lights reaches the minimum, the fitness value of turning on 5 street lights is also smaller than the fitness value of turning on four street lights. The maximum fitness value of turning on 5 street lights is k1(1 / 25+1 / 5), which is smaller than the minimum fitness value of turning on 5 street lights, k1×1 / 4. When turning on 5 street lights, the second judgment criterion is used to distinguish the fitness values of individuals.
[0055] The control module reads the switch signal from the database according to the current location of the pedestrian and the location information that the pedestrian has passed, and turns the street light on and off.
[0056] S13, when the image acquisition unit at the next street light determines that the pedestrian has arrived at the next street light, it updates the pedestrian's moving path and the probability of the pedestrian moving along each moving path, and turns the street light on and off based on the updated pedestrian's moving path and the probability of the pedestrian moving along each moving path, and repeats step S13 until the pedestrian arrives at the destination.
[0057] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.
Claims
1. A smart street lamp management method based on artificial intelligence data analysis, characterized in that: The following steps are involved: S11, the image acquisition unit at the street lamp acquires a scene image of the road at the street lamp, performs recognition analysis on the acquired scene image, recognizes pedestrians in the scene image, and if there are pedestrians in the scene image, uploads the location information of the pedestrian and the current street lamp; S12, generating a moving path of the pedestrian and a probability of the pedestrian moving along each moving path according to the current position of the pedestrian and the position information that the pedestrian has passed; Turning street lights on and off based on pedestrians’ movement paths and the probability of pedestrians moving along each movement path; S13, when the image acquisition unit at the next street light determines that the pedestrian has arrived at the next street light, it updates the pedestrian's moving path and the probability of the pedestrian moving along each moving path, and turns the street light on and off based on the updated pedestrian's moving path and the probability of the pedestrian moving along each moving path, and repeats step S13 until the pedestrian arrives at the destination.
2. The intelligent street lamp management method based on artificial intelligence data analysis according to claim 1 is characterized in that: In step S12, the step of generating the pedestrian's moving path and the probability of the pedestrian moving along each moving path according to the pedestrian's current location and the location information that the pedestrian has passed through further includes the following steps: Obtain the historical movement path data of pedestrians in the park, and add the street lamp positions that the current pedestrian has passed to the set U in the order of movement according to the street lamp positions that the current pedestrian has passed; let the elements in the set U be relevant elements, and filter out the movement paths that contain all relevant elements and whose arrangement order is the same as that of the set U from the historical movement path data, and calculate the probability of the selected movement path appearing: path i =n i / n, where path i represents the probability of the i-th selected moving path appearing, n i represents the number of times the i-th filtered moving path appears, and n represents the total number of times all moving paths appear.
3. The intelligent street lamp management method based on artificial intelligence data analysis according to claim 1 is characterized in that: In step S12, the method of turning on and off the street lamp based on the moving paths of pedestrians and the probability of pedestrians moving along each moving path also includes the following steps: S31, generating a pedestrian's field of vision according to the pedestrian's position and moving direction, and adding roads included in the moving path whose probability of appearing in the pedestrian's field of vision is not 0 to the demand domain; S32, testing the lighting function of the street lamp. When the street lamp is turned on and the brightness of the road reaches a specified value under the lighting of the street lamp, the road is added to the influence domain of the street lamp to obtain the influence domain of all street lamps in the park. S33, if the influence domain of the street lamp overlaps with the demand domain, the street lamp is added to the candidate set V, and the probability of a pedestrian passing each street lamp in the candidate set V is calculated: P j =∑path j , where P j represents the probability of a pedestrian passing the jth street light in the set V, path j represents the probability of the moving path containing the jth street light appearing.
4. The intelligent street lamp management method based on artificial intelligence data analysis according to claim 3 is characterized in that: In step S12, the method of turning on and off the street lamp based on the moving paths of pedestrians and the probability of pedestrians moving along each moving path also includes the following steps: S41, binary-encode the streetlights in the candidate set V to obtain individuals, where each binary bit of the individual corresponds to a streetlight in the candidate set V; if the streetlight needs to be turned on, the encoding value of the bit corresponding to the streetlight is 1, and if the streetlight does not need to be turned on, the encoding value of the bit corresponding to the streetlight is 0; S42, obtaining the probability of a pedestrian passing each street lamp in the candidate set V, creating an initial population that meets the constraint condition according to the probability of a pedestrian passing each street lamp in the candidate set V, and setting the number of iterations to 0; the constraint condition is that after the influence domains of all street lamps that need to be turned on are added together, the demand domain can be completely covered; S43, calculate the fitness value of each individual in the population, fit = k1(1 / x + k2×y), where fit represents the fitness value of the individual, x represents the number of street lamps that need to be turned on corresponding to the individual, let the street lamps that need to be turned on corresponding to the individual be the relevant street lamps, y represents the average probability of pedestrians passing the relevant street lamps, let P u represents the probability of a pedestrian passing the uth relevant street light, then y = ∑P u / n u , n u represents the number of relevant street lamps; k1 is the scaling factor, which is set according to the needs; k2 is the restriction coefficient, which is used to compare the pros and cons of individuals who turn on the same number of street lamps. The number of turned-on street lamps is used as the first judgment criterion. The fewer the number of turned-on street lamps, the higher the fitness of the individual; when there are individuals with the same number of turned-on street lamps, the average probability of pedestrians passing the relevant street lamps is used as the second judgment criterion. The higher the average probability of pedestrians passing the relevant street lamps, the higher the fitness of the individual. k2 is set to a constant less than 1 / N(N-1), where N is the number of street lamps in the alternative set V; S44, select individuals in the population according to their fitness values and pass the selected individuals on to the next generation; S45, performing a crossover operation on the individuals selected in step S44, exchanging chromosomes to generate new individuals, and determining whether the addition of the influence domains of all street lamps that need to be turned on corresponding to the new individuals can completely cover the demand domain. If not, the new individuals are excluded from the population. If they can completely cover the demand domain, the new individuals are retained and become valid solutions. S46, mutate the individuals with a certain probability to create new genetic genes; determine whether the influence domains of all street lamps that need to be turned on corresponding to the new individuals obtained after the mutation can completely cover the demand domain after adding up; if not, exclude the new individuals from the population; if it can completely cover the demand domain, retain the new individuals and make them valid solutions; use all valid solutions as the new generation population; S47, determine whether the termination condition is met. If so, stop the algorithm, find the individual with the highest fitness value, and determine the on and off status of the street light according to the binary code of the individual with the highest fitness value; if not, add 1 to the number of iterations and return to step S43; the termination condition refers to the number of iterations reaching the set value.
5. The intelligent street lamp management method based on artificial intelligence data analysis according to claim 4 is characterized in that: In step S12, the following steps are also included: When the number of pedestrians on the road is not one, each pedestrian corresponds to a demand domain, and the demand domains of all pedestrians are added together. The overall alternative set is obtained by taking the union of the alternative sets of all pedestrians. Each pedestrian corresponds to the average probability of passing the relevant street lamp. The average probability of all pedestrians passing the relevant street lamp is added together to obtain the overall average probability of passing the relevant street lamp. Repeat steps S41 and S45 to determine the on and off status of the street lamp, wherein the fitness value is determined according to the number of street lamps that need to be turned on and the overall average probability of passing the relevant street lamp.
6. The intelligent street lamp management method based on artificial intelligence data analysis according to claim 5 is characterized in that: In step S12, the following steps are also included: The historical movement path data of pedestrians in the park is obtained, and the scenes of pedestrians arriving at each street lamp on different movement paths are pre-stored. The on and off states of the street lamps in each scene are pre-determined and stored in the database. When the image acquisition unit at the street lamp recognizes the pedestrian, the position information of the street lamp passed by the pedestrian is obtained, the position information of the pedestrian has passed is obtained, and the on and off states of the street lamps corresponding to the current position information of the pedestrian have passed are obtained from the database to control the street lamps.
7. The intelligent street light management system based on artificial intelligence data analysis is characterized by: include: Data analysis module, control module, data storage module and street light module; The output end of the street light module is connected to the input end of the data storage module and the control module, and is used to identify pedestrians on the road and provide road lighting for pedestrians through street lights; the output end of the data storage module is connected to the input end of the data analysis module, and is used to store the historical movement data of pedestrians in the park; the output end of the data analysis module is connected to the input end of the control module, and the street lights that need to be turned on and off are determined based on the location of the pedestrians and the location information that the pedestrians have passed; the output end of the control module is connected to the input end of the street light module, and is used to generate a control signal for turning the street lights on and off.
8. The intelligent street light management system based on artificial intelligence data analysis according to claim 7 is characterized in that: The street lamp module also includes an image capture unit, a first communication unit and a lighting unit; the image capture unit is used to capture images on the road at the street lamp and identify pedestrians on the road; the communication unit is used to upload the location information of the street lamp and the pedestrian information to the control module and the data storage module; the lighting unit is used to provide lighting for pedestrians on the road.
9. The intelligent street light management system based on artificial intelligence data analysis according to claim 8 is characterized in that: The data analysis module also includes a database, a path analysis unit, an artificial intelligence unit and a second communication unit; the path analysis unit generates the pedestrian's moving path and the probability of the pedestrian moving along each moving path based on the pedestrian's current position and the position information that has been passed; the artificial intelligence unit determines the on and off status of the street lamps through a genetic algorithm, and provides the required lighting services to pedestrians by turning on the least number of street lamps. When the number of turned-on street lamps is the same, the street lamps are turned on according to the average probability of pedestrians passing by the street lamps to provide the required lighting services to pedestrians; the database is used to store the iterative results of the genetic algorithm; and the second communication unit is used to exchange data with the control module.
10. The intelligent street light management system based on artificial intelligence data analysis according to claim 9 is characterized in that: The control module further includes a third communication unit and a control unit; the third communication unit is used to obtain the current position information and the passed position information of the pedestrian from the street lamp module, and obtain the on / off state of the street lamp from the database according to the current position information and the passed position information of the pedestrian; The control unit is used to generate a switch-on and switch-off control signal for the street lamp.