Lamp management method considering pedestrian and wildlife trajectories in local communities

Abstract

The present application belongs to the field of automatic control technology, and particularly relates to a street lamp management method and system considering the trajectories of pedestrians and wild animals in a local community, and a device thereof. The scheme realizes the identification of animals or pedestrians in the community and the tracking of the motion trajectories of the pedestrians through the proximity radar, infrared grating and target tracking radar installed on the street lamp. By using the real-time detected data, the present application predicts the motion direction of the pedestrians and turns on the street lamp on the road in advance. On the other hand, the present application predicts the future motion trajectory of the wild animals according to the historical data of the trajectories of the wild animals and the current appearing position, and then judges whether the pedestrians and the wild animals will meet, and turns on the street lamp at the position where the two are likely to meet in advance, so as to realize the safety warning. The scheme of the present application can dynamically adjust the on-off state of the street lamp according to the pedestrians and the animals in the community, and thus effectively reduces the risk and harm encountered by the pedestrians and the animals.

Description

Technical Field

[0001] The present invention belongs to the field of automatic control technology, and in particular relates to a streetlight management method, system and device that take into account the trajectories of pedestrians and wild animals in a local community. Background Art

[0002] As the environment improves, wild animals, or pets kept or abandoned by residents, are increasingly appearing in large residential areas, parks, or industrial parks. When pedestrians encounter certain wild animals (such as wild boars), the animals can pose a direct threat to pedestrians' safety, especially if they feel threatened or frightened. If the number of wild animals in a particular area is high, this can even disrupt residents' normal activities. Especially in the evening or at night, the sudden appearance of wild animals (such as deer and wild boars) in human communities can lead to collisions with pedestrians or vehicles, posing a threat to residents' lives and property. Some animals (such as stray dogs or mother leopards) may attack pedestrians and cause injury when defending their territory or their cubs. Pedestrians passing through areas with wild animals, coming into contact with animal feces, or being bitten, can increase their risk of contracting diseases (such as rabies). Feeling anxious or averse to the presence of wild animals can lead to stress when outdoors, reducing their quality of life.

[0003] Regarding encounters between wildlife and humans, if the movement paths, timing, and frequency of aggressive animals could be predicted promptly and users could be informed of the risks, this would provide sufficient reaction time to avoid the threat. However, existing technologies lack effective solutions for tracking or predicting wildlife movements. Furthermore, at night, streetlights are often dim, limiting pedestrian visibility. This can prevent pedestrians from spotting distant wildlife, shortening their reaction time in the event of an encounter. Summary of the Invention

[0004] In order to solve the problem that the existing technology lacks means to effectively reduce the risks and hazards encountered by pedestrians and animals, the present invention provides a street light management method, system and device that take into account the trajectories of pedestrians and wild animals in a local community.

[0005] The technical solution provided by the present invention is:

[0006] A streetlight management method that considers the trajectories of pedestrians and wild animals in a local community includes the following steps:

[0007] S1: Build a geographic image based on the spatial distribution of roads within the community; and create a connected graph G representing the animal movement paths within the community, using each streetlight as a node and the lines between streetlights as edges.

[0008] S2: Use proximity radar installed on street lamps to detect living targets in the community in real time.

[0009] S3: When the distance of any approaching radar to a living target is less than a preset threshold, the type of living creature is identified by the infrared grating on the corresponding street lamp and the following decision is made:

[0010] (1) If the living creature is a person, the target tracking radar on the corresponding street light is turned on. The target tracking radar detects the spatial position of the target person on the geographic image and determines the pedestrian's movement direction and speed based on the pedestrian's spatial position changes.

[0011] The pedestrian's future movement trajectory is predicted based on the pedestrian's movement direction and speed, and the street lights in the direction of the pedestrian's movement are lit in advance. After the pedestrian passes, the street lights are restored to a low-power state according to a preset delay.

[0012] (2) If the living organism is a wild animal, the time and order of the wild animal's appearance at the nodes corresponding to each street light are recorded, and each node is connected in order to obtain a wild animal's appearance trajectory.

[0013] Based on the historical data of wildlife trajectories within the community, the connected graph G is transformed into a directed graph G′. The edge weights of the directed graph G′ are updated based on the newly added wildlife trajectories, and the average speed of the wildlife on each edge in the directed graph is recorded.

[0014] Among them, edge weights are used to represent the probability of wild animals appearing.

[0015] S4: When there are pedestrians in the community, if any node detects the presence of wild animals, the movement trajectory of the wild animals is predicted based on the directed graph G′; then it is determined whether the animals and pedestrians will encounter each other, and street lights are lit at the locations where wild animals and pedestrians may encounter each other to provide early warning.

[0016] As a further improvement of the present invention, in step S1, the data format of the connectivity graph is: G = (V, E);

[0017] Among them, V is the node v corresponding to each street light i The set of nodes, i = 1…n, n represents the number of street lights in the community. E is the set of edges between nodes, any node v i and v j The edge between them is denoted as e ij , j=1…n.

[0018] As a further improvement to the present invention, the streetlights used in the community have three operating modes: off, low power, and on. During preset daytime hours, the streetlights are in off mode. During preset nighttime busy hours, the streetlights are in on mode. During preset nighttime idle hours, the streetlights are initially in low power mode and switch to on mode upon command.

[0019] and / or

[0020] There are also multiple cameras installed in the community. When there are pedestrians in the community, the camera's pan / tilt is adjusted according to the detected spatial position, movement direction and movement speed of the pedestrians to achieve real-time tracking of the target person.

[0021] As a further improvement of the present invention, in step S3, the infrared grating includes multiple sensing points evenly arranged at different heights on the street lamp pole in the vertical direction; the infrared grating determines the type of the approaching living target based on the spatial distribution of the sensing signal in the vertical direction.

[0022] As a further improvement of the present invention, the process of converting the connected graph G into a directed graph G′ based on the historical data of wild animal tracks in the community is as follows:

[0023] (i) The direction of the line from the node at the previous moment to the node at the next moment in the wild animal's trajectory is taken as the direction of the animal's movement between the two nodes.

[0024] (ii) Based on the historical data of wildlife trajectories, count the number of times wildlife appears in a specified direction between any two nodes in the connected graph G.

[0025] (iii) Calculate the probability P of a wild animal appearing in a specified direction between any two nodes based on the number of appearances ij , and the probability P will appear ij as the edge weight of the corresponding edge in the directed graph G′;

[0026]

[0027] In the above formula, N i Indicates that wild animals appearing in historical data are from node v i The number of times it departs and arrives at any neighboring node; N ij Indicates that wild animals appearing in historical data are from node v i and depart to neighbor node v j times.

[0028] (iv) The direction of the maximum edge weight starting from the specified node is taken as the high-frequency direction of the current node, and the high-frequency direction of each node is recorded.

[0029] (v) Calculate the average time it takes for wild animals to pass through any two nodes based on the historical data of wild animal trajectories, and then calculate the average speed of wild animals passing through any edge based on the distance between nodes.

[0030] As a further improvement of the present invention, in step S4, the method for predicting the movement trajectory of wild animals based on the directed graph G′ is as follows:

[0031] S041: The location where the wild animal is first detected is used as the initial location of the movement trajectory.

[0032] S042, taking the initial position as the starting point, taking the high-frequency direction of the current node in the directed graph as the movement direction, and the average speed of the edge in the high-frequency direction as the movement speed of the wild animal, predicting the position and time of the wild animal arriving at the next node.

[0033] S043, taking the next node predicted each time as the starting point, repeat step S042 to determine the movement direction and movement speed of the wild animal at the current node, and obtain the end point of the current path.

[0034] S044: When the predicted next node is a node on the edge of the directed graph, the lines from the initial position to all the predicted nodes are used as the predicted movement trajectory of the wild animal.

[0035] As a further improvement to the present invention, the historical dataset of wildlife trajectories collected is divided into multiple time periods based on their appearance times. A corresponding directed graph G′ is then created for each of these time periods. The directed graph G′ for each time period is then combined to predict encounters between pedestrians and wildlife.

[0036] As a further improvement of the present invention, in step S4, the method for determining whether a pedestrian and a wild animal will encounter each other is as follows:

[0037] First, the spatial characteristics of the predicted movement trajectories of pedestrians and wild animals are analyzed to calculate whether the closest position between the two is less than the preset safe space distance: if so, it is considered that the two may meet, otherwise it is considered that the two will not meet.

[0038] Secondly, when pedestrians and wild animals are likely to meet, the time characteristics of the predicted movement trajectories of pedestrians and wild animals are further analyzed to determine whether the corresponding distance intervals between the two that are less than the safe distance overlap in the time domain: if so, it is considered that the two will meet; otherwise, it is considered that the two will not meet.

[0039] The present invention also includes a streetlight management system that utilizes the aforementioned streetlight management method that considers the trajectories of pedestrians and wildlife within a local community to adaptively adjust the operating mode of each streetlight within the community. The streetlight management system includes multiple proximity radars, multiple infrared barriers, multiple target tracking radars, at least one image acquisition device, and a backend server.

[0040] Proximity radars are installed on various streetlights within the community and are used to detect living targets around them. Infrared gratings are installed on various streetlights within the community and are used to distinguish the types of living targets detected by the proximity radars. Target tracking radars are installed on streetlights within the community and are used to track the positions of pedestrians identified by the infrared gratings. Image acquisition equipment includes cameras and a pan / tilt head (PTZ) to capture real-time image data of pedestrians in the community.

[0041] The backend server is electrically connected to the controllers of each proximity radar, infrared grating, target tracking radar, image acquisition equipment, and each street lamp in the community. The backend server is used to: (1) obtain the real-time detection signal of the proximity radar to determine whether a living target appears in the community. (2) turn on the infrared grating and distinguish the type of living target according to the detection signal of the infrared grating. (3) turn on the target tracking radar and image acquisition equipment to track pedestrians in the community and predict their movement trajectory and movement speed; and switch the street lamps on the pedestrian path to the lighting mode. (4) record the time and order of wild animals appearing at the nodes corresponding to each street lamp, and connect each node in order to obtain a wild animal appearance trajectory. (5) dynamically update the directed graph G′ in the community according to the newly added wild animal appearance trajectory. (6) predict the movement trajectory and movement speed of wild animals appearing in the community based on the directed graph G′. (7) judge whether the movement trajectory of pedestrians and wild animals will meet based on the movement trajectory of the two, and light up the street lamps at the corresponding positions if they do.

[0042] The present invention also includes a streetlight management device comprising a memory, a processor, and a computer program stored in the memory and executed by the processor. When the processor executes the computer program, the computer program employs the aforementioned streetlight management method that considers the trajectories of pedestrians and wildlife in a local community, thereby adaptively adjusting the operating mode of each streetlight within the community based on the trajectories of pedestrians and wildlife within the community.

[0043] The technical solution provided by the present invention has the following beneficial effects:

[0044] The present invention designs a streetlight management method that takes into account the trajectories of pedestrians and wild animals in a local community. The method uses a living presence-sensing radar and an infrared grating installed on each streetlight to form a scanning device for detecting humans or animals. The method also records the frequency, activity trajectories, movement speed, and time patterns of large, medium, and small animals in the community, and then generates a directed graph that can predict the movement trajectories of animals in the community.

[0045] After identifying a user through an infrared grating, the solution of the present invention also uses a target tracking radar to monitor the user's movement speed and spatial position in real time. It then uses geographic imagery within the community to predict the user's movement trajectory and adjust the operating state of the streetlights to provide immediate lighting. The solution also combines the predicted movement trajectories of animals and pedestrians to determine whether the two will intersect. If there is a risk of an encounter, the streetlight at the corresponding location is illuminated to provide a warning to the user.

[0046] The streetlight management method provided by this invention not only reduces streetlight operating power consumption but also effectively mitigates the risks and hazards encountered by pedestrians and animals. Compared to traditional sensing devices that completely traverse every space, this solution offers higher response efficiency, enabling the user to activate the streetlight at the current location and, within a valid timeframe, activate sensing devices within a range of N streetlights centered around the user to provide forecasts and warnings. This solution is particularly suitable for use in communities where wild animals and pets are frequently present, demonstrating its high practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flowchart of the steps of the street light management method that takes into account the trajectories of pedestrians and wild animals in a local community, provided in Example 1 of the present invention.

[0048] Figure 2 This is the connectivity graph established according to the node corresponding to each street lamp in the community in Example 1 of the present invention.

[0049] Figure 3 This is a schematic diagram of the principle of using infrared grating to identify the type of target in Example 1 of the present invention.

[0050] Figure 4 This is a flowchart of converting a connected graph into a directed graph based on historical data of wild animal appearance trajectories in a community in Example 1 of the present invention.

[0051] Figure 5 This is a flow chart of the steps for predicting the movement trajectory of wild animals based on a directed graph in Example 1 of the present invention.

[0052] Figure 6 This is a flow chart of the steps for determining whether pedestrians and wild animals will encounter each other in Example 1 of the present invention.

[0053] Figure 7This is an architectural diagram of the street light management system provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0055] Example 1

[0056] In view of the fact that wild animals may enter human communities at night and thus endanger the safety of life and property of community residents, this embodiment provides a street light management method that takes into account the trajectories of pedestrians and wild animals in local communities. The idea of ​​the technical solution provided by this embodiment is: through the proximity radar (living presence sensing radar), infrared grating and target tracking radar installed on the street lights, it is possible to identify animals or pedestrians in the community and track the movement trajectory of pedestrians. Using the data detected in real time, on the one hand, the direction of movement of pedestrians can be predicted, and the street lights on the pedestrians' path can be "lit up" in advance. This not only ensures the lighting function of the street lights but also reduces the lighting power of the street lights when necessary, achieving energy conservation and emission reduction. On the other hand, the future movement trajectory of wild animals can be predicted based on the historical data of the wild animal's appearance trajectory and the current appearance location, and then when pedestrians and wild animals are likely to meet, the street lights ahead can be lit in advance, thereby achieving safety warnings.

[0057] In the solution of this embodiment, the street lights used in the community have three working modes: off, low power consumption, and on. Among them, during the preset daytime period, the street lights are in the off mode. During the preset nighttime busy period, the street lights are in the on mode. During the preset nighttime idle period, the street lights are in the low power consumption mode in the initial state, and are switched to the on mode according to instructions. Among them, both the low power consumption mode and the on mode have lighting effects, the difference is that the brightness of the street lights in the low power consumption mode is relatively dim, while the street lights in the on mode have higher power and are brighter. This embodiment switches the street lights to the low power consumption mode when there are fewer pedestrians at night, and then switches the street lights at the corresponding positions to the on mode when pedestrians appear, or when there is a safety risk of encounter between pedestrians and animals, so as to achieve the effect of saving energy.

[0058] Specifically, if Figure 1 As shown, the streetlight management method provided in this embodiment that takes into account the trajectories of pedestrians and wild animals in a local community includes the following steps:

[0059] S1: Build a geographic image based on the spatial distribution of roads within the community; and create a connected graph G representing the animal movement paths within the community, using each streetlight as a node and the lines between streetlights as edges.

[0060] In the scheme of the embodiment, the established geographic image is mainly used to realize the prediction of the motion trajectory of the user. Generally, the user will only move along the road and the open square or lawn within the community. Therefore, the motion trajectory of the user can be predicted by combining the real-time motion information of the pedestrian and the geographic image information containing the elements such as the road, the building and the terrain.

[0061] Unlike the pedestrian, the wild animal will not move along the existing road when moving, and it will often pass through the hedge or the fence, etc. Therefore, the motion trajectory prediction method of the pedestrian is not applicable to the wild animal. In the scheme of the embodiment, it is considered that although the wild animal will not move along the road, the trajectory of the wild animal appearing in the community still has statistical characteristics, for example, when most animals appear in an area, they will mostly appear along a fixed route or several routes. Therefore, the embodiment establishes a connected graph taking the street lamp as the node, and combines the historical data of the appearing trajectory of the wild animal to weight the connected graph, converts the connected graph into a directed graph, and then represents the statistical characteristics of the wild animal appearing in the community by using the directed graph. Finally, the directed graph is used to assist in completing the prediction of the motion trajectory of the wild animal.

[0062] Figure 2 For a connected graph in a typical community, the circular dots in the graph are the nodes in the connected graph, and the connection lines between the nodes are the edges of the connected graph. In the embodiment, the data format of the connected graph established according to the spatial positions of the street lamps in the community is G=(V, E). Wherein, V is a set of nodes v i corresponding to the street lamps in the community, i=1…n, and n represents the number of street lamps in the community. E is a set of edges between the nodes, and the edge between any node v i and v j is denoted as e ij , j=1…n.

[0063] S2: Real-time detection of living targets appearing in the community by the proximity radar installed on the street lamp.

[0064] In the embodiment, the proximity radar can be used to detect whether there is a living target near the installation position, and the embodiment installs a proximity radar on each street lamp in the community. The detection range of each proximity radar basically covers all positions in the community. In other embodiments, in order to further expand the detection range of the proximity radar and eliminate the dead angle, the proximity radar can also be installed in other areas where the wild animals appear frequently and no street lamp is set.

[0065] S3: When any proximity radar detects that the distance of the living target is less than a preset threshold, the type of the living organism is distinguished by the infrared light grid on the corresponding street lamp.

[0066] As Figure 3As shown, the infrared grating in this embodiment includes multiple sensing points evenly arranged in the vertical direction at different heights on the streetlight pole; the infrared grating determines the type of approaching living targets based on the spatial distribution of the sensing signals in the vertical direction. For example, in a typical solution, more than five sensors can be set up within a height range corresponding to human height. For most pedestrians, they can usually trigger all or some of the infrared sensing points. For most wild animals, however, they can only trigger one or several sensing points located at lower positions. The server can distinguish the type of living organism based on the number and / or height information of the sensing signals in the infrared grating triggered by each approaching target.

[0067] Based on the identified target type, this embodiment makes the following decisions:

[0068] (1) If the living creature is a person, the target tracking radar on the corresponding street light is turned on. The target tracking radar detects the spatial position of the target person on the geographic image and determines the pedestrian's movement direction and speed based on the pedestrian's spatial position changes.

[0069] In the actual application of the solution of this embodiment, the least squares method can be used to linearly fit the pedestrian's movement path and calculate the movement speed based on the various spatial position points collected during the pedestrian's movement. The process is as follows:

[0070] The server constructs a grid with due north as the Y axis, due east as the X axis, and (0, 0) as the origin (Note: The maximum monitoring radius of the radar is d. The coordinate points within the set time t are marked, and the obtained coordinate points and direction frequencies are substituted into the following matrix model:

[0071] y=θ0+θ1x1+θ2x2

[0072] Where y represents the ordinate of the target being tracked, ω0 is the function intercept, x1 is the current abscissa, and x2 is the direction frequency. If the current direction coincides with the direction of the current position with the highest frequency, then x2 is given so that the difference between y2 and the nearest y1 and y3 is the same. θ1 and θ2 are the regression coefficients to be measured. After obtaining multiple sets of y0, they are expressed as the following matrix:

[0073]

[0074] in, represents the predicted value of y; θ is the coefficient to be estimated, and X represents the coordinate matrix.

[0075] Adding a column matrix of 1 to θ0 can be expressed as the following matrix form:

[0076]

[0077] The least squares estimate θ is calculated by the residual sum of squares SSE (Sum of Squares for Error), and the calculation formula is as follows:

[0078]

[0079] Here, ε represents the residual.

[0080] The residual sum of squares formula is further expressed as the following matrix form:

[0081] SSE=(y-Xθ) T (y-Xθ)

[0082] Find the minimum value of SSE, that is, the first-order derivative after taking the partial derivative of θ is 0, and find the minimum estimate:

[0083] θ=(X T X) -1 X T Y

[0084] Where Y represents the coordinate matrix, and using the matrix derivation rule we can get:

[0085]

[0086]

[0087] Take the value of θ that gives a derivative of 0:

[0088]

[0089] The minimum estimated value θ can be obtained:

[0090] θ=(X T X) -1 X T Y

[0091] Post-analysis directly into Obtain the linear correlation function D(x). The slope of the D(x) function is the target travel direction A. Compare this with the slope of the line connecting the initial and final values ​​of the recorded position coordinates. If the difference is significant, use the slope of D(x).

[0092] At the same time, we can use the polyfit(x,y,n) function in Matlab to perform linear fitting and get F(x). Then we can get the speed of the organism based on time and perform linear integration on L:

[0093] L=μ L F(x)ds

[0094] The distance L is obtained, and L / t gives a velocity value V. This velocity is then compared with the velocity measured by the radar itself. If the error is large, a warning alert is sent to the server, indicating that troubleshooting is required. Otherwise, the calculated value is returned. The resulting fitted function is the animal's movement path, from which the velocity V and path L can be derived.

[0095] The pedestrian's future movement trajectory is predicted based on the pedestrian's movement direction and speed, and the street lights in the direction of the pedestrian's movement are lit in advance. After the pedestrian passes, the street lights are restored to a low-power state according to a preset delay.

[0096] (2) If the living organism is a wild animal, the time and order of the wild animal's appearance at the nodes corresponding to each street light are recorded, and each node is connected in order to obtain a wild animal's appearance trajectory.

[0097] Based on historical data on wildlife sightings within a community, the connected graph G is transformed into a directed graph G′. Edge weights in G′ are updated based on newly added wildlife sightings, and the average speed of wildlife along each edge in the directed graph is recorded. Edge weights represent the probability of wildlife appearance.

[0098] like Figure 4 As shown, the process of converting the connected graph G into the directed graph G′ based on the historical data of wild animal trajectories in the community is as follows:

[0099] (i) The direction of the line from the node at the previous moment to the node at the next moment in the wild animal's trajectory is taken as the direction of the animal's movement between the two nodes.

[0100] (ii) Based on the historical data of wildlife trajectories, count the number of times wildlife appears in a specified direction between any two nodes in the connected graph G.

[0101] (iii) Calculate the probability P of a wild animal appearing in a specified direction between any two nodes based on the number of appearances ij , and the probability P will appear ij as the edge weight of the corresponding edge in the directed graph G′;

[0102]

[0103] In the above formula, N i Indicates that wild animals appearing in historical data are from node v i The number of times it departs and arrives at any neighboring node; N ij Indicates that wild animals appearing in historical data are from node v i and depart to neighbor node v j times.

[0104] (iv) The direction of the maximum edge weight starting from the specified node is taken as the high-frequency direction of the current node, and the high-frequency direction of each node is recorded.

[0105] (v) Calculate the average time it takes for wild animals to pass through any two nodes based on the historical data of wild animal trajectories, and then calculate the average speed of wild animals passing through any edge based on the distance between nodes.

[0106] S4: When there are pedestrians in the community, if any node detects the presence of wild animals, the movement trajectory of the wild animals is predicted based on the directed graph G′; then it is determined whether the animals and pedestrians will encounter each other, and street lights are lit at the locations where wild animals and pedestrians may encounter each other to provide early warning.

[0107] In this embodiment, if Figure 5 As shown in Figure 2, the method for predicting the movement trajectory of wild animals based on the directed graph G′ is as follows:

[0108] S041: The location where the wild animal is first detected is used as the initial location of the movement trajectory.

[0109] S042, taking the initial position as the starting point, taking the high-frequency direction of the current node in the directed graph as the movement direction, and the average speed of the edge in the high-frequency direction as the movement speed of the wild animal, predicting the position and time of the wild animal arriving at the next node.

[0110] S043, taking the next node predicted each time as the starting point, repeat step S042 to determine the movement direction and movement speed of the wild animal at the current node, and obtain the end point of the current path.

[0111] S044: When the predicted next node is a node on the edge of the directed graph, the lines from the initial position to all the predicted nodes are used as the predicted movement trajectory of the wild animal.

[0112] In the solution of this embodiment, Figure 6 As shown, the method for determining whether pedestrians and wild animals will encounter each other is as follows:

[0113] First, the spatial characteristics of the predicted movement trajectories of pedestrians and wild animals are analyzed to calculate whether the closest position between the two is less than the preset safe space distance: if so, it is considered that the two may meet, otherwise it is considered that the two will not meet.

[0114] Secondly, when pedestrians and wild animals are likely to meet, the time characteristics of the predicted movement trajectories of pedestrians and wild animals are further analyzed to determine whether the corresponding distance intervals between the two that are less than the safe distance overlap in the time domain: if so, it is considered that the two will meet; otherwise, it is considered that the two will not meet.

[0115] In a further optimization of this embodiment, multiple cameras are installed within the community. When pedestrians are detected within the community, the server adjusts the camera's pan / tilt function based on the detected pedestrian's spatial position, direction of movement, and speed to enable real-time tracking of the target person. The cameras in this embodiment are designed to record video footage of pedestrians as they appear within the community, providing visual evidence for potential damage.

[0116] In other embodiments, technicians can also divide the collected historical data sets of wildlife trajectories into multiple time periods based on the time of appearance, and create corresponding directed graphs G′ based on the data sets of different time periods. Then, the directed graphs G′ of the corresponding time periods can be combined to predict whether pedestrians and wildlife will encounter each other.

[0117] For example, technicians divided the trajectories of wild animals into three categories: 06:00-20:00, 20:00-24:00, and 24:00-06:00, and then established directed graphs corresponding to the three time periods. Because the types and movement routes of wild animals appearing in different time periods may vary, the high-frequency directions and movement speeds of wild animals corresponding to the same node in the directed graphs of the three time periods may also vary. On this basis, when wild animals are detected again within a specified time period, the directed graph of the corresponding time period can be queried, and the movement trajectories of wild animals at future moments can be predicted based on this directed graph. Compared with the previous solution, the optimized directed graph is more accurate in predicting the movement trajectories of wild animals.

[0118] Furthermore, the solution of this embodiment can also record the body size data of each wild animal's movement trajectory, and then create corresponding directed graphs for wild animals of different body sizes. In subsequent applications, when a new wild animal appears in the community, the corresponding directed graph is queried based on the animal's body size, and the animal's movement trajectory at a certain moment in time is predicted.

[0119] Example 2

[0120] Based on the solution of Example 1, this embodiment provides a streetlight management system, which is used to adopt the streetlight management method of Example 1 that takes into account the trajectories of pedestrians and wild animals in a local community to achieve adaptive adjustment of the working mode of each streetlight in the community. Figure 7 As shown, the street light management system includes: multiple proximity radars, multiple infrared gratings, multiple target tracking radars, at least one image acquisition device and a background server.

[0121] Proximity radars are installed on various streetlights within the community and are used to detect living targets around them. Infrared gratings are installed on various streetlights within the community and are used to distinguish the types of living targets detected by the proximity radars. Target tracking radars are installed on streetlights within the community and are used to track the positions of pedestrians identified by the infrared gratings. Image acquisition equipment includes cameras and a pan / tilt head (PTZ) to capture real-time image data of pedestrians in the community.

[0122] The backend server is electrically connected to the controllers of each proximity radar, infrared grating, target tracking radar, image acquisition equipment, and each street lamp in the community. The backend server receives signals from various detection devices and dynamically adjusts the working status of the street lamp according to the received signals, including: (1) obtaining the real-time detection signal of the proximity radar to determine whether a living target appears in the community. (2) turning on the infrared grating when a living target appears again, and distinguishing the type of the living target based on the detection signal of the infrared grating. (3) when the living target is identified as a pedestrian, turning on the target tracking radar and image acquisition equipment to track the pedestrian in the community and predict its movement trajectory and movement speed; and switching the street lamps on the pedestrian path to the lighting mode. (4) recording the time and order of the appearance of wild animals at the nodes corresponding to each street lamp, and connecting each node in order to obtain a wild animal appearance trajectory. (5) dynamically updating the directed graph G′ in the community based on the newly added wild animal appearance trajectory. (6) combining the directed graph G′ to predict the movement trajectory and movement speed of wild animals appearing in the community. (7) Based on the movement trajectories of pedestrians and wild animals, determine whether the two will meet. If so, light up the street lights at the corresponding positions.

[0123] In practical applications, the streetlight management system provided by this embodiment can not only autonomously control the operating status of streetlights but also proactively send users warnings about wildlife. For example, users can log in to the corresponding app to obtain their real-time location, which is pushed to them by the backend server, for navigation purposes. They can also receive timely warnings about wildlife intrusions.

[0124] Example 3

[0125] Based on Examples 1 and 2, this embodiment further provides a streetlight management device, which includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it adopts the streetlight management method that considers the trajectories of pedestrians and wild animals in a local community as described in Example 1, and implements adaptive adjustment of the operating mode of each streetlight within the community based on the trajectories of pedestrians and wild animals within the community. The streetlight management device in this embodiment is the backend server in the streetlight management system provided in Example 2.

[0126] The streetlight management device provided in this embodiment is essentially a computer device used to implement the solution in Example 1. The computer device can be a program-executable smart terminal, tablet computer, laptop computer, desktop computer, rack server, blade server, tower server, or cabinet server (including a standalone server or a server cluster consisting of multiple servers).

[0127] The computer device in this embodiment includes at least, but is not limited to, a memory and a processor that can be interconnected via a system bus. The memory (i.e., a readable storage medium) includes flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic storage, a magnetic disk, an optical disk, and the like. In some embodiments, the memory can be an internal storage unit of the computer device, such as the hard disk or internal memory of the computer device. In other embodiments, the memory can also be an external storage device of the computer device, such as a plug-in hard disk equipped with the computer device, a smart media card (SMC), a secure digital (SD) card, a flash memory card, and the like. Of course, the memory can also include both the internal storage unit of the computer device and its external storage devices. In this embodiment, the memory is generally used to store the operating system and various application software installed on the computer device. In addition, the memory can also be used to temporarily store various types of data that have been output or are about to be output.

[0128] In some embodiments, the processor may be a central processing unit (CPU), a graphics processing unit (GPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor is generally used to control the overall operation of the computer device. In this embodiment, the processor is used to run program code stored in the memory or process data.

[0129] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the patent for this invention shall be determined by the appended claims.

Claims

1. A streetlight management method that considers the trajectories of pedestrians and wild animals in a local community, characterized by: It includes the following steps: S1: Build a geographic image based on the spatial distribution of roads within the community; and create a connected graph G representing the animal movement paths within the community, using each streetlight as a node and the lines between streetlights as edges; S2: Use proximity radar installed on street lamps to detect living targets in the community in real time; S3: When the distance of any proximity radar to a living target is less than a preset threshold, the type of living creature is identified by the infrared grating on the corresponding street lamp and the following decision is made: (1) If the living being is a person, the target tracking radar on the corresponding street lamp is turned on; the spatial position of the target person on the geographic image is detected by the target tracking radar, and the direction and speed of movement of the pedestrian are determined based on the change in the spatial position of the pedestrian; The system predicts the pedestrian's future trajectory based on the pedestrian's direction and speed, and then lights up the streetlights in the direction of the pedestrian's movement in advance. After the pedestrian passes, the streetlights are restored to a low-power state for a preset period of time. (2) If the living organism is a wild animal, the time and order of the wild animal's appearance at the nodes corresponding to each street light are recorded, and each node is connected in order to obtain a wild animal's appearance trajectory; The connected graph G is converted into a directed graph G′ based on the historical data of wildlife trajectories in the community; the edge weights of the directed graph G′ are updated based on the newly added wildlife trajectories, and the average speed of the wildlife on each edge in the directed graph is recorded; Wherein, the edge weight is used to represent the probability of occurrence of wild animals; S4: When there are pedestrians in the community, if any node detects the presence of a wild animal, the movement trajectory of the wild animal is predicted based on the directed graph G′; then, it is determined whether the animal and the pedestrian will encounter each other, and street lights at the location where the wild animal and the pedestrian may encounter are lit to provide an early warning.

2. The streetlight management method according to claim 1, wherein: In step S1, the data format of the connectivity graph is: G = (V, E); Among them, V is the node v corresponding to each street light i The set of nodes, i = 1…n, n represents the number of street lights in the community; E is the set of edges between nodes, any node v i and v j The edge between them is denoted as e ij , j=1…n.

3. The streetlight management method according to claim 1, wherein: The streetlights used in the community have three operating modes: off, low power consumption, and on. During the preset daytime hours, the streetlights are in off mode; during the preset nighttime busy hours, the streetlights are in on mode; during the preset nighttime idle hours, the streetlights are initially in low power consumption mode and switch to on mode upon command. and / or There are also multiple cameras installed in the community. When there are pedestrians in the community, the camera's pan / tilt is adjusted according to the detected spatial position, movement direction and movement speed of the pedestrians to achieve real-time tracking of the target person.

4. The streetlight management method according to claim 1, wherein: In step S3, the infrared grating includes a plurality of sensing points evenly arranged in the vertical direction at different heights on the street lamp pole; the infrared grating determines the type of the approaching living target based on the spatial distribution of the sensing signals in the vertical direction.

5. The streetlight management method according to claim 2, wherein: The process of converting the connected graph G into a directed graph G′ based on the historical data of wild animal trajectories in the community is as follows: (i) The direction of the line connecting the node at the earlier moment to the node at the later moment in the wild animal's trajectory is taken as the direction of movement of the animal between the two nodes; (ii) Based on the historical data of wildlife trajectories, count the number of appearances of wildlife in a specified direction between any two nodes in the connected graph G; (iii) Calculate the probability P of a wild animal appearing in a specified direction between any two nodes based on the number of appearances ij , and the probability P will appear ij as the edge weight of the corresponding edge in the directed graph G′; In the above formula, N i Indicates that wild animals appearing in historical data are from node v i The number of times it departs and arrives at any neighboring node; N ij Indicates that wild animals appearing in historical data are from node v i and depart to neighbor node v j the number of times; (iv) The direction of the maximum edge weight starting from the specified node is taken as the high-frequency direction of the current node, and the high-frequency direction of each node is recorded; (v) Calculate the average time it takes for wild animals to pass through any two nodes based on the historical data of wild animal trajectories, and then calculate the average speed of wild animals passing through any edge based on the distance between nodes.

6. The streetlight management method according to claim 1, wherein: In step S4, the method for predicting the movement trajectory of wild animals based on the directed graph G′ is as follows: S041: The location where the wild animal is first detected is used as the initial location of the movement trajectory; S042, starting from the initial position, taking the high-frequency direction of the current node in the directed graph as the movement direction, and the average speed of the edges passing through the high-frequency direction as the movement speed of the wild animal, to predict the position and time of the wild animal's arrival at the next node; S043, taking the next node predicted each time as the starting point, repeating step S042 to determine the movement direction and movement speed of the wild animal at the current node, and obtaining the end point of the current path; S044: When the predicted next node is a node on the edge of the directed graph, the lines from the initial position to all the predicted nodes are used as the predicted movement trajectory of the wild animal.

7. The streetlight management method according to claim 1, wherein: The collected historical dataset of wildlife trajectories is divided into multiple time periods according to the time of appearance, and corresponding directed graphs G′ are established based on the datasets of different time periods. Then, the directed graphs G′ of the corresponding time periods are combined to predict whether pedestrians and wild animals will encounter each other.

8. The streetlight management method considering the trajectories of pedestrians and wild animals in a local community as claimed in claim 1, characterized in that: In step S4, the method for determining whether a pedestrian and a wild animal will encounter each other is as follows: First, the spatial characteristics of the predicted movement trajectories of pedestrians and wild animals are analyzed to calculate whether the closest position between the two is less than the preset safe space distance. If so, it is considered that the two may meet; otherwise, it is considered that the two will not meet. Secondly, when pedestrians and wild animals are likely to meet, the time characteristics of the predicted movement trajectories of pedestrians and wild animals are continued to determine whether the corresponding distance intervals between the two that are less than the safe distance overlap in the time domain: if so, it is considered that the two will meet; otherwise, it is considered that the two will not meet.

9. A streetlight management system, characterized by: It is used to adopt the streetlight management method considering the trajectories of pedestrians and wild animals in a local community as described in any one of claims 1 to 8 to achieve adaptive adjustment of the working mode of each streetlight within the community; The street light management system includes: Multiple proximity radars are installed on various streetlights in the community and are used to detect living targets around the streetlights; A plurality of infrared gratings, which are respectively installed on various street lamps in the community and are used to distinguish the types of living targets detected by the proximity radar; Multiple target tracking radars, which are installed on street lamps in the community and are used to track the positions of pedestrians identified by the infrared grating; At least one image acquisition device, comprising a camera and a pan / tilt head, the image acquisition device being used to acquire real-time image data of pedestrians appearing in the community; A backend server is electrically connected to the controllers of each of the proximity radars, infrared gratings, target tracking radars, image acquisition equipment, and each streetlight in the community; the backend server is used to: (1) obtain the real-time detection signal of the proximity radar to determine whether a living target appears in the community; (2) turn on the infrared grating and distinguish the type of the living target according to the detection signal of the infrared grating; (3) turn on the target tracking radar and image acquisition equipment to track pedestrians in the community and predict their movement trajectory and movement speed, and switch the streetlights on the pedestrian path to the lighting mode; (4) record the time and order of wild animals appearing at the nodes corresponding to each streetlight, and connect each node in order to obtain a wild animal appearance trajectory; (5) dynamically update the directed graph G′ in the community according to the newly added wild animal appearance trajectory; (6) predict the movement trajectory and movement speed of wild animals appearing in the community in combination with the directed graph G′; (7) judge whether the movement trajectory of pedestrians and the movement trajectory of wild animals will meet based on the movement trajectory of the pedestrians and the movement trajectory of wild animals, and if so, light up the streetlights at the corresponding positions.

10. A streetlight management device comprising a memory, a processor, and a computer program stored in the memory and running in the processor, characterized in that: When the processor executes the computer program, it adopts the street light management method that takes into account the trajectories of pedestrians and wild animals in the local community as described in any one of claims 1-8, so as to adaptively adjust the working mode of each street light within the community according to the trajectories of pedestrians and wild animals in the community.

Images