A method, device, equipment and medium for adjusting the cleaning speed of a sanitation vehicle

Abstract

The application discloses a kind of sanitation vehicle's cleaning speed adjustment method, device, equipment and medium, method includes by the automatic driving system of automatic driving sanitation vehicle carried on its front area corresponding real-time image is obtained, according to real-time image and semantic map combination judge whether there is accumulation obstacle in front area;If there is, then the driving speed of the automatic driving sanitation vehicle corresponding is reduced to the accumulation cleaning speed, and the road surface grid area of preset specification is constructed, and the obstacle point cloud corresponding to the accumulation obstacle is continuously detected, according to the point cloud dynamic cumulative value corresponding to each grid in the road surface grid area, the comparison result of combination preset accumulation threshold value, the rotation speed of the automatic driving sanitation vehicle corresponding is adjusted, so that the way of point cloud combination perception image is adjusted to the rotation speed of the automatic driving sanitation vehicle, improves the cleaning quality of the automatic driving sanitation vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent cleaning, in particular to a cleaning speed adjustment method, device and equipment of a sanitation vehicle and a medium. BACKGROUND

[0002] With the acceleration of urbanization, the road garbage of the city is also increasing, and the sanitation industry is facing great pressure. However, many urban roads still rely on sanitation workers to use brooms for cleaning. This working method is extremely inefficient and requires a large amount of labor. At present, the intelligent level of road cleaning vehicles is still relatively low, and a large amount of manpower is still needed to participate in cleaning operations, which also increases the burden of enterprises.

[0003] In order to reduce the workload of sanitation workers and reduce the economic burden of enterprises, many cleaning vehicle production enterprises at home and abroad have begun to develop autonomous sanitation vehicles to enable cleaning vehicles to work more efficiently.

[0004] However, the current autonomous sanitation vehicle usually maintains a constant speed during the edge cleaning operation, which is prone to cause garbage and leaves to splash onto the road edge when encountering garbage and leaf accumulation, resulting in poor cleaning effect. SUMMARY

[0005] The present application provides a sanitation vehicle cleaning speed adjustment method, device, equipment and medium, which solves the technical problem that the current autonomous sanitation vehicle usually maintains a constant speed during the edge cleaning operation, which is prone to cause garbage and leaves to splash onto the road edge when encountering garbage and leaf accumulation, resulting in poor cleaning effect.

[0006] The first aspect of the present application provides a sanitation vehicle cleaning speed adjustment method, comprising:

[0007] Obtain the real-time image corresponding to the front area of the autonomous sanitation vehicle;

[0008] According to the real-time image and the semantic map, it is judged whether the front area exists accumulated obstacles or not;

[0009] If so, the driving speed of the autonomous sanitation vehicle is reduced to the accumulated cleaning speed;

[0010] Construct a road surface grid area of a predetermined specification, and continuously detect the obstacle point cloud corresponding to the accumulated obstacles;

[0011] According to the road surface grid area and the obstacle point cloud, the rotating speed of the brush corresponding to the autonomous sanitation vehicle is adjusted.

[0012] Optionally, the step of determining whether the front area has accumulated obstacles according to the real-time image and the semantic map comprises:

[0013] detecting whether there is a fine obstacle in the real-time image;

[0014] if so, positioning the fine obstacle and determining whether the fine obstacle is in the to-be-cleaned track corresponding to the autonomous driving sanitation vehicle according to the semantic map;

[0015] if so, determining that the front area has accumulated obstacles;

[0016] if not or if the fine obstacle is not detected, determining that the front area does not have accumulated obstacles.

[0017] Optionally, the step of determining whether the fine obstacle is in the to-be-cleaned track corresponding to the autonomous driving sanitation vehicle if so comprises:

[0018] if a fine obstacle is detected in the real-time image, positioning the fine obstacle and determining the position of the fine obstacle;

[0019] dividing the front area according to the lane lines in the semantic map to obtain a plurality of to-be-cleaned areas

[0020] determining whether the position of the fine obstacle and the autonomous driving sanitation vehicle are in the same to-be-cleaned area;

[0021] if so, determining that the fine obstacle is in the to-be-cleaned track corresponding to the autonomous driving sanitation vehicle;

[0022] if not, determining that the fine obstacle is not in the to-be-cleaned track corresponding to the autonomous driving sanitation vehicle.

[0023] Optionally, the steps of constructing a road grid area of a preset specification and continuously detecting obstacle point clouds corresponding to the accumulated obstacles comprise:

[0024] calling a camera component to construct a detection area according to a preset distance;

[0025] dividing the detection area according to a preset specification to generate a road grid area composed of a plurality of grids;

[0026] continuously detecting obstacle point clouds corresponding to the accumulated obstacles by calling a laser radar.

[0027] Optionally, the step of adjusting the brush rotation speed corresponding to the autonomous driving sanitation vehicle according to the road grid area and the obstacle point clouds comprises:

[0028] mapping the obstacle point cloud to each grid in the road surface grid area;

[0029] respectively comparing the point cloud dynamic cumulative value of the obstacle point cloud in each grid with a preset accumulation threshold value;

[0030] if there is a pending grid whose point cloud dynamic cumulative value is greater than the preset accumulation threshold value within a first preset time period, a speed reduction instruction is generated and sent to a planning control component;

[0031] by the planning control component responding to the speed reduction instruction, the corresponding brush rotation speed of the autonomous sanitation vehicle is reduced to an accumulation cleaning rotation speed.

[0032] Optionally, if there is a pending grid whose point cloud dynamic cumulative value is greater than the preset accumulation threshold value within a first preset time period, a speed reduction instruction is generated and sent to a planning control component, the step comprises:

[0033] if there is a pending grid whose point cloud dynamic cumulative value is greater than the preset accumulation threshold value within a first preset time period, it is judged whether the number of the pending grid is greater than or equal to a preset grid threshold value;

[0034] if yes, a speed reduction instruction is generated and sent to a planning control component;

[0035] if no, the step of respectively comparing the point cloud dynamic cumulative value of the obstacle point cloud in each grid with a preset accumulation threshold value is executed.

[0036] Optionally, the method further comprises:

[0037] if there is no pending grid whose point cloud dynamic cumulative value is greater than the preset accumulation threshold value within a first preset time period, the step of respectively comparing the point cloud dynamic cumulative value of the obstacle point cloud in each grid with a preset accumulation threshold value is executed;

[0038] if the point cloud dynamic cumulative value is less than or equal to the preset accumulation threshold value and lasts for a second preset time period, or it is determined that the front area does not exist an accumulation obstacle, a speed-up instruction is generated and sent to a planning control component;

[0039] by the planning control component, the corresponding brush rotation speed of the autonomous sanitation vehicle is increased to a regular operation rotation speed, and the driving speed is increased to a regular operation speed.

[0040] The second aspect of the application provides a sanitation vehicle cleaning speed adjustment device, comprising:

[0041] a real-time image acquisition module for acquiring real-time images corresponding to the front area of the autonomous sanitation vehicle;

[0042] A pile-up obstacle judgment module is configured to judge whether the front area has a pile-up obstacle according to the real-time image and the semantic map.

[0043] A vehicle speed adjustment module is configured to reduce the driving vehicle speed of the autonomous sanitation vehicle to a pile-up cleaning vehicle speed if the pile-up obstacle exists.

[0044] A grid and point cloud creation module is configured to construct a road surface grid area of a preset specification and continuously detect an obstacle point cloud corresponding to the pile-up obstacle.

[0045] A rotation speed adjustment module is configured to adjust the brush rotation speed of the autonomous sanitation vehicle according to the road surface grid area and the obstacle point cloud.

[0046] The third aspect of the present application provides an electronic device including a memory and a processor, the memory storing a computer program, and the computer program being executed by the processor to make the processor execute the steps of the sanitation vehicle cleaning speed adjustment method according to any one of the first aspect of the present application.

[0047] The fourth aspect of the present application provides a computer readable storage medium storing a computer program, and the computer program being executed to implement the sanitation vehicle cleaning speed adjustment method according to any one of the first aspect of the present application.

[0048] As can be seen from the above technical solutions, the present application has the following advantages:

[0049] The autonomous sanitation vehicle carries an autonomous driving system to obtain a real-time image corresponding to a front area, and judges whether the front area has a pile-up obstacle according to the real-time image and a semantic map. If the pile-up obstacle exists, the driving vehicle speed of the autonomous sanitation vehicle is reduced to a pile-up cleaning vehicle speed, a road surface grid area of a preset specification is constructed, and an obstacle point cloud corresponding to the pile-up obstacle is continuously detected. According to the dynamic cumulative value of the point cloud corresponding to each grid in the road surface grid area and the comparison result of the preset pile-up threshold, the brush rotation speed of the autonomous sanitation vehicle is adjusted. Thus, the brush and vehicle speed of the autonomous sanitation vehicle are adjusted by means of point cloud and perception image, and the cleaning quality of the autonomous sanitation vehicle is improved. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0051] Figure 1A step flow chart of a sweeping speed adjustment method of a sanitation vehicle provided for the first embodiment of the present application is shown in the figure.

[0052] Figure 2 A step flow chart of a sweeping speed adjustment method of a sanitation vehicle provided for the second embodiment of the present application is shown in the figure.

[0053] Figure 3 A structure block diagram of a sweeping speed adjustment device of a sanitation vehicle provided for the third embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0054] The embodiments of the present application provide a sanitation vehicle sweeping speed adjustment method, device, equipment and medium, and are used for solving the technical problem that the sweeping brush of the automatic driving sanitation vehicle usually keeps constant rotating speed in the edge cleaning operation process at present, and the garbage and leaves are easily splashed onto the road edge when encountering garbage and leaf accumulation, resulting in poor cleaning effect.

[0055] In order to make the invention purpose, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the following described embodiments are only part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0056] Please refer to Figure 1 , Figure 1 A step flow chart of a sanitation vehicle sweeping speed adjustment method provided for the first embodiment of the present application is shown in the figure.

[0057] The sanitation vehicle sweeping speed adjustment method provided by the present application comprises:

[0058] Step 101, acquiring real-time images corresponding to the front area of the automatic driving sanitation vehicle;

[0059] The real-time image refers to the area image of the automatic driving sanitation vehicle in a certain range of the front area, which can be acquired through the wide-angle camera and fisheye camera arranged on the automatic driving sanitation vehicle. For example, the real-time image in the front 10-15m range of the automatic driving sanitation vehicle, which includes but is not limited to the road surface, the road edge and various obstacles on the road surface such as garbage and leaves.

[0060] It should be noted that the automatic driving sanitation vehicle can be equipped with an automatic driving system, and is provided with a laser radar and a camera assembly in communication connection with the automatic driving system. The automatic driving system includes a planning control component for adjusting the driving speed and brush rotation speed of the vehicle. The camera assembly includes, but is not limited to, a fisheye camera and a wide-angle camera, and the like perception component. The brush rotation speed can be divided into 6000 rpm, 10000 rpm and 12000 rpm corresponding to three operation modes of cleaning, regular cleaning and intensive cleaning.

[0061] In the embodiment of the present application, the method can be applied to the automatic driving system in the automatic driving sanitation vehicle. The wide-angle camera and the fisheye camera provided on the automatic driving sanitation vehicle are called by the automatic driving system to obtain real-time images corresponding to the front area of the automatic driving sanitation vehicle, so as to provide data basis for subsequent cleaning speed adjustment.

[0062] In step 102, whether there is a stacked obstacle in the front area is judged according to the real-time image and the semantic map.

[0063] The semantic map refers to a high-precision map containing various semantic information. The point cloud information of the physical world is obtained by the laser radar, and is finely restored, so that the lanes, cars, medians, roadside trees, signs, blue sky and many other different objects and concepts in the road can be distinguished. Semantic information refers to multi-level, rich-dimension information that can enable the unmanned vehicle to better understand the driving rules, perceive the road traffic conditions and plan the driving route, and is covered in the high-precision map.

[0064] In the embodiment of the present application, the object is located in the real-time image to determine whether there is a fine obstacle in the real-time image and to locate it. Then, the front area is divided according to the lane line and other route marks in the semantic map, and whether there is a stacked obstacle in the front area is judged in combination with the location of the fine obstacle.

[0065] It should be noted that the stacked obstacle refers to the fine obstacle existing in the front area and in the same lane or the same cleaning area as the to-be-cleaned track of the automatic driving sanitation vehicle. The fine obstacle refers to the stacked objects such as fallen leaves, tree leaves or road garbage.

[0066] In step 103, if there is, the driving speed of the automatic driving sanitation vehicle is reduced to the stacked cleaning speed.

[0067] If it is detected in real time that there is a stacked obstacle in the front area, it indicates that the automatic driving sanitation vehicle needs to switch modes to clean the front area at low speed. At this time, the planning control component can be used to adjust the driving speed of the automatic driving sanitation vehicle to the stacked cleaning speed.

[0068] It should be noted that the automatic driving system can set multiple driving speeds at the same time to adapt to different cleaning modes. For example, the normal working speed of the vehicle can be set to 7 km / h, and the accumulation cleaning speed can be set to 50% of the normal working speed, such as 3.5 km / h.

[0069] In step 104, a road grid area of a preset specification is constructed, and the obstacle point cloud corresponding to the accumulation obstacle is continuously detected.

[0070] After the driving speed of the automatic driving sanitation vehicle is reduced to the accumulation cleaning speed, the fisheye camera can be started to identify the road segmentation. The road in front of the vehicle within a range of 2-3 m is obtained by the fisheye camera, and is divided into several grid cells of the same size to construct a road grid area of a preset specification.

[0071] At the same time, the automatic driving system can also call the laser radar to continuously detect the obstacle point cloud corresponding to the accumulation obstacle to obtain the data basis for adjusting the brush rotation speed of the subsequent automatic driving sanitation vehicle.

[0072] In step 105, the brush rotation speed of the automatic driving sanitation vehicle is adjusted according to the road grid area and the obstacle point cloud.

[0073] When the laser radar detects the obstacle point cloud, a seg-fusion value is accumulated. The seg-fusion value is a dynamic value accumulated according to the laser radar point cloud fusion.

[0074] After the obstacle point cloud is detected in real time, it can be mapped to the road grid area to give the depth and height of each pixel. The actual point cloud dynamic accumulation value of each grid is counted in the road grid area, and the point cloud dynamic accumulation value and the preset accumulation threshold of the road grid area are compared to adjust the brush rotation speed of the automatic driving sanitation vehicle.

[0075] It should be noted that the automatic driving sanitation vehicle usually cleans the road edge by setting the brush itself. When passing through the garbage pile in the prior art, since the speed of the vehicle and the rotation speed of the brush are not reduced to a suitable level, the leaves and garbage will splash onto the road edge, which will cause the road edge to be unable to be cleaned or the garbage to be blown to the side of the lane, and the cleaning effect will be greatly reduced. More resources will be wasted for re-cleaning. At this time, the vehicle driving speed can be reduced while the brush rotation speed is adjusted to improve the cleaning quality of the automatic driving sanitation vehicle.

[0076] In the embodiment of the present application, the real-time image corresponding to the front area of the automatic driving sanitation vehicle is obtained by the automatic driving system mounted on the automatic driving sanitation vehicle, and it is judged whether there is a stacked obstacle in the front area according to the real-time image and the semantic map; if there is, the driving speed of the automatic driving sanitation vehicle is reduced to the stacking cleaning speed, a road grid area of a preset specification is constructed, and the obstacle point cloud corresponding to the stacked obstacle is continuously detected, the point cloud dynamically accumulated value of each grid in the road grid area is obtained, and the scanning speed of the automatic driving sanitation vehicle is adjusted according to the comparison result of the preset stacking threshold, so that the scanning and speed of the automatic driving sanitation vehicle are adjusted by the point cloud combined with the perception image, and the cleaning quality of the automatic driving sanitation vehicle is improved.

[0077] Please refer to Figure 2 , Figure 2 The step flow chart of the sanitation vehicle cleaning speed adjustment method provided in the second embodiment of the present application.

[0078] The sanitation vehicle cleaning speed adjustment method provided by the present application comprises:

[0079] Step 201, obtaining the real-time image corresponding to the front area of the automatic driving sanitation vehicle;

[0080] In the embodiment of the present application, the specific implementation process of step 201 is similar to that of step 101, which will not be repeated here.

[0081] Step 202, judging whether there is a stacked obstacle in the front area according to the real-time image and the semantic map;

[0082] Optionally, step 202 can include the following sub-steps S11-S14:

[0083] S11, detecting whether there is a fine obstacle in the real-time image;

[0084] S12, if there is, positioning the fine obstacle, and judging whether the fine obstacle is in the to-be-cleaned track corresponding to the automatic driving sanitation vehicle according to the semantic map;

[0085] In the embodiment of the present application, the image recognition algorithm can be used to position the fine obstacle in the real-time image to determine whether there is a fine obstacle in it.

[0086] For example, the position of the fine obstacle in the real-time image is identified in the form of surrounding the fine obstacle by a bounding box, and the image recognition algorithm can be realized by support vector machine, convolutional neural network or RCNN, etc., which is not limited in the embodiment of the present application.

[0087] If it is determined that there is a fine obstacle in the real-time image, the fine obstacle can be further positioned, and whether the position of the fine obstacle is on the corresponding cleaning track of the autonomous sanitation vehicle is determined according to the semantic map.

[0088] Further, the step S12 can include the following sub-steps:

[0089] If the fine obstacle is detected in the real-time image, the fine obstacle is positioned, and the position of the fine obstacle is determined.

[0090] The front area is divided according to the lane line in the semantic map to obtain a plurality of cleaning areas.

[0091] Whether the position of the fine obstacle and the autonomous sanitation vehicle are in the same cleaning area is determined.

[0092] If yes, it is determined that the fine obstacle is on the corresponding cleaning track of the autonomous sanitation vehicle.

[0093] If no, it is determined that the fine obstacle is not on the corresponding cleaning track of the autonomous sanitation vehicle.

[0094] In the embodiment of the application, if the fine obstacle is detected in the real-time image, the fine obstacle can be further positioned by an image frame or other identification method, and the position of the fine obstacle, such as the road area where the fine obstacle is located, is determined.

[0095] At the same time, the front area can be divided according to the lane line identified in the semantic map, and the front area can be divided by a wide-angle camera combined with the lane line to obtain a plurality of divided cleaning areas. It is further determined whether the position of the fine obstacle and the position of the autonomous sanitation vehicle are in the same cleaning area. If yes, it is determined that the fine obstacle is on the corresponding cleaning track of the autonomous sanitation vehicle, and belongs to the cleaning range of the autonomous sanitation vehicle. If not, it is indicated that the autonomous sanitation vehicle does not pass through the fine obstacle according to the current cleaning plan, and it is determined that the fine obstacle is not on the corresponding cleaning track of the autonomous sanitation vehicle.

[0096] S13, if yes, it is determined that the front area has accumulated obstacles;

[0097] S14, if no or no fine obstacle is detected, it is determined that the front area has no accumulated obstacles.

[0098] In the embodiment, if it is determined that the fine obstacle is on the corresponding cleaning track of the autonomous sanitation vehicle, it is determined that the front area has accumulated obstacles. If the fine obstacle is not on the cleaning track, or no fine obstacle is detected, it is determined that the front area has no accumulated obstacles.

[0099] Step 203, if there is, the corresponding driving speed of the autonomous sanitation vehicle is reduced to the accumulation cleaning speed;

[0100] In the embodiment of the application, the specific implementation process of step 203 is similar to step 103, and will not be repeated again.

[0101] Step 204, constructing a road surface grid area of a preset specification, and continuously detecting the obstacle point cloud corresponding to the accumulation obstacle;

[0102] Optionally, step 204 can include the following sub-steps:

[0103] Calling the camera component to construct a detection area according to a preset distance;

[0104] Grid dividing the detection area according to a preset specification to generate a road surface grid area composed of multiple grids;

[0105] Calling the laser radar to continuously detect the obstacle point cloud corresponding to the accumulation obstacle.

[0106] The camera component includes but is not limited to a fisheye camera and a wide-angle camera.

[0107] After the autonomous sanitation vehicle is reduced to the accumulation cleaning speed, the camera component can be called to construct a detection area according to a preset distance, such as 2-3m, and the detection area is grid divided according to a preset specification, thereby generating a road surface grid area composed of multiple grids.

[0108] At the same time, the autonomous driving system can also call the laser radar to continuously detect the obstacle point cloud corresponding to the accumulation obstacle to accumulate a seg-fusion value, which is a dynamic value accumulated according to laser radar point cloud fusion.

[0109] Step 205, mapping the obstacle point cloud to each grid in the road surface grid area;

[0110] In this embodiment, while the laser radar continuously detects the obstacle point cloud, it can also be mapped to each grid in the road surface grid area to assign the seg-fusion value, i.e., the point cloud dynamic accumulation value, accumulated by it to each grid.

[0111] Step 206, respectively comparing the point cloud dynamic accumulation value of the obstacle point cloud in each grid and the preset accumulation threshold;

[0112] And when the road surface has a leaf or garbage pile formed, the seg-fusion value in the grid will break through the upper limit of the threshold of the grid, such as 50%, in turn, at this time, the dynamic accumulation value of the point cloud in each grid and the preset accumulation threshold can be compared respectively to determine the point cloud dynamic accumulation in each grid.

[0113] In step 207, if there is a pending grid whose point cloud dynamic accumulation value is greater than the preset accumulation threshold within the first preset time length, a speed reduction instruction is generated and sent to the planning control component.

[0114] In an example of the present application, the point cloud dynamic accumulation value of a single grid can be used as the generation judgment standard of the speed reduction instruction. To prevent false positives, when the point cloud dynamic accumulation value is greater than the preset accumulation threshold, it can be further judged whether the duration of the situation meets the first preset time length. That is, the point cloud dynamic accumulation value is greater than the preset accumulation threshold within the first preset time length, at this time, the grid is determined as a pending grid, and a speed reduction instruction is generated and sent to the planning control component.

[0115] It should be noted that the preset accumulation threshold can be set to 50%. If the point cloud dynamic accumulation value in a certain grid is higher than 50% within the first preset time length, such as 3 seconds, it indicates that there is a high possibility of garbage or leaf accumulation in the area passed through by the grid.

[0116] Optionally, step 207 can include the following sub-steps:

[0117] If there is a pending grid whose point cloud dynamic accumulation value is greater than the preset accumulation threshold within the first preset time length, it is judged whether the number of pending grids is greater than or equal to the preset grid threshold.

[0118] If yes, a speed reduction instruction is generated and sent to the planning control component.

[0119] If no, the step of comparing the point cloud dynamic accumulation value of the obstacle point cloud in each grid with the preset accumulation threshold is executed.

[0120] In the embodiment of the present application, if there is a pending grid whose point cloud dynamic accumulation value is greater than the preset accumulation threshold within the first preset time length, to further improve the detection accuracy and prevent false positives caused by sudden increase of seg-fusion value in 1-2 local grids, it can be further judged whether the number of pending grids is greater than or equal to the preset grid threshold, such as 5. If yes, it indicates that the current judgment is accurate, a speed reduction instruction can be generated and sent to the planning control component. If no, it indicates that it may be a false positive, at this time, step 206 can be executed to loop to detect the point cloud dynamic accumulation in the road grid area until the autonomous sanitation vehicle completes the current road cleaning plan.

[0121] Optionally, when judging whether the number of pending grids is greater than or equal to the preset grid threshold, the positions of the pending grids can be adjacent or can not be adjacent.

[0122] For example, the fisheye camera divides the front road surface into 10 grid meshes of the same size, and only when the threshold that the seg-fusion value in the same mesh is higher than 50% for 3 seconds and 5 meshes are simultaneously higher than 50% is met, the condition of reducing the rotation speed of the sweeper is triggered, and the speed reduction instruction is issued to the planning control component.

[0123] In step 208, the planning control component responds to the speed reduction instruction to reduce the rotation speed of the corresponding sweeper of the autonomous sanitation vehicle to the accumulation cleaning rotation speed.

[0124] In the embodiment of the application, when the planning control component receives the speed reduction instruction, the planning control component adjusts the rotation speed of the corresponding sweeper of the autonomous sanitation vehicle and reduces it to the accumulation cleaning rotation speed.

[0125] In a specific implementation, the upstream perception input is given to the planning control component, and an instruction is automatically triggered and sent to the planning control component to reduce the rotation speed of the sweeper from 10,000 rpm of the regular cleaning rotation speed to 6,000 rpm. The reduction of the vehicle speed in coordination with the reduction of the rotation speed of the sweeper can effectively reduce the situation of flying leaves and garbage.

[0126] Optionally, the method can further include the following steps S21-S23:

[0127] S21, if there is no pending grid in which the point cloud dynamic accumulation value is greater than the preset accumulation threshold within the first preset time length, then jump to the step of comparing the point cloud dynamic accumulation value of the obstacle point cloud in each grid with the preset accumulation threshold, respectively;

[0128] S22, if the point cloud dynamic accumulation value is less than or equal to the preset accumulation threshold and lasts for a second preset time length, or it is determined that there is no accumulated obstacle in the front area, then a speed-up instruction is generated and issued to the planning control component;

[0129] S23, the planning control component increases the rotation speed of the corresponding sweeper of the autonomous sanitation vehicle to the regular operation rotation speed, and increases the driving speed to the regular operation speed.

[0130] In another example of the application, if there is no pending grid in which the point cloud dynamic accumulation value is greater than the preset accumulation threshold within the first preset time length, then jump to step 206 to continuously compare the point cloud dynamic accumulation value of the obstacle point cloud in each grid with the preset accumulation threshold in real time until the autonomous sanitation vehicle completes the cleaning plan or the autonomous driving system is turned off.

[0131] If it is determined that there is no accumulation obstacle in the front area, or the point cloud dynamic accumulation value is less than or equal to the preset accumulation threshold and lasts for a second preset time length, it indicates that there is no accumulation obstacle in the front area of the autonomous driving sanitation vehicle, and in order to improve the cleaning efficiency, a speed-up instruction can be generated and sent to the planning control component, and the corresponding brush rotating speed of the autonomous driving sanitation vehicle is increased to the normal working rotating speed by the planning control component, and the driving speed is increased to the normal working speed.

[0132] In a specific implementation, the wide-angle camera determines that there is no garbage accumulation or the garbage is not on the track of the vehicle at a distance of 10-15 m, or the seg-fusion in the current fisheye camera segmented road grid is reduced to below 50% and lasts for more than 3 s, and then it is determined that there is no accumulation of leaves and garbage on the road (relatively dispersed). If the seg-fusion is below the 50% threshold for 3 s, a condition is triggered to return an instruction to the planning control component control dbw. The vehicle speed can be increased to the normal cleaning speed of 7 km / h, and the brush rotating speed can be increased to 10,000 rpm. In this way, the cleaning quality and efficiency are guaranteed, and the situation that leaves and garbage are splashed onto the roadbed and other lanes, causing the machine to be unable to clean completely and requiring manual or machine re-cleaning, is reduced.

[0133] In the embodiment of the application, the automatic driving system carried on the autonomous driving sanitation vehicle is used to obtain real-time images corresponding to the front area, and whether there is an accumulation obstacle in the front area is determined according to the real-time images and a semantic map; if there is, the driving speed of the autonomous driving sanitation vehicle is reduced to an accumulation cleaning speed, a road grid area of a preset specification is constructed, and an obstacle point cloud corresponding to the accumulation obstacle is continuously detected, the brush rotating speed of the autonomous driving sanitation vehicle is adjusted according to the comparison result of the point cloud dynamic accumulation value of each grid in the road grid area and the preset accumulation threshold, so that the brush and the speed of the autonomous driving sanitation vehicle are adjusted by combining the point cloud and the perception image, and the cleaning quality of the autonomous driving sanitation vehicle is improved.

[0134] Please refer to Figure 3 , Figure 3 A structure block diagram of a sanitation vehicle cleaning speed adjustment device is provided for the third embodiment of the application.

[0135] The sanitation vehicle cleaning speed adjustment device provided by the embodiment of the application comprises:

[0136] The real-time image acquisition module 301 is configured to acquire real-time images corresponding to the front area of the autonomous driving sanitation vehicle.

[0137] The accumulation obstacle determination module 302 is configured to determine whether there is an accumulation obstacle in the front area according to the real-time images and a semantic map.

[0138] The vehicle speed adjustment module 303 is configured to reduce the driving speed of the autonomous sanitation vehicle to the accumulation cleaning speed if the accumulation obstacle exists.

[0139] The grid and point cloud creation module 304 is configured to construct a road grid area of a preset specification, and continuously detect an obstacle point cloud corresponding to the accumulation obstacle.

[0140] The rotation speed adjustment module 305 is configured to adjust the rotation speed of the autonomous sanitation vehicle according to the road grid area and the obstacle point cloud.

[0141] Optionally, the accumulation obstacle judgment module 302 comprises:

[0142] The fine obstacle detection submodule is configured to detect whether a fine obstacle exists in the real-time image.

[0143] The trajectory overlap judgment submodule is configured to position the fine obstacle if the fine obstacle exists, and determine whether the fine obstacle is in the to-be-cleaned trajectory of the autonomous sanitation vehicle according to the semantic map.

[0144] The accumulation obstacle existence determination submodule is configured to determine that the front area has the accumulation obstacle if the fine obstacle is in the to-be-cleaned trajectory.

[0145] The accumulation obstacle non-existence determination submodule is configured to determine that the front area does not have the accumulation obstacle if the fine obstacle is not in the to-be-cleaned trajectory or the fine obstacle is not detected.

[0146] Optionally, the trajectory overlap judgment submodule is specifically configured to:

[0147] If the fine obstacle is detected in the real-time image, the fine obstacle is positioned to determine the position of the fine obstacle.

[0148] The front area is divided according to the lane line in the semantic map to obtain a plurality of to-be-cleaned areas.

[0149] It is determined whether the position of the fine obstacle and the autonomous sanitation vehicle are in the same to-be-cleaned area.

[0150] If yes, it is determined that the fine obstacle is in the to-be-cleaned trajectory of the autonomous sanitation vehicle.

[0151] If no, it is determined that the fine obstacle is not in the to-be-cleaned trajectory of the autonomous sanitation vehicle.

[0152] Optionally, the grid and point cloud creation module 304 is specifically configured to:

[0153] The camera component is called to construct a detection area according to a preset distance.

[0154] The detection area is divided into a road grid area composed of a plurality of grids according to a preset specification.

[0155] The laser radar is called to continuously detect the obstacle point cloud corresponding to the accumulated obstacle.

[0156] Optionally, the rotating speed adjustment module 305 comprises:

[0157] a point cloud mapping submodule, configured to map the obstacle point cloud to each grid in the road surface grid area;

[0158] a comparison submodule, configured to compare the point cloud dynamic cumulative value of the obstacle point cloud in each grid with the preset accumulation threshold value respectively;

[0159] a speed reduction instruction generation submodule, configured to generate a speed reduction instruction and send the speed reduction instruction to the planning control component if there is a pending grid whose point cloud dynamic cumulative value is greater than the preset accumulation threshold value within the first preset time length;

[0160] a rotating speed reduction submodule, configured to reduce the rotating speed of the automatic driving sanitation vehicle to the accumulated cleaning rotating speed through the planning control component in response to the speed reduction instruction.

[0161] Optionally, the speed reduction instruction generation submodule is specifically configured to:

[0162] if there is a pending grid whose point cloud dynamic cumulative value is greater than the preset accumulation threshold value within the first preset time length, determine whether the number of the pending grid is greater than or equal to a preset grid threshold value;

[0163] if yes, generate a speed reduction instruction and send the speed reduction instruction to the planning control component;

[0164] if no, jump to the step of comparing the point cloud dynamic cumulative value of the obstacle point cloud in each grid with the preset accumulation threshold value respectively.

[0165] Optionally, the rotating speed adjustment module 305 further comprises:

[0166] a jump submodule, configured to, if there is no pending grid whose point cloud dynamic cumulative value is greater than the preset accumulation threshold value within the first preset time length, jump to the step of comparing the point cloud dynamic cumulative value of the obstacle point cloud in each grid with the preset accumulation threshold value respectively;

[0167] a speed increase instruction generation submodule, configured to, if the point cloud dynamic cumulative value is less than or equal to the preset accumulation threshold value and lasts for a second preset time length, or it is determined that there is no accumulated obstacle in the front area, generate a speed increase instruction and send the speed increase instruction to the planning control component;

[0168] a speed increase submodule, configured to increase the rotating speed of the automatic driving sanitation vehicle to the normal working rotating speed and increase the driving speed to the normal working speed through the planning control component.

[0169] The embodiment of the present application provides a kind of electronic equipment, including memory and processor, the computer program is stored in the memory, the computer program is executed by the processor, so that the processor executes the steps of the sweeping speed adjustment method of the sanitation vehicle as any embodiment of the present application described.

[0170] The embodiment of the present application provides a kind of computer readable storage medium, which stores computer program, the computer program is executed and realizes the sweeping speed adjustment method of the sanitation vehicle as any embodiment of the present application described.

[0171] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device, module and sub-module can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

[0172] In several embodiments provided by the present application, it should be understood that the disclosed device and method can be implemented by other manners.For example, the above-described device embodiment is only schematic, for example, the division of the unit is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed.In addition, the coupling or direct coupling or communication connection between the shown or discussed units can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0173] The unit described as a separate component can be or can not be physically separated, and the component shown as a unit can be or can not be a physical unit, that is, it can be located in one place, or it can be distributed to a plurality of network units.According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment scheme.

[0174] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.The above integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0175] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0176] The above-described and above-embodied examples are only used to illustrate the technical solutions of the present application, rather than limit the same. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features thereof. Such modifications or replacements 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 application.

Claims

1. A method of adjusting a cleaning speed of a sanitation vehicle, characterized by, The method comprises the following steps: acquiring real-time images corresponding to a front area of an autonomous sanitation vehicle; judging whether there is a piled obstacle in the front area according to the real-time images and a semantic map; if there is, reducing a driving speed of the autonomous sanitation vehicle to a piled cleaning speed; constructing a road surface grid area of a preset specification, and continuously detecting obstacle point clouds corresponding to the piled obstacle; adjusting a brush rotating speed of the autonomous sanitation vehicle according to the road surface grid area and the obstacle point clouds; the step of adjusting the brush rotating speed of the autonomous sanitation vehicle according to the road surface grid area and the obstacle point clouds comprises: mapping the obstacle point clouds to each grid in the road surface grid area; respectively comparing a point cloud dynamic accumulation value of the obstacle point clouds in each grid with a preset piled threshold value; if there is a pending grid in which the point cloud dynamic accumulation value is greater than the preset piled threshold value within a first preset time length, generating a speed reduction instruction and delivering the speed reduction instruction to a planning control component; reducing the brush rotating speed of the autonomous sanitation vehicle to a piled cleaning rotating speed through the planning control component in response to the speed reduction instruction.

2. The method of claim 1, wherein, the step of judging whether there is a piled obstacle in the front area according to the real-time images and the semantic map comprises: detecting whether there is a fine obstacle in the real-time images; if there is, positioning the fine obstacle and judging whether the fine obstacle is in a to-be-cleaned track of the autonomous sanitation vehicle according to the semantic map; if yes, determining that there is a piled obstacle in the front area; if no or the fine obstacle is not detected, determining that there is no piled obstacle in the front area.

3. The method of claim 2, wherein, the step of positioning the fine obstacle and judging whether the fine obstacle is in the to-be-cleaned track of the autonomous sanitation vehicle according to the semantic map if there is comprises: if the fine obstacle is detected in the real-time images, positioning the fine obstacle to determine a fine obstacle position; dividing the front area according to lane lines in the semantic map to obtain a plurality of to-be-cleaned areas; judging whether the fine obstacle position and the autonomous sanitation vehicle are in the same to-be-cleaned area; if yes, determining that the fine obstacle is in the to-be-cleaned track of the autonomous sanitation vehicle; if no, determining that the fine obstacle is not in the to-be-cleaned track of the autonomous sanitation vehicle.

4. The method of claim 1, wherein, the step of constructing a road surface grid area of a preset specification and continuously detecting obstacle point clouds corresponding to the piled obstacle comprises: calling a camera component to construct a detection area according to a preset distance; grid-dividing the detection area according to a preset specification to generate a road surface grid area composed of a plurality of grids; continuously detecting obstacle point clouds corresponding to the piled obstacle by calling a laser radar.

5. The method of claim 1, wherein, the step of generating a speed reduction instruction and delivering the speed reduction instruction to a planning control component if there is a pending grid in which the point cloud dynamic accumulation value is greater than the preset piled threshold value within a first preset time length comprises: If there is a pending grid whose point cloud dynamic accumulation value is greater than the preset accumulation threshold value within a first preset time length, it is determined whether the number of the pending grid is greater than or equal to a preset grid threshold value; If yes, a speed reduction instruction is generated and sent to a planning control component; If no, the step of respectively comparing the point cloud dynamic accumulation value of the obstacle point cloud in each grid with the preset accumulation threshold value is executed.

6. The method of claim 1, wherein, The method further comprises: If there is no pending grid whose point cloud dynamic accumulation value is greater than the preset accumulation threshold value within a first preset time length, the step of respectively comparing the point cloud dynamic accumulation value of the obstacle point cloud in each grid with the preset accumulation threshold value is executed. If the point cloud dynamic accumulation value is less than or equal to the preset accumulation threshold value and lasts for a second preset time length, or it is determined that the front area does not exist a piled obstacle, a speed-up instruction is generated and sent to the planning control component; The planning control component increases the corresponding brush rotating speed of the autonomous driving sanitation vehicle to a regular working rotating speed, and increases the driving speed to a regular working speed.

7. A sweeping speed adjustment device for a sanitation vehicle, characterized in that, It comprises: A real-time image acquisition module for acquiring real-time images corresponding to the front area of the autonomous driving sanitation vehicle; An accumulated obstacle judgment module for judging whether there is an accumulated obstacle in the front area according to the real-time images and a semantic map; A vehicle speed adjustment module for reducing the driving speed of the autonomous driving sanitation vehicle to a piled cleaning speed if there is; A grid and point cloud creation module for constructing a road grid area of a preset specification and continuously detecting obstacle point clouds corresponding to the piled obstacle; A rotating speed adjustment module for adjusting the brush rotating speed of the autonomous driving sanitation vehicle according to the road grid area and the obstacle point clouds; The rotating speed adjustment module comprises: A point cloud mapping submodule for mapping the obstacle point clouds to each grid in the road grid area; A comparison submodule for respectively comparing the point cloud dynamic accumulation value of the obstacle point cloud in each grid with a preset accumulation threshold value; A speed reduction instruction generation submodule for generating a speed reduction instruction and sending it to the planning control component if there is a pending grid whose point cloud dynamic accumulation value is greater than the preset accumulation threshold value within a first preset time length; A rotating speed reduction submodule for reducing the brush rotating speed of the autonomous driving sanitation vehicle to a piled cleaning speed by the planning control component in response to the speed reduction instruction.

8. An electronic device, comprising: It comprises a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the sanitation vehicle cleaning speed adjustment method according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed to realize the sanitation vehicle cleaning speed adjustment method according to any one of claims 1-6.

Images