Garbage point location identification method and device

Abstract

The present invention provides a method and device for identifying the location of garbage points. By recognizing the behavior that the collection and transportation starts to decelerate at time point t1, reaches a low speed at time point t2, and then accelerates at time point t3, where t1 < t2 < t3, the present invention can completely capture the short-term停留 events of sanitation vehicles, more accurately learn the locations of garbage points, and effectively improve the management efficiency of urban garbage collection and transportation.

Description

Technical Field

[0001] The invention relates to a method and a device for identifying a garbage point location. Background Art

[0002] With the development of the Internet of Things and artificial intelligence technologies, smart IoT devices are being used more and more widely in urban management, IoT monitoring dimensions are becoming more diverse, and the amount of IoT data is growing exponentially. The industry models in refined urban management solutions based on IoT devices have increasingly higher requirements for the accuracy and reliability of their data.

[0003] The number of garbage collection points in an urban area generally ranges from 10,000 to 50,000. On average, 40% of the garbage collection points are mobile garbage collection points, which are not fixed in place and have a certain degree of mobility every day. Accurate management of the location of garbage collection points is critical to the refined operation of urban garbage collection and transportation, but it is unrealistic to rely on manpower for maintenance.

[0004] There are two basic methods in the industry. One is to judge whether it is a collection point based on the residence time of the garbage collection truck, which will cause a lot of misjudgment and its accuracy is less than 60%. The other is to purchase expensive data collection equipment to collect vehicle operation data. However, there are many brands of sanitation vehicles in the industry, and many types of vehicles do not open their data protocols. Therefore, although the data collection equipment can successfully collect data for a certain type of vehicle, it cannot successfully collect data for all models of the entire fleet. Summary of the invention

[0005] The object of the present invention is to provide a method and device for identifying the location of a garbage point.

[0006] In order to solve the above problems, the present invention provides a method for identifying the location of garbage points, comprising:

[0007] Step S1, obtaining a real-time operation data of a sanitation vehicle at a preset time interval, each real-time operation data includes: timestamp, vehicle speed, vehicle position and vehicle weight;

[0008] Step S2: if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as deceleration, and then step S1 is continued; otherwise, step S3 is executed;

[0009] Step S3, if there is only one real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, then the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as low speed, and then step S1 is continued; if the vehicle speed in the current real-time operation data is greater than the preset low-speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then all real-time operation data in the model stack are cleared, and then step S1 is continued;

[0010] Step S4: if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is greater than the vehicle speed in the real-time operation data at the top of the stack, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as accelerated, and then step S1 is continued; if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the vehicle speed in the real-time operation data at the top of the stack, the real-time operation data at the top of the stack is replaced by the current real-time operation data, and the real-time operation data at the top of the stack is marked as low speed, and then step S1 is continued;

[0011] Step S5: if there are three real-time operation data in the model stack, after assembling the three real-time operation data into a set of candidate model data, all the real-time operation data in the model stack are cleared;

[0012] Step S6, identifying the location of the garbage point based on the candidate model data.

[0013] Furthermore, in the above method, step S6, identifying the location of the garbage point based on the candidate model data, includes:

[0014] If, in a set of candidate model data, the vehicle weight in the real-time operation data marked as low speed is greater than the vehicle weight in the real-time operation data marked as deceleration, then the set of candidate model data is used as a filtered set of candidate model data;

[0015] Based on the filtered data of each set of candidate models, the locations of garbage spots are identified.

[0016] Furthermore, in the above method, based on the filtered candidate model data, identifying the location of the garbage point includes:

[0017] Merge the filtered candidate model data groups with similar positions into candidate model data with the same position;

[0018] Based on the candidate model data at the same location, the location of the garbage point is identified.

[0019] Furthermore, in the above method, based on the candidate model data at the same location, identifying the location of the garbage point includes:

[0020] It is determined whether the candidate model data at the same location meets the operating index requirements of the sanitation vehicle to which it belongs. If so, the location is used as the location of the garbage point.

[0021] Furthermore, in the above method, after the location is taken as the location of the garbage point, the method further includes:

[0022] Based on the map information and the location of the garbage point, the type corresponding to the location of the garbage point is obtained.

[0023] According to another aspect of the present invention, there is also provided a garbage point location identification device, comprising:

[0024] The first device is used to obtain a real-time operation data of the sanitation vehicle at a preset time interval, and each real-time operation data includes: a timestamp, a vehicle speed, a vehicle position and a vehicle weight;

[0025] The second device is used for, if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, pushing the current real-time operation data to the top of the model stack, marking the real-time operation data at the top of the stack as deceleration, and continuing to execute the first device, otherwise executing the third device in step;

[0026] The third device is used for, if there is only one real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, then pushing the current real-time operation data to the top of the model stack, marking the real-time operation data at the top of the stack as low speed, and then continuing to execute the first device of step; if the vehicle speed in the current real-time operation data is greater than the preset low-speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then clearing all the real-time operation data in the model stack and continuing to execute the first device of step;

[0027] The fourth device is used for, if there are two real-time operation data in the model stack and the vehicle speed in the current real-time operation data is greater than the vehicle speed in the real-time operation data at the top of the stack, pushing the current real-time operation data to the top of the model stack, marking the real-time operation data at the top of the stack as accelerated, and continuing to execute the first device of step; if there are two real-time operation data in the model stack and the vehicle speed in the current real-time operation data is less than or equal to a preset low-speed threshold, replacing the real-time operation data at the top of the stack with the current real-time operation data, marking the real-time operation data at the top of the stack as low speed, and continuing to execute the first device;

[0028] A fifth device, for clearing all the real-time operation data in the model stack after assembling the three real-time operation data into a set of candidate model data if there are three real-time operation data in the model stack;

[0029] The sixth device is used to identify the location of the garbage point based on the candidate model data.

[0030] Further, in the above-mentioned device, the sixth device is used to use the group of candidate model data as a filtered group of candidate model data if, in a group of candidate model data, the vehicle weight in the real-time operation data marked as low speed is greater than the vehicle weight in the real-time operation data marked as deceleration;

[0031] Based on the filtered data of each set of candidate models, the locations of garbage spots are identified.

[0032] Furthermore, in the above-mentioned device, the sixth device is used to merge the filtered groups of candidate model data with similar positions into candidate model data at the same position; and identify the position of the garbage point based on the candidate model data at the same position.

[0033] According to another aspect of the present invention, there is further provided a computer-readable storage medium having computer-executable instructions stored thereon, wherein when the computer-executable instructions are executed by a processor, the processor is caused to:

[0034] Step S1, obtaining a real-time operation data of a sanitation vehicle at a preset time interval, each real-time operation data includes: timestamp, vehicle speed, vehicle position and vehicle weight;

[0035] Step S2: if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as deceleration, and then step S1 is continued; otherwise, step S3 is executed;

[0036] Step S3, if there is only one real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, then the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as low speed, and then step S1 is continued; if the vehicle speed in the current real-time operation data is greater than the preset low-speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then all real-time operation data in the model stack are cleared, and then step S1 is continued;

[0037] Step S4: if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is greater than the vehicle speed in the real-time operation data at the top of the stack, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as accelerated, and then step S1 is continued; if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the vehicle speed in the real-time operation data at the top of the stack, the real-time operation data at the top of the stack is replaced by the current real-time operation data, and the real-time operation data at the top of the stack is marked as low speed ls, and then step S1 is continued;

[0038] Step S5: if there are three real-time operation data in the model stack, after assembling the three real-time operation data into a set of candidate model data, all the real-time operation data in the model stack are cleared;

[0039] Step S6, identifying the location of the garbage point based on the candidate model data.

[0040] According to another aspect of the present invention, there is also provided a computer device, comprising:

[0041] Processor; and

[0042] a memory arranged to store computer executable instructions which, when executed, cause the processor to:

[0043] Step S1, obtaining a real-time operation data of a sanitation vehicle at a preset time interval, each real-time operation data includes: timestamp, vehicle speed, vehicle position and vehicle weight;

[0044] Step S2: if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as deceleration, and then step S1 is continued; otherwise, step S3 is executed;

[0045] Step S3, if there is only one real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, then the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as low speed, and then step S1 is continued; if the vehicle speed in the current real-time operation data is greater than the preset low-speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then all real-time operation data in the model stack are cleared, and then step S1 is continued;

[0046] Step S4, if there are two real-time job data in the model stack, and the vehicle speed in the current real-time job data is greater than the vehicle speed in the real-time job data at the top of the stack, then push the current real-time job data onto the top of the model stack, mark the real-time job data at the top of the stack as accelerated, and then continue to execute Step S1; if there are two real-time job data in the model stack, and the vehicle speed in the current real-time job data is less than or equal to the vehicle speed in the real-time job data at the top of the stack, then replace the real-time job data at the top of the stack with the current real-time job data, mark the real-time job data at the top of the stack as low speed ls, and then continue to execute Step S1;

[0047] Step S5, if there are three real-time job data in the model stack, then assemble these three real-time job data into a group of candidate model data, and then clear all the real-time job data in the model stack;

[0048] Step S6, based on the candidate model data, identify the location of the garbage point.

[0049] Compared with the prior art, the present invention can completely capture the short停留事件 of the sanitation vehicle by identifying the behavior of decelerating dc at time point t1, then running at low speed ls at time point t2, and then accelerating ac at time point t3, where t1 < t2 < t3, learn the location of the garbage point more accurately, and effectively improve the management efficiency of urban garbage collection and transportation.

[0050] In addition, the present invention first analyzes the time period when the vehicle is running at low speed or even stationary, and obtains reliable weighing data during this time period to ensure accurate screening of each group of candidate model data.

[0051] Furthermore, the present invention can avoid the problem of directly adding a large number of and overly redundant garbage points every time a new garbage point is learned by merging the filtered groups of candidate model data with similar positions into candidate model data at the same position. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a schematic diagram of a garbage point location recognition device according to an embodiment of the present invention;

[0053] Figure 2 is a flowchart of a garbage point location recognition method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0055] Such as Figure 1 and 2As shown, the present invention provides a method for identifying a garbage point location, comprising steps S1 to S6:

[0056] Step S1, obtaining a real-time operation data of a sanitation vehicle at a preset time interval, each real-time operation data includes: timestamp, vehicle speed, vehicle position and vehicle weight;

[0057] Specifically, Figure 1 As shown in the figure, when a sanitation operation vehicle equipped with IoT sensor equipment is started and driving, the sensor will regularly upload data, including GPS positioning data, i.e. vehicle location, speed data, mileage data, compression equipment opening and closing status data, etc. These vehicle operation data can be uploaded to the behavior database of the cloud server through the 4G network.

[0058] Specifically, the behavior database can store the original data of the daily operation behavior of the operation vehicle, including vehicle trajectory data, compression vehicle opening and closing data, fuel consumption data, etc. These data reflect the performance of various indicators of the vehicle throughout the entire operation cycle.

[0059] The machine learning model of garbage collection and transportation points can include the following indicators: vehicle position (hereinafter represented by GPS), vehicle weighing (hereinafter represented by LW), deceleration (hereinafter represented by DC), acceleration (hereinafter represented by AC), low speed (hereinafter represented by LS, low speed can include the stationary behavior of sanitation vehicles) and compression state (hereinafter represented by CS, 0: uncompressed, 1: compressed, this indicator is optional because not all vehicle models can obtain the compression state). When calculating vehicles that can obtain the compression state, cs = 1 means that they are being collected and transported. When calculating vehicles that cannot obtain the compression state, the process of deceleration-low speed-acceleration must be met, and there must be an incremental change in LW during the low-speed stage to be considered a potential collection and transportation behavior, where LS <= 5 kilometers per hour. For the GPS where the potential collection and transportation behavior occurs, it can be considered as a potential garbage point.

[0060] Step S2: if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as deceleration dc, and then step S1 is continued; otherwise, step S3 is executed;

[0061] For example, if the model stack is empty and the current vehicle speed is <= 5 km / h, the data is pushed into the stack, the top data of the stack is marked as deceleration dc, and step S1 is continued, otherwise step S3 is executed;

[0062] Step S3, if there is only one real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, then the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as low speed ls, and then step S1 is continued; if the vehicle speed in the current real-time operation data is greater than the preset low-speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then all real-time operation data in the model stack are cleared, and then step S1 is continued;

[0063] For example, if there is only one element in the model stack, this operation is performed, otherwise step S4 is performed. If the current speed is <= 5 km / h and is less than the first element in the model stack, the current data is pushed into the stack, and the top data is marked as low speed ls, otherwise the stack is cleared and step S1 is continued;

[0064] Step S4: if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is greater than the vehicle speed in the real-time operation data at the top of the stack, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as acceleration ac, and then step S1 is continued; if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the vehicle speed in the real-time operation data at the top of the stack, the real-time operation data at the top of the stack is replaced by the current real-time operation data, and the real-time operation data at the top of the stack is marked as low speed ls, and then step S1 is continued;

[0065] For example, if the model stack has two elements, and the current speed, that is, the speed of the third element, is greater than the speed of the second element, the third element is pushed into the stack, and the data on the top of the stack is marked as accelerated ac, which is considered to be data that meets the model; if the model stack has two elements, and the current speed, that is, the speed of the third element, is less than or equal to the speed of the second element, the second element on the top of the stack is popped out of the stack, and the third element is pushed into the stack as the second element;

[0066] Step S5: if there are three real-time operation data in the model stack, after assembling the three real-time operation data into a set of candidate model data, all the real-time operation data in the model stack are cleared;

[0067] Here, for data that meets the model, a set of candidate model data can be assembled. Then the model stack is cleared, and step S1 is continued to enter a new round of model screening. In this way, only a data stack with a maximum of 3 elements is needed to complete stateful streaming computing.

[0068] Step S6: Based on the candidate model data, identify the locations of the garbage points.

[0069] Here, for the vehicle operation behavior time-series point data group: The operation sanitation vehicle can upload real-time operation data at a cycle of every 5 seconds. The present invention can form a group of vehicle operation behavior data with time-series characteristics within a certain time period as a group of candidate model data.

[0070] The present invention has a concept of low speed in the collection and transportation model, which is proposed based on two aspects: First, in actual garbage collection and transportation, the garbage truck is not necessarily completely stationary. There is also a scenario where the vehicle is driving at a low speed and the bucket hanger hangs the roadside trash can onto the vehicle. Second, in some scenarios, the collection and transportation is very fast, but when the vehicle data acquisition frequency is relatively short, such as once every 5 seconds, it cannot fully capture the short-term停留 events of the sanitation vehicle.

[0071] Collect the collection and transportation operation behavior data such as the positioning data of the vehicle GPS and the opening and closing data of the garbage truck compression device every 10 seconds. Combine the urban geography and infrastructure data, including geographical data such as urban schools, hospitals, administrative organs, street shops, residential communities, industrial parks, etc., and infrastructure data such as intersections and transfer stations, so as to more accurately analyze the specific locations of the garbage points.

[0072] The present invention can fully capture the short-term停留 events of the sanitation vehicle by identifying the behavior of starting to decelerate dc at time point t1, then having a low speed ls at time point t2, and then accelerating ac at time point t3, where t1 < t2 < t3. It can more accurately learn the locations of the garbage points and effectively improve the management efficiency of urban garbage collection and transportation.

[0073] In an embodiment of the garbage point location identification method of the present invention, in step S6, based on the candidate model data, identifying the locations of the garbage points includes:

[0074] Step S61: If in a group of candidate model data, the vehicle weight in the real-time operation data marked as low speed ls is greater than the vehicle weight in the real-time operation data marked as deceleration dc, then use this group of candidate model data as a filtered group of candidate model data;

[0075] Step S62: Based on the filtered groups of candidate model data, identify the locations of the garbage points.

[0076] Here, the vehicle weight increment lw change within its model time period can be calculated. If it meets the reasonable change range, then transfer its model data to the next round of screening.

[0077] In order to obtain reliable weighing data, the time period when the vehicle is at a low speed or even stationary can be analyzed first, and the weighing data reported during this time period is highly reliable.

[0078] The time series data of the sanitation vehicle collection and operation behavior events can be set to meet two conditions:

[0079] 1) At time point t1, the collection and transportation starts to decelerate dc, and then to low speed ls at time stamp t2, and then to accelerate ac at time stamp t3. Among them, t1 <t2<t3;

[0080] 2) If the weight obtained at low speed has incremental changes, then this set of time series data obtains a collection and transportation event consisting of data from t1 to t3. Each element in the event includes vehicle behavior data such as timestamp, positioning, speed, mileage and weight.

[0081] In an embodiment of the garbage point location identification method of the present invention, step S62, based on the filtered candidate model data, identifies the location of the garbage point, including:

[0082] Step S621, merging the filtered groups of candidate model data with similar positions into candidate model data with the same position;

[0083] Here, if Figure 1 As shown, the model database can be established for each of the multiple operation indicators according to the operation sanitation requirements, and various sanitation operation mathematical models can be established based on the operation indicator requirements.

[0084] The model database can be used to store garbage collection and transportation operation models learned based on historical data and machine algorithms.

[0085] The behavior model calculation engine can start calculations based on the time series data of the vehicle operation behavior in the behavior database and the model design requirements in the model database. During the calculation process, GPS points with close spatial distances can be merged (indicating that the garbage points in this batch are moving points of the same garbage points in the real world), and the calculation results can be written into the view database.

[0086] Step S622, identifying the location of the garbage point based on the candidate model data at the same location.

[0087] Here, by matching the real-time operation data stream of the operation vehicle with the garbage collection and transportation model, the candidate model data that meets the model rules can be written into the view database.

[0088] Based on the filtered groups of candidate model data, the data can be compared with the historical data in the view database, and the filtered candidate model data within the longitude and latitude range of 20 meters can be merged.

[0089] Potential garbage spots within a preset radius can be set as garbage spots with similar locations. The radius of potential garbage spots can be set to 10 meters, that is, garbage spots that appear within a radius of 10 meters of an existing garbage spot will be considered existing garbage spots. Since garbage spots are mobile and GPS positioning also has a certain amount of drift, if new garbage spots are added directly every time they are learned, it will cause massive and overly redundant garbage spots. According to the current interval between mobile garbage spots, a radius of 10 meters is enough to cover a range of 20 meters, which is enough to cover most of the scenarios where there is occasional drift or slight changes in the collection and transportation location.

[0090] In an embodiment of the garbage point location identification method of the present invention, step S622, based on the candidate model data of the same location, identifies the location of the garbage point, including:

[0091] Step S6221, determine whether the candidate model data at the same location meets the operating index requirements of the sanitation vehicle to which it belongs. If so, use the location as the garbage point.

[0092] Here, the candidate model data at the same location can be counted to obtain the number of times each potential garbage point meets the operating indicator requirements in the vehicle dimension and date dimension, and finally the actual garbage point location is obtained based on the requirements of garbage collection and transportation.

[0093] The general operating indicators of the sanitation mechanized operation assessment require that the operating vehicles operate in accordance with certain operating specifications within a specified time period. For example, the assessment in XXX City requires that sanitation collection and transportation vehicles completely collect and transport garbage points on the preset road section between 18:00 and 6:00 in the morning.

[0094] There are different types of sanitation vehicles, including, for example: dry garbage collection vehicles, wet garbage collection vehicles; different types of vehicles may have different operating index requirements at the workplace, such as dry garbage collection vehicles are required to collect and transport garbage twice a day at the specified time; wet garbage collection vehicles are required to collect and transport garbage three times a day at the specified time.

[0095] In an embodiment of the method for identifying the location of a garbage point of the present invention, after taking the location as the location of the garbage point in step S63, the method further includes:

[0096] Based on the map information and the location of the garbage point, the type corresponding to the location of the garbage point is obtained.

[0097] Here, if Figure 1 As shown, data presentation in different dimensions can be provided according to the location of garbage points in the view database. Combined with map information, the location of garbage points is marked with corresponding types, such as schools, supermarkets, and parks, etc.

[0098] According to another aspect of the present invention, there is also provided a garbage point location identification device, comprising:

[0099] The first device is used to obtain a real-time operation data of the sanitation vehicle at a preset time interval, and each real-time operation data includes: a timestamp, a vehicle speed, a vehicle position and a vehicle weight;

[0100] The second device is used for, if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, pushing the current real-time operation data to the top of the model stack, marking the real-time operation data at the top of the stack as deceleration, and continuing to execute the first device, otherwise executing the third device in step;

[0101] The third device is used for, if there is only one real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, then pushing the current real-time operation data to the top of the model stack, marking the real-time operation data at the top of the stack as low speed, and then continuing to execute the first device of step; if the vehicle speed in the current real-time operation data is greater than the preset low-speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then clearing all the real-time operation data in the model stack and continuing to execute the first device of step;

[0102] The fourth device is used for, if there are two real-time operation data in the model stack and the vehicle speed in the current real-time operation data is greater than the vehicle speed in the real-time operation data at the top of the stack, pushing the current real-time operation data into the top of the model stack, marking the real-time operation data at the top of the stack as accelerated, and continuing to execute the first device of step; if there are two real-time operation data in the model stack and the vehicle speed in the current real-time operation data is less than or equal to the vehicle speed in the real-time operation data at the top of the stack, replacing the real-time operation data at the top of the stack with the current real-time operation data, marking the real-time operation data at the top of the stack as low speed, and continuing to execute the first device;

[0103] A fifth device, for clearing all the real-time operation data in the model stack after assembling the three real-time operation data into a set of candidate model data if there are three real-time operation data in the model stack;

[0104] The sixth device is used to identify the location of the garbage point based on the candidate model data.

[0105] Further, in the above-mentioned device, the sixth device is used to use the group of candidate model data as a filtered group of candidate model data if, in a group of candidate model data, the vehicle weight in the real-time operation data marked as low speed is greater than the vehicle weight in the real-time operation data marked as deceleration;

[0106] Based on the filtered data of each set of candidate models, the locations of garbage spots are identified.

[0107] Furthermore, in the above-mentioned device, the sixth device is used to merge the filtered groups of candidate model data with similar positions into candidate model data at the same position; and identify the position of the garbage point based on the candidate model data at the same position.

[0108] According to another aspect of the present invention, there is further provided a computer-readable storage medium having computer-executable instructions stored thereon, wherein when the computer-executable instructions are executed by a processor, the processor is caused to:

[0109] Step S1, obtaining a real-time operation data of a sanitation vehicle at a preset time interval, each real-time operation data includes: timestamp, vehicle speed, vehicle position and vehicle weight;

[0110] Step S2: if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as deceleration, and then step S1 is continued; otherwise, step S3 is executed;

[0111] Step S3, if there is only one real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, then the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as low speed, and then step S1 is continued; if the vehicle speed in the current real-time operation data is greater than the preset low-speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then all real-time operation data in the model stack are cleared, and then step S1 is continued;

[0112] Step S4: if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is greater than the vehicle speed in the real-time operation data at the top of the stack, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as accelerated, and then step S1 is continued; if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the vehicle speed in the real-time operation data at the top of the stack, the real-time operation data at the top of the stack is replaced by the current real-time operation data, and the real-time operation data at the top of the stack is marked as low speed ls, and then step S1 is continued;

[0113] Step S5: if there are three real-time operation data in the model stack, after assembling the three real-time operation data into a set of candidate model data, all the real-time operation data in the model stack are cleared;

[0114] Step S6, identifying the location of the garbage point based on the candidate model data.

[0115] According to another aspect of the present invention, there is also provided a computer device, comprising:

[0116] Processor; and

[0117] a memory arranged to store computer executable instructions which, when executed, cause the processor to:

[0118] Step S1, obtaining a real-time operation data of a sanitation vehicle at a preset time interval, each real-time operation data includes: timestamp, vehicle speed, vehicle position and vehicle weight;

[0119] Step S2: if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as deceleration, and then step S1 is continued; otherwise, step S3 is executed;

[0120] Step S3: if there is only one real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, then the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as low speed, and then step S1 is continued; if the vehicle speed in the current real-time operation data is greater than the preset low-speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then all real-time operation data in the model stack are cleared, and then step S1 is continued;

[0121] Step S4: if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is greater than the vehicle speed in the real-time operation data at the top of the stack, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as accelerated, and then step S1 is continued; if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the vehicle speed in the real-time operation data at the top of the stack, the real-time operation data at the top of the stack is replaced by the current real-time operation data, and the real-time operation data at the top of the stack is marked as low speed ls, and then step S1 is continued;

[0122] Step S5: if there are three real-time operation data in the model stack, after assembling the three real-time operation data into a set of candidate model data, all the real-time operation data in the model stack are cleared;

[0123] Step S6, identifying the location of the garbage point based on the candidate model data.

[0124] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0125] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0126] Obviously, those skilled in the art can make various changes and modifications to the invention without departing from the spirit and scope of the invention. Thus, if these modifications and variations of the invention fall within the scope of the claims of the invention and their equivalents, the invention is also intended to include these modifications and variations.

Claims

1. A method for identifying garbage point locations, characterized in that: include: Step S1, obtaining a real-time operation data of a sanitation vehicle at a preset time interval, each real-time operation data includes: timestamp, vehicle speed, vehicle position and vehicle weight; Step S2: if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as deceleration, and then step S1 is continued; otherwise, step S3 is executed; Step S3: if there is only one real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, then the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as low speed, and then step S1 is continued; if the vehicle speed in the current real-time operation data is greater than the preset low speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then all real-time operation data in the model stack are cleared, and then step S1 is continued; Step S4: if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is greater than the vehicle speed in the real-time operation data at the top of the stack, the current real-time operation data is pushed to the top of the model stack, and the real-time operation data at the top of the stack is marked as accelerated, and then step S1 is continued; if there are two real-time operation data in the model stack, and the vehicle speed in the current real-time operation data is less than or equal to the vehicle speed in the real-time operation data at the top of the stack, the real-time operation data at the top of the stack is replaced by the current real-time operation data, and the real-time operation data at the top of the stack is marked as low speed, and then step S1 is continued; Step S5: if there are three real-time operation data in the model stack, after assembling the three real-time operation data into a set of candidate model data, all the real-time operation data in the model stack are cleared; Step S6, identifying the location of the garbage point based on the candidate model data; Step S6, identifying the location of the garbage point based on the candidate model data, includes: If, in a set of candidate model data, the vehicle weight in the real-time operation data marked as low speed is greater than the vehicle weight in the real-time operation data marked as deceleration, then the set of candidate model data is used as a filtered set of candidate model data; Based on the filtered data of each group of candidate models, the location of garbage points is identified; Based on the filtered data of each set of candidate models, the location of garbage points is identified, including: Merge the filtered candidate model data groups with similar positions into candidate model data with the same position; Based on the candidate model data at the same location, the location of the garbage point is identified.

2. The method for identifying the location of a garbage point according to claim 1, characterized in that: Based on the candidate model data at the same location, identify the location of the garbage point, including: It is determined whether the candidate model data at the same location meets the operating index requirements of the sanitation vehicle to which it belongs. If so, the location is used as the location of the garbage point.

3. The method for identifying the location of a garbage point according to claim 1, characterized in that: After the location is taken as the location of the garbage point, it also includes: Based on the map information and the location of the garbage point, the type corresponding to the location of the garbage point is obtained.

4. A garbage point location identification device, characterized in that: include: The first device is used to obtain a real-time operation data of the sanitation vehicle at a preset time interval, and each real-time operation data includes: a timestamp, a vehicle speed, a vehicle position and a vehicle weight; The second device is used for, if the model stack is empty and the vehicle speed in the current real-time operation data is less than or equal to the preset low-speed threshold, pushing the current real-time operation data to the top of the model stack, marking the real-time operation data at the top of the stack as deceleration, and continuing to execute the first device, otherwise executing the third device in step; The third device is used for, if there is only one real-time operation data in the model stack and the vehicle speed in the current real-time operation data is less than the vehicle speed in the real-time operation data at the top of the stack, pushing the current real-time operation data to the top of the model stack, marking the real-time operation data at the top of the stack as low speed, and then continuing to execute the first device of step; if the vehicle speed in the current real-time operation data is greater than a preset low-speed threshold, or the vehicle speed in the current real-time operation data is greater than or equal to the vehicle speed in the real-time operation data at the top of the stack, then clearing all the real-time operation data in the model stack and continuing to execute the first device of step; The fourth device is used for, if there are two real-time operation data in the model stack and the vehicle speed in the current real-time operation data is greater than the vehicle speed in the real-time operation data at the top of the stack, pushing the current real-time operation data into the top of the model stack, marking the real-time operation data at the top of the stack as accelerated, and continuing to execute the first device of step; if there are two real-time operation data in the model stack and the vehicle speed in the current real-time operation data is less than or equal to the vehicle speed in the real-time operation data at the top of the stack, replacing the real-time operation data at the top of the stack with the current real-time operation data, marking the real-time operation data at the top of the stack as low speed, and continuing to execute the first device; A fifth device, for clearing all the real-time operation data in the model stack after assembling the three real-time operation data into a set of candidate model data if there are three real-time operation data in the model stack; A sixth device for identifying the location of a garbage point based on the candidate model data; The sixth device is used to use the group of candidate model data as a filtered group of candidate model data if the vehicle weight in the real-time operation data marked as low speed is greater than the vehicle weight in the real-time operation data marked as deceleration in a group of candidate model data; Based on the filtered data of each group of candidate models, the location of garbage points is identified; The sixth device is used to merge the filtered candidate model data groups with similar positions into candidate model data at the same position; and identify the position of the garbage point based on the candidate model data at the same position.

5. A computer-readable storage medium having computer-executable instructions stored thereon, wherein: When the computer executable instructions are executed by a processor, the processor is caused to: perform the method according to any one of claims 1 to 3.

6. A computer device, wherein: include: processor; as well as A memory arranged to store computer executable instructions which, when executed, cause the processor to: perform a method as claimed in any one of claims 1 to 3.

Images