CN115759660A

CN115759660A - Scheduling method, device, equipment and medium for unmanned vehicle

Info

Publication number: CN115759660A
Application number: CN202211482454.8A
Authority: CN
Inventors: 黄超
Original assignee: Guangzhou Weride Technology Co Ltd
Current assignee: Guangzhou Weride Technology Co Ltd
Priority date: 2022-11-24
Filing date: 2022-11-24
Publication date: 2023-03-07

Abstract

The invention discloses a method, a device, equipment and a medium for scheduling an unmanned vehicle, wherein the method comprises the steps of obtaining a vehicle operation area through a cloud scheduling platform and carrying out gridding to obtain a gridding area diagram; determining grid weight and a plurality of path points corresponding to each grid according to historical order data corresponding to each grid in the gridding area graph; planning a path according to a plurality of path points, and constructing vehicle tour routes corresponding to each grid; after the grid weights are standardized, dispatching the unmanned vehicle to each grid; the method comprises the steps of sending corresponding scheduling instructions to each unmanned vehicle according to a vehicle tour route, enabling the unmanned vehicles to carry out circular tour according to the vehicle tour route, carrying out region scheduling and circular tour on the unmanned vehicles according to grid weights and instruction scheduling modes through a cloud scheduling platform, and accordingly reducing the number of personnel participating in vehicle scheduling and improving scheduling efficiency while effectively reducing labor cost.

Description

Scheduling method, device, equipment and medium for unmanned vehicle

Technical Field

The invention relates to the technical field of unmanned vehicle scheduling, in particular to a scheduling method, device, equipment and medium of an unmanned vehicle.

Background

With the improvement of economic development level, the traffic demand of people is continuously increased, people not only need common public transportation modes such as urban buses and long-distance line passenger transportation, but also sometimes need convenient, fast, safe and comfortable passenger transportation modes, and taxis are popular to people due to the flexible operation characteristics of the taxis.

With the continuous maturity of the unmanned technology, the unmanned taxi becomes one of the development trends of the unmanned system, and the dispatching of the existing unmanned taxi generally dispatches the vehicle to a certain area through a manual dispatching mode by a transport person, and then controls the vehicle to cruise in the area through a security guard. However, in the above scheme, a plurality of persons are required to participate in scheduling in the vehicle scheduling process, and the scheduling efficiency is low.

Disclosure of Invention

The invention provides a method, a device, equipment and a medium for scheduling an unmanned vehicle, which solve the technical problems that the existing unmanned vehicle scheduling scheme needs a plurality of persons to participate in scheduling in the vehicle scheduling process and the scheduling efficiency is low.

The invention provides a method for dispatching an unmanned vehicle, which is applied to a cloud dispatching platform and comprises the following steps:

acquiring a vehicle operation area and gridding to obtain a gridding area graph;

determining grid weight and a plurality of path points corresponding to each grid according to historical order data corresponding to each grid in the gridding area graph;

planning a path according to the multiple path points, and constructing vehicle tour routes corresponding to the grids respectively;

dispatching the unmanned vehicle to each grid according to the grid weight;

and sending a corresponding scheduling instruction to each unmanned vehicle according to the vehicle tour route, so that the unmanned vehicles perform circular tour according to the vehicle tour route.

Optionally, the step of obtaining the vehicle operation area and performing meshing to obtain a meshed area map includes:

acquiring a vehicle operation area;

positioning a plurality of boundary points corresponding to the vehicle operation area along a preset direction, and constructing a quadrilateral area by adopting all the boundary points;

and gridding the quadrilateral area according to a preset unit specification to obtain a gridding area diagram.

Optionally, the step of determining a grid weight and a plurality of route points corresponding to each grid according to historical order data corresponding to each grid in the gridding area graph includes:

clustering historical order data corresponding to each grid in the gridding area graph to obtain a plurality of clustering clusters;

selecting a preset number of target clustering clusters from the plurality of clustering clusters according to the order use frequency from high to low, and determining the positions of the target clustering clusters as the passing points;

and respectively taking the historical order data corresponding to each grid as the grid weight corresponding to each grid.

Optionally, the step of dispatching the unmanned vehicle to each of the grids according to the grid weights comprises:

mapping each grid weight to a preset integer distribution space for standardization to obtain a plurality of corresponding vehicle distribution quantities;

dispatching the unmanned vehicles to each of the grids according to the assigned number of each of the vehicles.

Optionally, the method further comprises:

determining any unmanned vehicle which completes the order task as a vehicle to be scheduled;

searching all the grids from high to low according to the grid weight, and determining the number of vehicles of each grid at the current moment;

selecting grids with the number of the vehicles less than the number of the vehicles distributed as demand grids;

and dispatching the vehicle to be dispatched to the demand grid, and sending a corresponding dispatching instruction to the vehicle to be dispatched according to the vehicle tour route of the demand grid.

Optionally, the unmanned vehicle is further provided with a people stream identification component, the method further comprising:

when the unmanned vehicle executes the order task, identifying the personnel density corresponding to a plurality of driving positions in the driving path through the people flow identification component;

selecting the target driving positions with the personnel density larger than a preset density threshold value, and positioning a first grid to be added of each target driving position in the grid area graph;

increasing the driving position to the passing point corresponding to the first grid to be added from large to small according to the personnel density until the number of the passing points is equal to a preset number threshold;

and skipping to execute the step of planning the path according to the plurality of path points and constructing the vehicle tour route corresponding to each grid.

Optionally, the people flow identification component comprises a vehicle-mounted camera and a laser radar; when the unmanned vehicle executes the order task, the step of identifying the plurality of personnel densities corresponding to the driving path through the people flow identification component comprises the following steps:

when the unmanned vehicle executes the order task, acquiring street view image flow of a driving path in real time through the vehicle-mounted camera;

carrying out image recognition on the street view image stream frame by frame, and determining figure images of each driving position;

acquiring point cloud data on the travelling path in real time through the laser radar;

performing specific object recognition on the point cloud data, and determining target point cloud data which accords with character characteristics;

projecting the target point cloud data to the character images, and matching the number of people corresponding to each character image;

and when the unmanned vehicle leaves the driving path, determining the personnel density corresponding to each driving position by adopting the personnel number and combining the driving positions.

Optionally, the method further comprises:

when accessing a vehicle wireless communication technology V2X system, retrieving the V2X system according to a preset request;

if the street view image stream is searched, skipping to execute the step of carrying out image recognition on the street view image stream frame by frame and determining the figure image of each driving position;

and if the demand grid is searched, jumping to the step of dispatching the vehicle to be dispatched to the demand grid, and sending a corresponding dispatching instruction to the vehicle to be dispatched according to the vehicle tour route of the demand grid.

Optionally, the method further comprises:

when a third-party map is accessed, hot interest points on the third-party map are obtained, and a second to-be-added grid of each hot interest point in the grid area map is determined;

adding the hot interest points to the passing points corresponding to the second grid to be added until the number of the passing points is equal to a preset number threshold;

Optionally, the method further comprises:

when a map open platform is accessed and an initial map point is received, selecting a map recording point with the occurrence frequency greater than a preset frequency threshold value in the gridding area map from the initial map point, and determining a corresponding third grid to be added;

adding the map recording points to the passing points corresponding to the third to-be-added grid until the number of the passing points is equal to a preset number threshold;

The invention provides a dispatching device of an unmanned vehicle, which is applied to a cloud dispatching platform and comprises:

the regional gridding module is used for acquiring a vehicle operation region and gridding the vehicle operation region to obtain a gridding region diagram;

the weight and route point determining module is used for determining grid weight and a plurality of route points corresponding to each grid according to historical order data corresponding to each grid in the gridding area graph;

the path planning module is used for planning paths according to the path points and constructing vehicle tour routes corresponding to the grids respectively;

a vehicle scheduling module for scheduling unmanned vehicles to each of the grids according to the grid weights;

and the dispatching instruction sending module is used for sending corresponding dispatching instructions to each unmanned vehicle according to the vehicle tour route so that the unmanned vehicles carry out circular tour according to the vehicle tour route.

A third aspect of the present invention provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, causes the processor to perform the steps of the method for scheduling an unmanned vehicle according to any one of the first aspect of the present invention.

A fourth aspect of the invention provides a computer readable storage medium having stored thereon a computer program which, when executed, implements a method of scheduling an unmanned vehicle according to any one of the first aspects of the invention.

According to the technical scheme, the invention has the following advantages:

acquiring a vehicle operation area through a cloud dispatching platform and gridding the vehicle operation area to obtain a gridding area graph; determining grid weight and a plurality of path points corresponding to each grid according to historical order data corresponding to each grid in the gridding area graph; planning a path according to a plurality of path points, and constructing vehicle tour routes corresponding to each grid; after the grid weights are standardized, dispatching the unmanned vehicle to each grid; the method comprises the steps of sending corresponding scheduling instructions to each unmanned vehicle according to a vehicle tour route, enabling the unmanned vehicles to carry out circular tour according to the vehicle tour route, carrying out region scheduling and circular tour on the unmanned vehicles according to grid weights and instruction scheduling modes through a cloud scheduling platform, and accordingly reducing the number of personnel participating in vehicle scheduling and improving scheduling efficiency while effectively reducing labor cost.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.

Fig. 1 is a flowchart illustrating steps of a scheduling method for an unmanned vehicle according to an embodiment of the present invention;

FIG. 2 is a flowchart illustrating steps of a method for scheduling an unmanned vehicle according to a second embodiment of the present invention;

fig. 3 is a block diagram of a scheduling apparatus of an unmanned vehicle according to a third embodiment of the present invention.

Detailed Description

The embodiment of the invention provides a method, a device, equipment and a medium for scheduling an unmanned vehicle, which are used for solving the technical problems that a plurality of persons are required to participate in scheduling in the vehicle scheduling process and the scheduling efficiency is low in the conventional unmanned vehicle scheduling scheme.

In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1, fig. 1 is a flowchart illustrating a method for scheduling an unmanned vehicle according to an embodiment of the present invention.

step 101, acquiring a vehicle operation area and gridding the vehicle operation area to obtain a gridding area graph;

the cloud scheduling platform is a platform which is based on cloud computing and has high concurrency and high elasticity for data storage, calculation and scheduling, is in communication connection with the unmanned vehicle, and analyzes and processes various data acquired by the unmanned vehicle based on technologies such as a model algorithm and the like.

The vehicle operation area refers to the sum of areas operated by a plurality of unmanned vehicles which belong to the same cloud scheduling platform.

In the embodiment of the application, after the vehicle operation area is obtained through the cloud scheduling platform, the vehicle operation area is gridded according to the preset specification, and a gridding area graph is generated.

Step 102, determining grid weight and a plurality of path points corresponding to each grid according to historical order data corresponding to each grid in the gridding area graph;

after the gridding area map is generated, historical order data corresponding to each grid in the gridding area map may be further acquired, a plurality of sites with the highest use frequency, for example, 5 sites, in each grid may be determined according to the historical order data corresponding to each grid, and the determined plurality of sites may be determined as the passing points.

Meanwhile, the weight of each grid is calculated according to the historical order data, and the grid weight corresponding to each grid is obtained.

103, planning a path according to the multiple path points, and constructing vehicle tour routes corresponding to the grids respectively;

in the embodiment of the present application, after the multiple passing points corresponding to each grid are obtained through recalculation, the multiple passing points may be respectively adopted for path planning according to each grid, so as to respectively construct a vehicle tour route passing through the multiple passing points on each grid.

It should be noted that the vehicle tour route may be obtained by projecting a plurality of waypoints on the grid area map, associating each waypoint with an unmanned vehicle driving road, and setting the total length of the road as the nearest distance.

Step 104, after the grid weights are standardized, dispatching the unmanned vehicle to each grid;

by mapping the grid weight normalization to the integer distribution space, the integer distribution space can be set to, for example, 0 to 3 (log 10 (x)/log 10 (x) _max ) 3) for representing the number of vehicles that should be present daily within this grid, so that a corresponding number of unmanned vehicles can be dispatched to each grid according to the grid weights to complete the assignment of unmanned vehicles.

The type of the unmanned vehicle can be an unmanned taxi.

And 105, sending a corresponding scheduling instruction to each unmanned vehicle according to the vehicle tour route, so that the unmanned vehicles perform circular tour according to the vehicle tour route.

In a specific implementation, after the allocation of the unmanned vehicles is completed, corresponding scheduling instructions may be further generated according to the vehicle tour route corresponding to each grid and sent to each unmanned vehicle. When the unmanned vehicle receives the scheduling instruction, the unmanned vehicle can sequentially pass through all the passing points according to the vehicle tour route associated with the grid to carry out circular tour.

In the embodiment of the application, a vehicle operation area is obtained through a cloud dispatching platform and gridded to obtain a gridded area graph; determining grid weight and a plurality of path points corresponding to each grid according to historical order data corresponding to each grid in the gridding area graph; planning a path according to a plurality of path points, and constructing vehicle tour routes corresponding to each grid; after the grid weights are standardized, dispatching the unmanned vehicle to each grid; the method comprises the steps of sending corresponding scheduling instructions to each unmanned vehicle according to a vehicle tour route, enabling the unmanned vehicles to carry out circular tour according to the vehicle tour route, carrying out region scheduling and circular tour on the unmanned vehicles according to grid weights and instruction scheduling modes through a cloud scheduling platform, and accordingly reducing the number of personnel participating in vehicle scheduling and improving scheduling efficiency while effectively reducing labor cost.

Referring to fig. 2, fig. 2 is a flowchart illustrating steps of a method for scheduling an unmanned vehicle according to a second embodiment of the present invention.

step 201, obtaining a vehicle operation area and gridding the vehicle operation area to obtain a gridding area graph;

optionally, step 201 may comprise the following sub-steps:

acquiring a vehicle operation area;

In the embodiment of the application, a quadrilateral area such as a rectangular area is constructed by acquiring vehicle operation areas of a plurality of unmanned vehicles which belong to the same cloud scheduling platform for scheduling, positioning a plurality of boundary points of the vehicle operation areas along a preset direction such as south, east, west and north, and adopting all the boundary points as vertexes or outer middle points.

And carrying out gridding area segmentation on the quadrilateral area according to a preset unit specification, wherein the preset unit specification can be a grid specification with the longitude and latitude of 0.05, namely the distance of 5 km. The method includes the step of performing gridding area division on a quadrilateral area from left to right and from top to bottom, and thus obtaining a gridding area.

Step 202, clustering historical order data corresponding to each grid in the gridding area graph to obtain a plurality of clustering clusters;

historical order data refers to order request data placed at a location within the grid and may be represented by an amount of orders over a period of time, such as a day, a week, a month, etc.

In the embodiment of the application, after the gridding area graph is obtained, historical order data can be respectively obtained for each grid, and after the historical order data are clustered according to positions, a plurality of cluster clusters are obtained to reflect the use condition of the historical order data in the grid.

Step 203, selecting a preset number of target cluster clusters from the plurality of cluster clusters according to the order use frequency from high to low, and determining the positions of the target cluster clusters as the passing points;

after a plurality of clustering clusters are obtained, the order use frequency in each clustering cluster can be further calculated, a preset number of target clustering clusters are selected from the clustering clusters according to the order use frequency from high to low, and the positions of the target clustering clusters are determined as the passing points.

The order usage frequency can be determined by the order quantity in unit time, such as a day, a half day or by time period.

Step 204, respectively taking the historical order data corresponding to each grid as the grid weight corresponding to each grid;

in a specific implementation, the order quantity may be extracted from the historical order data corresponding to each grid, and the grid weight corresponding to each grid is taken as the respective order quantity, and the weight is relatively higher when the number is larger.

Step 205, planning a path according to a plurality of path points, and constructing vehicle tour routes corresponding to each grid;

in a specific implementation, the algorithm of the path planning may include, but is not limited to, dijkstra, a, D, LPA, D, and other algorithms, and after obtaining a plurality of waypoints, any waypoint may be used as a starting point, the vehicle may sequentially pass through the remaining waypoints along the driving lane of the unmanned vehicle, and finally the vehicle tour route corresponding to the mesh may be constructed by using the first waypoint as a key point.

In order to further save resource consumption of the unmanned vehicle, the vehicle tour route with the shortest route is selected after at least one vehicle tour route is obtained through calculation.

Step 206, after the grid weights are standardized, dispatching the unmanned vehicle to each grid;

optionally, step 206 may include the following sub-steps:

and dispatching the unmanned vehicles to each grid according to the distributed quantity of each vehicle.

The integer distribution space refers to an interval provided with a plurality of intermediate values, and belongs to one of standardized operation mapping spaces. For example, an integer distribution space of 0 to 3 can be obtained by a log function normalization method in which a grid weight is calculated and then substituted as x into log10 (x)/log 10 (x) _max ) And 3, calculating to obtain corresponding mapping values, and taking the maximum value of the interval where the mapping values are located as the vehicle distribution quantity.

In this embodiment, after the grid weight mapping is normalized to the preset integer distribution space, the vehicle distribution number corresponding to each grid is determined, and the unmanned vehicle is further dispatched to the starting point of each grid according to the vehicle distribution number, so as to complete the preliminary distribution of the grid.

And step 207, sending a corresponding scheduling instruction to each unmanned vehicle according to the vehicle tour route, so that the unmanned vehicles perform circular tour according to the vehicle tour route.

In the embodiment of the application, the scheduling instruction of each unmanned vehicle is generated according to the vehicle tour route corresponding to each grid and the information of each passing point and the associated lane.

And the cloud scheduling platform sends the scheduling instruction to each unmanned vehicle, so that each unmanned vehicle performs circular line tour in the grid according to the corresponding vehicle tour route.

Further, the method comprises the following steps S11-S14:

s11, determining any unmanned vehicle completing the order task as a vehicle to be scheduled;

s12, searching all grids from high to low according to the grid weight, and determining the number of vehicles of each grid at the current moment;

s13, selecting grids with the number of vehicles less than the number of vehicles distributed as demand grids;

and S14, dispatching the vehicle to be dispatched to the demand grid, and sending a corresponding dispatching instruction to the vehicle to be dispatched according to the vehicle tour route of the demand grid.

In the embodiment of the application, when any unmanned vehicle receives the order task, the unmanned vehicle drives to the position specified by the order task to execute the order task.

After the order task is completed, the starting point of the order task exists in the grid which is responsible for the unmanned vehicle, but the end point of the order task may not exist in the grid, so that in order to achieve dynamic scheduling of the unmanned vehicle in each grid, the unmanned vehicle which completes the order task can be determined as a vehicle to be scheduled, all grids are sequentially searched from high to low according to the weight of each grid, the number of vehicles in each grid at the current moment is determined, and the grid with the number of vehicles less than the number of vehicles distributed is selected as a demand grid. And dispatching the vehicle to be dispatched to the demand grid, generating a corresponding dispatching instruction according to the vehicle tour route of the demand grid, and sending the dispatching instruction to the vehicle to be dispatched, so that the vehicle to be dispatched can carry out circular tour along the vehicle tour route.

In one example of the invention, the unmanned vehicle is further provided with a people stream identification component, the method further comprising the steps S21-S24 of:

s21, when the unmanned vehicle executes order tasks, identifying the personnel density corresponding to a plurality of driving positions in a driving path through a people stream identification component;

further, the people flow identification component comprises a vehicle-mounted camera and a laser radar, and S21 may comprise the following sub-steps:

when the unmanned vehicle executes an order task, a street view image stream of a driving path is obtained in real time through the vehicle-mounted camera;

acquiring point cloud data on a travelling path in real time through a laser radar;

performing specific object recognition on the point cloud data, and determining target point cloud data which accord with character characteristics;

projecting the target point cloud data to the character images, and matching the number of the corresponding personnel of each character image;

when the unmanned vehicle leaves the driving path, the number of each person is combined with the driving position, and the person density corresponding to each driving position is determined.

In the embodiment of the application, when the unmanned vehicle receives the order task and executes the order task, the street view image stream of the driving path where the unmanned vehicle is located is obtained in real time through the vehicle-mounted camera so as to obtain the data basis of people stream identification.

The image of the street view image stream is performed frame by frame to determine the figure image at each driving position, wherein the driving position refers to the position where a certain number of people exist on the driving path. Meanwhile, point cloud data on a driving path are obtained in real time through a laser radar, specific object identification is carried out, target point cloud data which accord with character features are determined, and the target point cloud data are projected to character images so as to match personnel in each character image and count the number of the personnel. When the unmanned vehicle leaves the driving path, the number of the personnel corresponding to each character image is combined with each driving position, the personnel intensity corresponding to each driving position is generated and uploaded to the cloud dispatching platform until the unmanned vehicle completes the order task.

S22, selecting target driving positions with the personnel density larger than a preset density threshold value, and positioning a first grid to be added of each target driving position in a gridding area graph;

s23, increasing the driving position to the passing points corresponding to the first grid to be added from large to small according to the personnel density until the number of the passing points is equal to a preset number threshold;

and S24, jumping to execute a step of planning a path according to the multiple path points and constructing vehicle tour routes corresponding to the grids respectively.

After the cloud scheduling platform obtains the personnel density of each driving position, selecting a target driving position larger than a preset density threshold value, and positioning a first to-be-added grid of each target driving position in a grid area graph. And increasing the driving positions to the route points corresponding to the first grid to be added from large to small according to the personnel density until the quantity of the route points is equal to a preset quantity threshold value, filtering the excessive driving positions, and skipping to the step 205. And planning a path by using the newly added multiple passing points, and reconstructing a vehicle tour route corresponding to each grid.

It should be noted that the driving positions that are not added may be stored according to the grid first, and until the passing points are newly added again, the part of the driving positions are added to the passing points according to the storage sequence.

In another example of the present invention, the method further comprises steps S31-S33:

s31, when the vehicle wireless communication technology V2X system is accessed, the V2X system is retrieved according to a preset request;

s32, if the street view image stream is searched, skipping to execute the step of carrying out image recognition on the street view image stream frame by frame and determining the character image of each driving position;

and S33, if the demand grid is searched, skipping to the step of dispatching the vehicle to be dispatched to the demand grid, and sending a corresponding dispatching instruction to the vehicle to be dispatched according to the vehicle tour route of the demand grid.

The V2X system refers to Vehicle to X, and a wireless communication technology system, namely, information exchange between a Vehicle and the outside, can automatically select a driving route with the best road condition by analyzing real-time traffic information. Meanwhile, street view image streams can be cached, and vehicle requirements sent by the requirement grids can be responded to record the requirement grids.

In the embodiment of the application, when the cloud scheduling platform is accessed to the V2X system, the V2X system may be retrieved according to a preset request, such as an image stream request or a demand grid request, and if a street view image stream is retrieved, the street view image stream may be directly acquired from the V2X system, and the street view image stream is subjected to image recognition frame by frame to determine a character image of each driving position.

If the demand grid exists, the step S14 can be skipped to dispatch the vehicle to be dispatched, which exists at the current moment, to the demand grid, generate a corresponding dispatching instruction and send the dispatching instruction to the vehicle to be dispatched, so that the vehicle to be dispatched can carry out the circular line tour along the vehicle tour route corresponding to the demand grid.

Optionally, the method further comprises the following steps S41-S43:

s41, when a third-party map is accessed, hot interest points on the third-party map are obtained, and a second to-be-added grid of each hot interest point in the grid area map is determined;

s42, adding the hot interest points to the passing points corresponding to the second grid to be added until the number of the passing points is equal to a preset number threshold;

and S43, skipping to execute a step of planning a path according to the multiple path points and constructing vehicle tour routes corresponding to the grids respectively.

The popular Interest points refer to vehicle demand points of a vehicle operation area in a third-party map, for example, POI is an abbreviation of "Point of Interest", and chinese can be translated into "Interest points". In the geographic information system, one POI may be a house, a shop, a mailbox, a bus station, etc. Usually, the third-party map has a taxi taking function, positions of taxi taking requests can be counted at the same time, and the positions are sorted according to the request quantity to obtain popular points of interest.

In this embodiment, if the cloud scheduling platform accesses the third-party map, it indicates that the new data source is updated at this time. The hot interest points can be obtained from the third-party map, a second grid to be added of each hot interest point in the gridding area graph is positioned, the hot interest points are added to the passing points corresponding to the second grid to be added again until the number of the passing points is equal to the preset number threshold. And filtering the excessive driving positions, and skipping to the step 205. And planning the path by using the newly added multiple passing points, and reconstructing a vehicle tour route corresponding to each grid.

In an alternative embodiment of the invention, the method further comprises the following steps S51-S53:

s51, when a map open platform is accessed and an initial map point is received, selecting a map recording point with the occurrence frequency greater than a preset frequency threshold value in a gridding area map from the initial map point, and determining a corresponding third grid to be added;

s52, adding the map recording points to the passing points corresponding to the third to-be-added grid until the number of the passing points is equal to the preset number threshold;

and S53, jumping to execute a step of planning a path according to the multiple path points and constructing vehicle tour routes corresponding to the grids respectively.

The map open platform refers to a platform as a service, which is provided with a plurality of API interfaces in different forms to provide different systems and different types of map data, such as a high-level map open platform.

In this embodiment, when the cloud scheduling platform is accessed to the map opening platform, the estimation interface of the map opening platform sends a plurality of initial map points to the cloud scheduling platform, and the cloud scheduling platform counts the initial map points with the occurrence frequency greater than the preset frequency threshold in the gridding area map as the map recording points while acquiring the initial map points, and simultaneously locates the grid where the map recording points are located as the third grid to be added. And adding the map record points to the passing points corresponding to the third to-be-added grid until the number of the passing points is equal to the preset number threshold. And filtering the excessive driving positions, and skipping to the step 205. And planning a path by using the newly added multiple passing points, and reconstructing a vehicle tour route corresponding to each grid.

In the embodiment of the application, a vehicle operation area is obtained through a cloud dispatching platform and gridded to obtain a gridded area graph; determining grid weight and a plurality of path points corresponding to each grid according to historical order data corresponding to each grid in the gridding area graph; planning a path according to a plurality of path points, and constructing vehicle tour routes corresponding to each grid; after the grid weights are standardized, dispatching the unmanned vehicle to each grid; the method comprises the steps of sending corresponding scheduling instructions to each unmanned vehicle according to a vehicle tour route, enabling the unmanned vehicles to carry out circular tour according to the vehicle tour route, carrying out region scheduling and circular tour on the unmanned vehicles through a cloud scheduling platform according to grid weights and an instruction scheduling mode, and accordingly reducing the number of personnel participating in vehicle scheduling, and improving scheduling efficiency while effectively reducing labor cost.

Referring to fig. 3, fig. 3 is a block diagram of a scheduling apparatus of an unmanned vehicle according to a third embodiment of the present invention.

The embodiment of the invention provides a scheduling device of an unmanned vehicle, which is applied to a cloud scheduling platform and comprises the following components:

the regional gridding module 301 is used for acquiring a vehicle operation region and gridding the vehicle operation region to obtain a gridded region map;

a weight and route point determination module 302, configured to determine a grid weight and multiple route points corresponding to each grid according to historical order data corresponding to each grid in the gridding area graph;

the path planning module 303 is configured to perform path planning according to the multiple access points, and construct vehicle tour routes corresponding to each grid;

the vehicle scheduling module 304 is used for scheduling the unmanned vehicle to each grid after the grid weight is standardized;

and the scheduling instruction sending module 305 is configured to send a corresponding scheduling instruction to each unmanned vehicle according to the vehicle tour route, so that the unmanned vehicle performs the circular tour according to the vehicle tour route.

Optionally, the area gridding module 301 is specifically configured to:

acquiring a vehicle operation area;

Optionally, the weight and route point determining module 302 is specifically configured to:

selecting a preset number of target clustering clusters from a plurality of clustering clusters from high to low according to the order use frequency, and determining the positions of the target clustering clusters as the passing points;

Optionally, the vehicle scheduling module 304 is specifically configured to:

Optionally, the apparatus further comprises:

the vehicle to be scheduled determining module is used for determining any unmanned vehicle which completes the order task as a vehicle to be scheduled;

the vehicle number counting module is used for searching all grids from high to low according to the grid weight and determining the number of vehicles of each grid at the current moment;

the demand grid selection module is used for selecting grids with the number of vehicles less than the number of vehicles distributed as demand grids;

and the instruction sending module is used for dispatching the vehicle to be dispatched to the demand grid and sending a corresponding dispatching instruction to the vehicle to be dispatched according to the vehicle tour route of the demand grid.

Optionally, the unmanned vehicle is further provided with a people stream recognition assembly, the apparatus further comprising:

the personnel density identification module is used for identifying the personnel density corresponding to a plurality of driving positions in the driving path through the people stream identification component when the unmanned vehicle executes the order task;

the first grid to be added positioning module is used for selecting a target driving position with the personnel density larger than a preset density threshold value and positioning a first grid to be added of each target driving position in a gridding area graph;

the first passing point adding module is used for adding the travelling position to the passing point corresponding to the first grid to be added from large to small according to the personnel density until the number of the passing points is equal to the preset number threshold;

and the first planning and skipping module is used for skipping and executing the steps of planning the path according to the multiple path points and constructing the vehicle tour routes corresponding to the grids respectively.

Optionally, the people flow identification component comprises a vehicle-mounted camera and a laser radar; the personnel intensity identification module is specifically configured to:

acquiring point cloud data on a driving path in real time through a laser radar;

Optionally, the apparatus further comprises:

the system retrieval module is used for retrieving the V2X system according to a preset request when the vehicle wireless communication technology V2X system is accessed;

the street view skipping module is used for skipping to execute the steps of carrying out image recognition on the street view image stream frame by frame and determining the character image of each driving position if the street view image stream is searched;

and the demand skipping module is used for skipping to the step of scheduling the vehicle to be scheduled to the demand grid and sending a corresponding scheduling instruction to the vehicle to be scheduled according to the vehicle tour route of the demand grid if the demand grid is retrieved.

Optionally, the apparatus further comprises:

the second to-be-added grid positioning module is used for acquiring hot interest points on the third-party map when the third-party map is accessed, and determining a second to-be-added grid of each hot interest point in the gridding area map;

the second passing point adding module is used for adding the hot interest points to the passing points corresponding to the second grid to be added until the number of the passing points is equal to the preset number threshold;

and the second planning and skipping module is used for skipping to execute the steps of planning the path according to the multiple path points and constructing the vehicle tour routes corresponding to the grids respectively.

Optionally, the apparatus further comprises:

the third grid to be added positioning module is used for selecting map recording points with the frequency of occurrence larger than a preset frequency threshold value in the gridding area map from the initial map points when the map open platform is accessed and the initial map points are received, and determining a corresponding third grid to be added;

a third passing point adding module, configured to add a map recording point to a passing point corresponding to a third to-be-added grid until the number of passing points is equal to a preset number threshold;

and the third planning and skipping module is used for skipping to execute the steps of planning the path according to the multiple path points and constructing the vehicle tour routes corresponding to the grids respectively.

It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described apparatuses and modules may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in the form of hardware, or may also be implemented in the form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A scheduling method of an unmanned vehicle is applied to a cloud scheduling platform, and comprises the following steps:

dispatching the unmanned vehicle to each grid according to the grid weight;

and sending a corresponding scheduling instruction to each unmanned vehicle according to the vehicle tour route, so that the unmanned vehicles carry out circular tour according to the vehicle tour route.

2. The method of claim 1, wherein the step of obtaining the vehicle operation area and gridding the vehicle operation area to obtain a gridded area map comprises:

acquiring a vehicle operation area;

3. The method of claim 1, wherein the step of determining grid weights and a plurality of approach points for each grid based on historical order data for each grid in the grid-enabled region graph comprises:

4. The method of claim 1, wherein said step of dispatching said unmanned vehicle to each of said grids according to said grid weights comprises:

5. The method of claim 4, further comprising:

6. The method of claim 5, wherein the unmanned vehicle is further provided with a people flow identification component, the method further comprising:

7. The method of claim 6, wherein the people flow identification component comprises an in-vehicle camera and a lidar; when the unmanned vehicle executes the order task, the step of identifying the plurality of personnel densities corresponding to the driving path through the people flow identification component comprises the following steps:

performing specific object recognition on the point cloud data, and determining target point cloud data which accord with character features;

8. The method of claim 7, further comprising:

9. The method of claim 1, further comprising:

10. The method of claim 1, further comprising:

and skipping to execute the step of planning the path according to the multiple path points and constructing the vehicle tour routes corresponding to the grids respectively.

11. The utility model provides a scheduling device of unmanned vehicles which characterized in that is applied to high in the clouds dispatch platform, the device includes:

the route planning module is used for planning a route according to the multiple route points and constructing vehicle tour routes corresponding to the grids respectively;

the vehicle scheduling module is used for scheduling the unmanned vehicle to each grid according to the grid weight;

12. An electronic device, comprising a memory and a processor, the memory having stored thereon a computer program that, when executed by the processor, causes the processor to perform the steps of the method of scheduling of unmanned vehicles according to any of claims 1-10.

13. A computer-readable storage medium, having a computer program stored thereon, wherein the computer program, when executed, implements the method of scheduling an unmanned vehicle of any of claims 1-10.