A route planning method, device, equipment and medium of an unmanned cleaning vehicle

Abstract

A route planning method, device, equipment and medium of an unmanned cleaning vehicle are disclosed. The method comprises the following steps: acquiring, by an automatic driving system in the unmanned cleaning vehicle, a region image corresponding to a preset-specification region in front of the unmanned cleaning vehicle; mapping the region image to a region on a preset semantic map; dividing the region on the semantic map according to road markings on the semantic map; determining at least two cleaning regions and identifying the number of region garbage corresponding to each cleaning region; calculating the absolute value of the garbage difference between adjacent cleaning regions using the number of region garbage; and adjusting the cleaning route of the unmanned cleaning vehicle according to the number of region garbage by a planning module if the absolute value of the garbage difference exceeds a preset lane-changing threshold. Thus, the number of garbage in the preset-specification region is effectively analyzed by combining the region image and object perception, and the cleaning route of the unmanned cleaning vehicle can be flexibly adjusted.

Description

Technical Field

[0001] This invention relates to the field of cleaning vehicle driving technology, and in particular to a route planning method, apparatus, equipment and medium for an unmanned cleaning vehicle. Background Technology

[0002] With the continuous development and progress of science and technology, more and more tools are becoming intelligent, especially the research on autonomous driving, which will bring great convenience to people's lives.

[0003] For example, unmanned cleaning vehicles can identify the location of garbage by analyzing images of the road surface and then clean it up.

[0004] However, the cleaning routes of unmanned cleaning vehicles are usually set based on routine cleaning tasks. The routes are relatively fixed and the cleaning coverage is limited. They cannot judge whether there is a lot of garbage or a relatively clean road surface, resulting in low cleaning flexibility. Summary of the Invention

[0005] This invention provides a route planning method, device, equipment, and medium for unmanned cleaning vehicles, which solves the technical problem that the cleaning routes of current unmanned cleaning vehicles are usually set based on routine cleaning tasks, the routes are relatively fixed and the cleaning coverage is limited, and it is impossible to judge the road surface with more garbage and relatively clean surfaces, resulting in low cleaning flexibility.

[0006] The first aspect of this invention provides a route planning method for an unmanned cleaning vehicle, comprising:

[0007] Acquire an image of the area corresponding to a preset specification area in front of the unmanned cleaning vehicle;

[0008] Based on the preset semantic map and the area image, at least two areas to be cleaned are determined and the amount of garbage corresponding to each area to be cleaned is identified.

[0009] The absolute value of the difference in waste between adjacent areas to be cleaned is calculated using the amount of waste in the area.

[0010] If the absolute value of the garbage difference exceeds the preset lane change threshold, the cleaning route of the unmanned cleaning vehicle will be adjusted by the planning module according to the amount of garbage in the area.

[0011] Optionally, the method further includes:

[0012] If the absolute value of the garbage difference does not exceed the preset lane change threshold, the unmanned cleaning vehicle maintains its cleaning route at the current moment through the planning module.

[0013] Optionally, the step of determining at least two areas to be cleaned based on a preset semantic map and the area image, and identifying the amount of garbage corresponding to each area to be cleaned, includes:

[0014] The road markings and road areas corresponding to the region image are located from the preset semantic map;

[0015] The road area is divided according to the road markings, and at least two areas to be cleaned are identified.

[0016] Garbage is identified in the area images corresponding to each of the areas to be cleaned, and the amount of garbage in each area is determined.

[0017] Optionally, the road markings include dashed variable lane lines and solid roadbed lines; the step of dividing the road area according to the road markings and determining at least two areas to be cleaned includes:

[0018] Count the number of dashed lines in the variable lane area within the road region;

[0019] If there is only one dashed line, then the road area is divided into two areas to be cleaned, with the dashed variable lane line as the center line and the solid roadbed line as the boundary line.

[0020] If there are two or more dashed lines, the road area is divided into multiple areas to be cleaned, with the variable lane dashed lines as equal dividing lines and the roadbed solid lines as boundary lines.

[0021] Optionally, the step of adjusting the cleaning route of the unmanned cleaning vehicle according to the amount of garbage in the area by the planning module if the absolute value of the garbage difference exceeds a preset lane-changing threshold includes:

[0022] If the absolute value of the garbage difference exceeds the preset lane-changing threshold, a lane-changing command is sent to the planning module;

[0023] The planning module allocates priority to each area to be cleaned based on the amount of garbage in the area, from most to least.

[0024] If the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is less than or equal to a preset threshold, the cleaning route of the unmanned cleaning vehicle will be directly adjusted to the highest priority cleaning area through the planning module.

[0025] If the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is greater than a preset threshold, the cleaning route of the unmanned cleaning vehicle will be continuously planned by the planning module and then adjusted to the highest priority cleaning area.

[0026] Optionally, the step of adjusting the cleaning route of the unmanned cleaning vehicle to the highest priority cleaning area after continuous lane-changing planning by the planning module includes:

[0027] The planning module adjusts the cleaning route of the unmanned cleaning vehicle to the adjacent area to be cleaned, and assigns a lane-changing cost to the highest priority area to be cleaned.

[0028] If the lane change cost is less than or equal to a preset cost threshold, the lane change cost is reduced by the planning module according to a preset running time gradient.

[0029] When the lane change cost equals the preset minimum cost, the cleaning route is adjusted to the highest priority area to be cleaned through the planning module.

[0030] Optionally, the method further includes:

[0031] If the cost of changing lanes exceeds a preset cost threshold, the planning module will maintain the cleaning route in the area to be cleaned at the current moment.

[0032] Optionally, the method further includes:

[0033] Once the cleaning route is adjusted to the highest priority area to be cleaned, the planning module assigns a priority cleaning marker to the current area of ​​the unmanned cleaning vehicle in the semantic map.

[0034] The planning module generates an updated cleaning route using the priority cleaning markers and determines whether the updated cleaning route meets the road U-turn conditions.

[0035] If the conditions are met, the unmanned cleaning vehicle will move according to the updated cleaning route via the planning module to complete the cleaning action.

[0036] If the conditions are not met, the area coordinates corresponding to the priority cleaning identifier are obtained and sent to the communication-associated collaborative unmanned cleaning vehicle.

[0037] The collaborative unmanned cleaning vehicle is equipped with a fixed operating route that passes through the coordinates of the area.

[0038] A second aspect of the present invention also provides a route planning device for an unmanned cleaning vehicle, comprising:

[0039] The area image acquisition module is used to acquire area images corresponding to a preset size area in front of the unmanned cleaning vehicle;

[0040] The area division and waste identification module is used to determine at least two areas to be cleaned based on a preset semantic map and the area image, and to identify the amount of waste in each area to be cleaned.

[0041] The garbage difference calculation module is used to calculate the absolute value of the garbage difference between adjacent areas to be cleaned, based on the amount of garbage in the area.

[0042] The cleaning route planning module is used to adjust the cleaning route of the unmanned cleaning vehicle according to the amount of garbage in the area if the absolute value of the garbage difference exceeds a preset lane change threshold.

[0043] A third aspect of the present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the route planning method for an unmanned cleaning vehicle as described in any of the first aspects of the present invention.

[0044] A fourth aspect of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the route planning method for an unmanned cleaning vehicle as described in any of the first aspects of the present invention.

[0045] As can be seen from the above technical solutions, the present invention has the following advantages:

[0046] This invention utilizes the autonomous driving system within an unmanned cleaning vehicle to acquire an image of a pre-defined area in front of the vehicle. The image is then mapped onto a pre-defined semantic map, which is used to divide the area according to road markings. At least two areas are identified as needing cleaning, and the amount of trash in each area is determined. The absolute value of the difference in trash quantity between adjacent areas is then calculated. If this difference exceeds a pre-defined lane-changing threshold, the cleaning route of the unmanned cleaning vehicle is adjusted by a planning module based on the amount of trash in each area. By combining area imagery and object perception, the amount of trash within a pre-defined area can be effectively analyzed, enabling flexible adjustment of the cleaning route. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1A flowchart illustrating the steps of a route planning method for an unmanned cleaning vehicle provided in Embodiment 1 of the present invention;

[0049] Figure 2 This is a flowchart illustrating the steps of a route planning method for an unmanned cleaning vehicle according to Embodiment 2 of the present invention.

[0050] Figure 3 This is a structural block diagram of a route planning device for an unmanned cleaning vehicle provided in Embodiment 3 of the present invention. Detailed Implementation

[0051] This invention provides a route planning method, apparatus, equipment, and medium for unmanned cleaning vehicles, which addresses the technical problem that current unmanned cleaning vehicles typically set their cleaning routes based on routine cleaning tasks, resulting in relatively fixed routes, limited cleaning coverage, inability to judge areas with heavy garbage or relatively clean surfaces, and low cleaning flexibility.

[0052] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0053] Please see Figure 1 , Figure 1 This is a flowchart illustrating the steps of a route planning method for an unmanned cleaning vehicle provided in Embodiment 1 of the present invention.

[0054] This invention provides a route planning method for an unmanned cleaning vehicle, comprising the following steps:

[0055] Step 101: Obtain an image of the area corresponding to the preset specification area in front of the unmanned cleaning vehicle;

[0056] Regional images refer to images acquired within a pre-defined area using front-facing wide-angle high-definition cameras and telephoto cameras mounted on the unmanned cleaning vehicle. The pre-defined area can be a road area with a length ranging from 25 to 50 meters.

[0057] In this embodiment of the invention, a front-facing wide-angle high-definition camera and a telephoto camera mounted on the unmanned cleaning vehicle can be used to acquire regional images within a preset specification area in front of the unmanned cleaning vehicle, so as to provide a data basis for subsequent adjustment of the cleaning route.

[0058] Step 102: Based on the preset semantic map and region image, determine at least two regions to be cleaned and identify the amount of garbage in each region.

[0059] After obtaining the area image, it is necessary to further identify the area to be cleaned and the amount of garbage to be collected by the unmanned cleaning vehicle. To do this, the road area corresponding to the area image can be located in the preset semantic map and divided according to the road markings to determine at least two areas to be cleaned.

[0060] At the same time, the sensing module carried by the unmanned cleaning vehicle identifies the amount of garbage in each area to be cleaned based on the acquired area images.

[0061] It should be noted that semantic maps refer to high-precision maps containing a variety of semantic information. They acquire point cloud information of the physical world using LiDAR and then refine it to distinguish various objects and concepts such as lanes, cars, medians, roadside trees, signs, and the blue sky. Semantic information refers to the multi-layered, rich-dimensional information contained within these high-precision maps that enables autonomous vehicles to better understand driving rules, perceive road traffic conditions, and plan routes.

[0062] Step 103: Calculate the absolute value of the difference in garbage quantity between adjacent areas to be cleaned;

[0063] In this embodiment of the invention, after the sensing module identifies the amount of garbage in each area to be cleaned, in order to determine whether the cleaning route needs to be adjusted for unmanned cleaning vehicles, the absolute value of the difference between garbage in each pair of adjacent areas to be cleaned can be calculated based on the amount of garbage in each area to be cleaned, thereby detecting whether there is a large difference in the amount of garbage between the areas to be cleaned.

[0064] It should be noted that the absolute value of the garbage difference refers to the absolute value of the difference between the amount of garbage in adjacent areas to be cleaned. For example, if the amount of garbage in area A to be cleaned is 70 and the amount of garbage in the adjacent area B to be cleaned is 20, then the absolute value of the garbage difference is 50.

[0065] Step 104: If the absolute value of the garbage difference exceeds the preset lane change threshold, the cleaning route of the unmanned cleaning vehicle will be adjusted according to the amount of garbage in the area through the planning module.

[0066] After calculating the absolute value of the garbage difference, if it exceeds the preset lane-changing threshold, it indicates that there are lanes with a large difference in the amount of garbage in the preset area in front of the unmanned cleaning vehicle. The planning module can then prioritize the cleaning areas according to the amount of garbage in each area, thereby adjusting the cleaning route of the unmanned cleaning vehicle.

[0067] The adjustment of the cleaning route of the unmanned cleaning vehicle may include, but is not limited to, changing lanes to adjacent areas to be cleaned or continuously changing lanes to non-adjacent areas to be cleaned.

[0068] In this embodiment of the invention, the autonomous driving system inside the unmanned cleaning vehicle acquires an image of a pre-defined area in front of the vehicle. The area is mapped onto a pre-defined semantic map, and then divided according to road markings on the semantic map. At least two areas to be cleaned are identified, and the amount of trash in each area is determined. The absolute value of the difference in trash quantity between adjacent areas is calculated using the trash quantity. If the absolute value of the trash difference exceeds a pre-defined lane-changing threshold, the cleaning route of the unmanned cleaning vehicle is adjusted by the planning module based on the amount of trash in each area. Thus, by combining area imagery and object perception, the amount of trash within the pre-defined area is effectively analyzed, enabling flexible adjustment of the cleaning route of the unmanned cleaning vehicle.

[0069] Please see Figure 2 , Figure 2 This is a flowchart illustrating the steps of a route planning method for an unmanned cleaning vehicle provided in Embodiment 2 of the present invention.

[0070] This invention provides a route planning method for an unmanned cleaning vehicle, comprising the following steps:

[0071] Step 201: Obtain an image of the area corresponding to the preset specification area in front of the unmanned cleaning vehicle;

[0072] In this embodiment of the invention, the specific implementation process of step 201 is similar to that of step 101, and will not be repeated here.

[0073] Step 202: Based on the preset semantic map and region image, determine at least two regions to be cleaned and identify the amount of garbage in each region.

[0074] Optionally, step 202 may include the following sub-steps S11-S13:

[0075] S11. Locate the road markings and road areas corresponding to the preset semantic map image;

[0076] In this embodiment of the invention, after obtaining the regional image, the road markings corresponding to the regional image can be obtained from a preset semantic map.

[0077] In practice, unmanned cleaning vehicles typically acquire semantic information about their surroundings based on their location in a semantic map. The area image corresponds to a pre-defined area in front of the unmanned cleaning vehicle. Therefore, the road area can be located in the semantic map according to the area image. The road markings and road areas can be located through the semantic information carried by the semantic map to provide the data basis for the subsequent division of areas to be cleaned.

[0078] S12. Divide the road area according to the road markings and identify at least two areas to be cleaned;

[0079] Furthermore, the road markings include dashed lines for variable lanes and solid lines for the roadbed. S12 may include the following sub-steps:

[0080] Count the number of dashed lane markings within the road area;

[0081] If there is only one dashed line, the road area is divided into two areas to be cleaned, with the dashed variable lane line as the center line and the solid roadbed line as the boundary line.

[0082] If there are two or more dashed lines, the road area is divided into multiple areas to be cleaned, with the variable lane dashed lines as equal dividing lines and the roadbed solid lines as boundary lines.

[0083] Dashed lane lines refer to white or yellow dashed lines on the road, while solid roadbed lines refer to white solid lines on the road.

[0084] In this embodiment of the invention, the number of dashed variable lane lines in the semantic map can be counted first. If there is only one dashed line, it indicates that the road area is a two-lane area. At this time, the dashed variable lane line can be used as the center line and the solid roadbed line can be used as the boundary line to divide the road area into two areas to be cleaned.

[0085] If there are two or more dashed lines, it indicates that the road area is a three-lane or multi-lane road. In this case, the dashed lane lines can be converted into equal-dividing lines, with the solid roadbed lines as the boundary lines, dividing the road area into multiple areas to be cleaned.

[0086] It should be noted that if the distance between two dashed variable lane lines is less than the width of the lane, it will be counted as one dashed variable lane line.

[0087] In practical implementation, for example, for a two-lane road, the dashed variable lane line in the middle and the solid roadbed line are used as the baseline for dividing the road area into two regions, A and B. For a three-lane road, the dashed variable lane line and the solid white lines near the roadbed on both sides are used as the baseline to divide the road into three regions, A, B, and C. Similarly, for a four-lane road, it is divided into four regions, A, B, C, and D.

[0088] S13. Perform garbage identification on the area images corresponding to each area to be cleaned, and determine the amount of garbage in each area.

[0089] After dividing the areas to be cleaned, the images of each area can be further divided according to the areas to be cleaned, so as to identify the garbage in the area images corresponding to each area to be cleaned and determine the amount of garbage in each area to be cleaned.

[0090] Among them, garbage identification can be achieved through the perception module in the autonomous driving system of the unmanned cleaning vehicle, which can be implemented through object recognition algorithms, including but not limited to convolutional neural networks, support vector machines, decision trees, random forests or deep learning models.

[0091] Optionally, the amount of garbage in a region can be represented by the Trash value output by the sensing module.

[0092] Step 203: Calculate the absolute value of the difference in waste between adjacent areas to be cleaned using the amount of waste in each area;

[0093] After identifying the amount of garbage in each area to be cleaned, the absolute value of the difference in garbage between adjacent areas to be cleaned can be determined by subtracting the amount of garbage in adjacent areas to be cleaned and taking the absolute value.

[0094] Taking a two-lane road as an example, if the Trash value of the left lane area is 75 and the Trash value of the right lane area is 15, then the absolute value of the difference in trash between adjacent areas to be cleaned is 60.

[0095] Furthermore, the method also includes the following steps:

[0096] If the absolute value of the garbage difference does not exceed the preset lane change threshold, the unmanned cleaning vehicle will maintain its cleaning route at the current moment through the planning module.

[0097] In one example of the present invention, if the absolute value of the difference in garbage does not exceed the preset lane-changing threshold, it indicates that the amount of garbage in the adjacent lane is similar to that in the current lane. In order to save the operating resources of the unmanned cleaning vehicle, it is not necessary to change lanes for the unmanned cleaning vehicle. The cleaning route of the unmanned cleaning vehicle at the current moment can be maintained by the planning module.

[0098] Step 204: If the absolute value of the garbage difference exceeds the preset lane change threshold, a lane change command is sent to the planning module.

[0099] In another example of the present invention, if the absolute value of the difference in the amount of garbage exceeds the preset lane-changing threshold, it indicates that there is a cleaning area ahead with a large difference in the amount of garbage. At this time, a lane-changing command can be sent to the planning module to call the planning module to adjust the cleaning route of the cleaning area according to the amount of garbage in the area.

[0100] In a specific implementation, the above lane change command can be used to enable the on-demand cleaning operation mode of the planning module. The on-demand cleaning operation mode can include the following steps 205-207.

[0101] Step 205: The planning module allocates the priority of each area to be cleaned according to the amount of garbage in the area, from most to least.

[0102] In this embodiment of the invention, after the planning module receives the lane change instruction, it can allocate the priority of each area to be cleaned according to the amount of garbage in each area, from most to least.

[0103] Taking a three-lane road system (A, B, and C) as an example, the amount of trash in area A to be cleaned is 50, in area B it is 110, and in area C it is 40. The lane-changing threshold is 50. If the unmanned cleaning vehicle is currently in area A, the calculated absolute difference in trash volume between A and B is 60, and between B and C it is 70, both exceeding the lane-changing threshold. Therefore, area B can be assigned the highest priority, area A the medium priority, and area C the lowest priority.

[0104] Step 206: If the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is less than or equal to a preset threshold, the cleaning route of the unmanned cleaning vehicle will be directly adjusted to the highest priority cleaning area through the planning module.

[0105] In this embodiment of the invention, after prioritizing each area to be cleaned, if the number of lanes between the highest priority area to be cleaned and the current area of ​​the unmanned cleaning vehicle is less than or equal to a preset threshold, it indicates that the highest priority area to be cleaned is either an adjacent lane or the current lane. At this time, the cleaning route of the unmanned cleaning vehicle can be directly adjusted to the highest priority area to be cleaned through the planning module, so that the unmanned cleaning vehicle can directly change lanes to the area to be cleaned.

[0106] It should be noted that the preset threshold can be set to 1, and this embodiment of the invention does not impose any restrictions on it.

[0107] Taking a three-lane road as an example, it is divided into areas A, B, and C. The difference between A and B is 60, and the difference between B and C is 70.

[0108] If the current vehicle is in area B, and the number of lanes to change to A and C is 1, then C, which has a higher difference, will be selected (because C has a higher Trash value and a higher priority for cleaning).

[0109] Step 207: If the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is greater than a preset threshold, the cleaning route of the unmanned cleaning vehicle will be continuously planned by the planning module and then adjusted to the highest priority cleaning area.

[0110] Optionally, step 207 may include the following sub-steps:

[0111] The planning module adjusts the cleaning route of the unmanned cleaning vehicle to the adjacent area to be cleaned, and assigns a lane-changing cost to the highest priority area to be cleaned.

[0112] If the lane change cost is less than or equal to the preset cost threshold, the lane change cost will be reduced by the planning module according to the preset running time gradient.

[0113] When the cost of changing lanes equals the preset minimum cost, the cleaning route is adjusted to the highest priority area to be cleaned through the planning module.

[0114] In this embodiment of the invention, if the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is greater than a preset threshold, it indicates that the unmanned cleaning vehicle needs to perform continuous lane changes to reach the highest priority cleaning area. To reduce the risk of continuous lane changes by the unmanned cleaning vehicle at low speeds and minimize the impact on other traffic participants, the cleaning route of the unmanned cleaning vehicle can be adjusted to an adjacent cleaning area through a planning module, and a lane-changing cost can be assigned to the highest priority cleaning area. If the lane-changing cost is less than or equal to a preset cost threshold, the lane-changing cost can be gradually reduced by the planning module according to a preset operating time gradient, i.e., the travel time of the unmanned cleaning vehicle. When the lane-changing cost equals a preset minimum cost value, the cleaning route can be adjusted to the highest priority cleaning area through the planning module to control the unmanned cleaning vehicle to perform a lane change.

[0115] Taking a three-lane road as an example, if the unmanned cleaning vehicle is currently in area A, it will first trigger a lane change to area B, while simultaneously assigning a lane change cost to area C to control the vehicle's continuous lane change behavior within a short period. If the cost is less than a preset cost threshold, this cost will decrease to 0 after the vehicle enters area B and travels for a certain period of time (approximately 5 seconds). Only when this cost drops to 0 will the planning module select area C, which has a higher cost difference, for the lane change.

[0116] Optionally, the method further includes:

[0117] If the cost of changing lanes exceeds the preset cost threshold, the cleaning route will be kept in the area to be cleaned at the current moment through the planning module.

[0118] In this embodiment of the invention, the lane-changing cost can be calculated based on the number of lanes in the road area and the number of lanes between the highest-priority area to be cleaned and the current area of ​​the unmanned cleaning vehicle. For example, if there are four lanes and the leftmost lane has the highest trash value, and the unmanned cleaning vehicle is currently cleaning in the rightmost lane, changing lanes three times consecutively is too risky for a low-speed cleaning vehicle. If the planning module calculates that the cost of consecutive lane changes exceeds a preset threshold, a penalty is generated, preventing consecutive lane changes. The planning module maintains the cleaning route within the current area to be cleaned and pauses the on-demand cleaning operation.

[0119] In another example of the present invention, after performing step 207, the method may further include the following steps:

[0120] Once the cleaning route is adjusted to the highest priority area to be cleaned, the planning module assigns a priority cleaning label to the current area of ​​the unmanned cleaning vehicle in the semantic map;

[0121] The planning module generates updated cleaning routes using priority cleaning markers and determines whether the updated cleaning routes meet the road U-turn conditions.

[0122] If the conditions are met, the unmanned cleaning vehicle will move according to the updated cleaning route through the planning module to complete the cleaning action;

[0123] If the conditions are not met, the coordinates of the area corresponding to the priority cleaning marker are obtained and sent to the communication-associated collaborative unmanned cleaning vehicle.

[0124] Among them, the collaborative unmanned cleaning vehicle has a fixed operating route with coordinates of the area it passes through.

[0125] In this embodiment of the invention, when the cleaning route is adjusted to the highest priority cleaning area, since the unmanned cleaning vehicle may not have finished cleaning the current lane, to ensure the quality of road cleaning, the planning module can assign a priority cleaning marker to the current area of ​​the unmanned cleaning vehicle in the semantic map. This priority cleaning marker, combined with the adjusted cleaning route, is used to re-plan the route to generate an updated cleaning route. Simultaneously, it is determined whether the updated cleaning route meets the road turning-around conditions, i.e., whether the unmanned cleaning vehicle can turn around and return to the area corresponding to the priority cleaning marker given the current battery level or resources. If it does, the planning module can move the unmanned cleaning vehicle according to the updated cleaning route to complete the cleaning action. If it does not meet the conditions, the coordinates of the area corresponding to the priority cleaning marker are obtained and sent to the associated collaborative unmanned cleaning vehicle. The collaborative unmanned cleaning vehicle has a fixed operating route that passes through the area coordinates, thereby realizing a collaborative working method between on-demand cleaning operation mode and fixed operation mode.

[0126] In this embodiment of the invention, the autonomous driving system inside the unmanned cleaning vehicle acquires an image of a pre-defined area in front of the vehicle. The area is mapped onto a pre-defined semantic map, and then divided according to road markings on the semantic map. At least two areas to be cleaned are identified, and the amount of trash in each area is determined. The absolute value of the difference in trash quantity between adjacent areas is calculated using the trash quantity. If the absolute value of the trash difference exceeds a pre-defined lane-changing threshold, the cleaning route of the unmanned cleaning vehicle is adjusted by the planning module based on the amount of trash in each area. Thus, by combining area imagery and object perception, the amount of trash within the pre-defined area is effectively analyzed, enabling flexible adjustment of the cleaning route of the unmanned cleaning vehicle.

[0127] Please see Figure 3 , Figure 3 This is a structural block diagram of a route planning device for an unmanned cleaning vehicle provided in Embodiment 3 of the present invention.

[0128] This invention provides a route planning device for an unmanned cleaning vehicle, comprising:

[0129] The area image acquisition module 301 is used to acquire the area image corresponding to the preset specification area in front of the unmanned cleaning vehicle;

[0130] The area division and waste identification module 302 is used to determine at least two areas to be cleaned based on a preset semantic map and area image, and to identify the amount of waste in each area to be cleaned.

[0131] The garbage difference calculation module 303 is used to calculate the absolute value of the garbage difference between adjacent areas to be cleaned based on the amount of garbage in the area.

[0132] The cleaning route planning module 304 is used to adjust the cleaning route of the unmanned cleaning vehicle according to the amount of garbage in the area if the absolute value of the garbage difference exceeds the preset lane change threshold.

[0133] Optionally, the device further includes:

[0134] The cleaning route maintenance module is used to maintain the cleaning route of the unmanned cleaning vehicle at the current moment through the planning module if the absolute value of the garbage difference does not exceed the preset lane change threshold.

[0135] Optionally, the area division and waste identification module 302 includes:

[0136] The road marking and area positioning submodule is used to locate the road markings and road areas corresponding to the area image from the preset semantic map;

[0137] The area division submodule is used to divide the road area according to the road markings and identify at least two areas to be cleaned;

[0138] The waste identification submodule is used to identify waste in the area images corresponding to each area to be cleaned, and to determine the amount of waste in each area.

[0139] Optionally, road markings include dashed lines for variable lanes and solid lines for the roadbed; the area division submodule is specifically used for:

[0140] Count the number of dashed lane markings within the road area;

[0141] If there is only one dashed line, the road area is divided into two areas to be cleaned, with the dashed variable lane line as the center line and the solid roadbed line as the boundary line.

[0142] If there are two or more dashed lines, the road area is divided into multiple areas to be cleaned, with the variable lane dashed lines as equal dividing lines and the roadbed solid lines as boundary lines.

[0143] Optionally, the cleaning route planning module 304 includes:

[0144] The lane change instruction issuing submodule is used to send a lane change instruction to the planning module if the absolute value of the garbage difference exceeds the preset lane change threshold.

[0145] The priority allocation submodule is used to allocate the priority of each area to be cleaned according to the amount of garbage in the area, from most to least, through the planning module.

[0146] The first route adjustment submodule is used to adjust the cleaning route of the unmanned cleaning vehicle directly to the highest priority cleaning area if the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is less than or equal to a preset threshold.

[0147] The second route adjustment submodule is used to adjust the cleaning route of the unmanned cleaning vehicle to the highest priority cleaning area after the planning module continuously plans lane changes if the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is greater than a preset threshold.

[0148] Optionally, the second route adjustment submodule is specifically used for:

[0149] The planning module adjusts the cleaning route of the unmanned cleaning vehicle to the adjacent area to be cleaned, and assigns a lane-changing cost to the highest priority area to be cleaned.

[0150] If the lane change cost is less than or equal to the preset cost threshold, the lane change cost will be reduced by the planning module according to the preset running time gradient.

[0151] When the cost of changing lanes equals the preset minimum cost, the cleaning route is adjusted to the highest priority area to be cleaned through the planning module.

[0152] Optionally, the second route adjustment submodule is also specifically used for:

[0153] If the cost of changing lanes exceeds the preset cost threshold, the cleaning route will be kept in the area to be cleaned at the current moment through the planning module.

[0154] Optionally, the device further includes:

[0155] The priority cleaning label assignment module is used to assign a priority cleaning label to the current area of ​​the unmanned cleaning vehicle in the semantic map through the planning module after the cleaning route is adjusted to the highest priority area to be cleaned.

[0156] The route update and U-turn judgment module is used to generate an updated cleaning route through the planning module using priority cleaning markers, and to determine whether the updated cleaning route meets the road U-turn conditions.

[0157] The mobile cleaning module is used to move the unmanned cleaning vehicle according to the updated cleaning route through the planning module to complete the cleaning action if the conditions are met.

[0158] The collaborative cleaning module is used to obtain the area coordinates corresponding to the priority cleaning mark if the conditions are not met, and send them to the collaborative unmanned cleaning vehicle with communication association.

[0159] Among them, the collaborative unmanned cleaning vehicle has a fixed operating route with coordinates of the area it passes through.

[0160] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the route planning method for an unmanned cleaning vehicle as described in any embodiment of the present invention.

[0161] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed, implements the route planning method for an unmanned cleaning vehicle as described in any embodiment of this invention.

[0162] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, modules, and sub-modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0163] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0164] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0165] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0166] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0167] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A route planning method for an unmanned cleaning vehicle, characterized in that, include: Acquire an image of the area corresponding to a preset specification area in front of the unmanned cleaning vehicle; Based on the preset semantic map and the area image, at least two areas to be cleaned are determined and the amount of garbage corresponding to each area to be cleaned is identified. The absolute value of the difference in waste between adjacent areas to be cleaned is calculated using the amount of waste in the area. If the absolute value of the garbage difference exceeds the preset lane change threshold, the cleaning route of the unmanned cleaning vehicle will be adjusted by the planning module according to the amount of garbage in the area. The step of adjusting the cleaning route of the unmanned cleaning vehicle according to the amount of garbage in the area by the planning module if the absolute value of the garbage difference exceeds a preset lane-changing threshold includes: If the absolute value of the garbage difference exceeds the preset lane-changing threshold, a lane-changing command is sent to the planning module; The planning module allocates priority to each area to be cleaned based on the amount of garbage in the area, from most to least. If the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is less than or equal to a preset threshold, the cleaning route of the unmanned cleaning vehicle will be directly adjusted to the highest priority cleaning area through the planning module. If the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is greater than a preset threshold, the cleaning route of the unmanned cleaning vehicle will be continuously planned by the planning module and then adjusted to the highest priority cleaning area.

2. The method according to claim 1, characterized in that, The method further includes: If the absolute value of the garbage difference does not exceed the preset lane change threshold, the unmanned cleaning vehicle maintains its cleaning route at the current moment through the planning module.

3. The method according to claim 1, characterized in that, The step of determining at least two areas to be cleaned based on a preset semantic map and the area image, and identifying the amount of garbage corresponding to each area to be cleaned, includes: The road markings and road areas corresponding to the region image are located from the preset semantic map; The road area is divided according to the road markings, and at least two areas to be cleaned are identified. Garbage is identified in the area images corresponding to each of the areas to be cleaned, and the amount of garbage in each area is determined.

4. The method according to claim 3, characterized in that, The road markings include dashed variable lane lines and solid roadbed lines; the step of dividing the road area according to the road markings and determining at least two areas to be cleaned includes: Count the number of dashed lines in the variable lane area within the road region; If there is only one dashed line, then the road area is divided into two areas to be cleaned, with the dashed variable lane line as the center line and the solid roadbed line as the boundary line. If there are two or more dashed lines, the road area is divided into multiple areas to be cleaned, with the variable lane dashed lines as equal dividing lines and the roadbed solid lines as boundary lines.

5. The method according to claim 1, characterized in that, The step of adjusting the cleaning route of the unmanned cleaning vehicle to the highest priority cleaning area after continuous lane-changing planning by the planning module includes: The planning module adjusts the cleaning route of the unmanned cleaning vehicle to the adjacent area to be cleaned, and assigns a lane-changing cost to the highest priority area to be cleaned. If the lane change cost is less than or equal to a preset cost threshold, the lane change cost is reduced by the planning module according to a preset running time gradient. When the lane change cost equals the preset minimum cost, the cleaning route is adjusted to the highest priority area to be cleaned through the planning module.

6. The method according to claim 5, characterized in that, The method further includes: If the cost of changing lanes exceeds a preset cost threshold, the planning module will maintain the cleaning route in the area to be cleaned at the current moment.

7. The method according to claim 1, characterized in that, The method further includes: Once the cleaning route is adjusted to the highest priority area to be cleaned, the planning module assigns a priority cleaning marker to the current area of ​​the unmanned cleaning vehicle in the semantic map. The planning module generates an updated cleaning route using the priority cleaning markers and determines whether the updated cleaning route meets the road U-turn conditions. If the conditions are met, the unmanned cleaning vehicle will move according to the updated cleaning route via the planning module to complete the cleaning action. If the conditions are not met, the area coordinates corresponding to the priority cleaning identifier are obtained and sent to the communication-associated collaborative unmanned cleaning vehicle. The collaborative unmanned cleaning vehicle is equipped with a fixed operating route that passes through the coordinates of the area.

8. A route planning device for an unmanned cleaning vehicle, characterized in that, include: The area image acquisition module is used to acquire area images corresponding to a preset size area in front of the unmanned cleaning vehicle; The area division and waste identification module is used to determine at least two areas to be cleaned based on a preset semantic map and the area image, and to identify the amount of waste in each area to be cleaned. The garbage difference calculation module is used to calculate the absolute value of the garbage difference between adjacent areas to be cleaned, based on the amount of garbage in the area. A cleaning route planning module is used to adjust the cleaning route of the unmanned cleaning vehicle according to the amount of garbage in the area if the absolute value of the garbage difference exceeds a preset lane change threshold. The cleaning route planning module includes: The lane change instruction issuing submodule is used to send a lane change instruction to the planning module if the absolute value of the garbage difference exceeds a preset lane change threshold. The priority allocation submodule is used to allocate the priority of each of the areas to be cleaned from the most to the least according to the amount of garbage in the area through the planning module. The first route adjustment submodule is used to adjust the cleaning route of the unmanned cleaning vehicle directly to the highest priority cleaning area through the planning module if the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is less than or equal to a preset threshold. The second route adjustment submodule is used to adjust the cleaning route of the unmanned cleaning vehicle to the highest priority cleaning area after the planning module continuously plans lane changes if the number of lanes between the highest priority cleaning area and the current area of ​​the unmanned cleaning vehicle is greater than a preset threshold.

9. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the steps of the route planning method for the unmanned cleaning vehicle as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the route planning method for the unmanned cleaning vehicle as described in any one of claims 1-7.

Images