Summary of the invention
The inventor of the disclosure thinks: the difficulty of indoor crusing robot essentially consist in indoor environment closing without GPS signal,
Environment is more complicated than outdoor for indoor environment, the more high aspect of requirement of inspection.These lead to the path of robot when indoor inspection
Plan extremely difficult, the indoor inspection especially for this environment complexity in warehouse will be particularly difficult.In view of the above technical problems,
The present disclosure proposes a kind of indoor inspection schemes easy to accomplish.
According to some embodiments of the present disclosure, a kind of paths planning method is provided, comprising: based on setting in robot
Laser radar acquired in environmental information, the laser in warehouse is constructed using drawing method is built accordingly according to storage area offline
Map;By merging location information acquired in the odometer and inertial sensor that are arranged in the robot and posture letter
Breath and the environmental information, the robot is positioned using adaptability Monte Carlo localization method;In the laser map
On, the polling path of the robot is dynamically planned according to specified inspection target point and inspection sequence.
Optionally, the laser map that drawing method constructs warehouse offline is built using gmapping in the warehouse small for area, packet
It includes: according to location information and environmental information acquired in the odometer and the laser radar, being estimated using particle filter
Count position and the posture of the robot;For occupying the warehouse map that method indicates with grid, according to swashing for the laser radar
Whether light repeatedly reaches the same grid, and to determine, whether there are obstacles on each grid, so that building is described with laser
Figure.
Optionally, the warehouse big for area builds drawing method using cartographer and constructs warehouse offline with laser
Figure, comprising: according to the data of the laser radar, position and the posture of the robot are estimated using figure Optimized model;It is tired
Add the frame data obtained with the laser scanning of the laser radar, to construct multiple local maps in the warehouse;Pass through spelling
The multiple local map is connect to construct global map;The global map is converted to the grid adapted to the localization method
Laser map.
Optionally, the environmental information includes shelf feature and cargo feature, makes shelf special when positioning the robot
The weight levied in particle filter is greater than cargo feature in the weight of particle filter.
Optionally, the paths planning method further include: control robot carries out inspection according to the polling path of planning, and
And specific movement is executed in specific inspection target point.
Optionally, the polling path of the robot is dynamically planned based on A* algorithm and dynamic window algorithm.
According to other embodiments of the disclosure, a kind of path planning system is provided, comprising: laser radar is configured as
Obtain the environmental information where robot;Odometer is configured as obtaining the location information of the robot;Inertial sensor,
It is configured as obtaining the posture information of the robot;And controller.Controller is configured as: being based on the environmental information, root
The laser map in warehouse is constructed offline using drawing method is built accordingly according to storage area;By merging the location information, appearance
State information and the environmental information, the robot is positioned using adaptability Monte Carlo localization method;It is described with laser
On figure, the polling path of the robot is dynamically planned according to specified inspection target point and inspection sequence.
Optionally, the laser map that drawing method constructs warehouse offline is built using gmapping in the warehouse small for area, packet
It includes: according to location information and environmental information acquired in the odometer and the laser radar, being estimated using particle filter
Count position and the posture of the robot;For occupying the warehouse map that method indicates with grid, according to swashing for the laser radar
Whether light repeatedly reaches the same grid, and to determine, whether there are obstacles on each grid, so that building is described with laser
Figure.
Optionally, the warehouse big for area builds drawing method using cartographer and constructs warehouse offline with laser
Figure, comprising: according to the data of the laser radar, position and the posture of the robot are estimated using figure Optimized model;It is tired
Add the frame data obtained with the laser scanning of the laser radar, to construct multiple local maps in the warehouse;Pass through spelling
The multiple local map is connect to construct global map;The global map is converted to the grid adapted to the localization method
Laser map.
Optionally, the environmental information includes shelf feature and cargo feature, makes shelf special when positioning the robot
The weight levied in particle filter is greater than cargo feature in the weight of particle filter.
Optionally, the controller is additionally configured to control robot and carries out inspection according to the polling path of planning, and
Specific movement is executed in specific inspection target point.
Optionally, the polling path of the robot is dynamically planned based on A* algorithm and dynamic window algorithm.
According to the other embodiment of the disclosure, a kind of path planning system is provided, comprising: memory and be coupled to institute
The processor of memory is stated, the processor is configured to executing above-mentioned based on the instruction being stored in the memory device
Paths planning method described in any one embodiment.
According to the still other embodiments of the disclosure, a kind of path planning system, comprising: implement for executing any of the above-described
The device of paths planning method described in example.
According to some embodiments of the present disclosure, a kind of robot is provided, including path described in any of the above-described a embodiment
Planning system.
According to other embodiments of the disclosure, a kind of computer readable storage medium is provided, computer is stored thereon with
Program, the program realize paths planning method described in any of the above-described a embodiment when being executed by processor.
In the above-described embodiments, warehouse is constructed with laser offline using drawing method is built accordingly according to storage area
Figure passes through location information, posture information and environmental information acquired in fusion setting odometer, inertial sensor and laser radar
Carry out positioning robot, according to specified inspection target point and inspection sequence come the polling path of dynamically planning robot.In this way
It can easily realize the indoor inspection of robot.
Specific embodiment
The various exemplary embodiments of the disclosure are described in detail now with reference to attached drawing.It should also be noted that unless in addition having
Body explanation, the unlimited system of component and the positioned opposite of step, numerical expression and the numerical value otherwise illustrated in these embodiments is originally
Scope of disclosure.
Simultaneously, it should be appreciated that for ease of description, the size of various pieces shown in attached drawing is not according to reality
Proportionate relationship draw.
Be to the description only actually of at least one exemplary embodiment below it is illustrative, never as to the disclosure
And its application or any restrictions used.
Technology, method and apparatus known to person of ordinary skill in the relevant may be not discussed in detail, but suitable
In the case of, the technology, method and apparatus should be considered as authorizing part of specification.
It is shown here and discuss all examples in, any occurrence should be construed as merely illustratively, without
It is as limitation.Therefore, the other examples of exemplary embodiment can have different values.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, then in subsequent attached drawing does not need that it is further discussed.
Fig. 1 shows the flow chart of some embodiments of the paths planning method according to the disclosure.
As shown in Figure 1, paths planning method includes: step 110, the laser map in warehouse is constructed;Step 120, localization machine
Device people;With step 130, the dynamically polling path of planning robot.
Firstly, in step 110, based on environmental information acquired in the laser radar being arranged in robot, according to storehouse
Library area constructs the laser map in warehouse using drawing method is built accordingly offline.Here, environmental information is for example including in warehouse
Shelf feature and shelf on cargo feature, legend such as can occupy method using grid to indicate with laser.
Gmapping builds drawing method, and figure speed is fast, parameter can easily be accommodated due to building, and compares and is suitble to small scene.Therefore, right
In the small warehouse of area, the laser map that drawing method constructs warehouse offline can be built using gmapping.Fig. 2A is shown according to this
The flow chart of disclosed some embodiments that drawing method building laser map is built using gmapping.
Gmapping builds the data that drawing method needs odometer and laser radar.The data of inertial sensor can be used as auxiliary
It helps, but is not required.As shown in Fig. 2, it includes: step 211A that gmapping, which builds drawing method, estimated using particle filter
The position of robot and posture;With step 214A, determine that whether there are obstacles on each grid of map, to construct grid
Lattice laser map.
In step 211A, according to location information acquired in odometer and laser radar and environmental information, particle is utilized
Filter estimates position and the posture of robot.
In step 214A, for occupying the warehouse map that method indicates with grid, according to the laser of institute's laser radar whether
Repeatedly reaching the same grid, whether there are obstacles on each grid to determine, to construct laser map.Can will be
No there are the grid tag of barrier is different colours, or makes not isolabeling and construct laser map.For example, can will deposit
It is black or grey in the grid tag of barrier, and there will be no the grid tags of barrier for white or without label.
Gmapping parameter can be adjusted as needed during entirely building figure.
Fig. 3 shows the schematic diagram of the laser map in the warehouse of the building of method shown in A according to fig. 2.
As shown in figure 3, there are the grids of barrier to be labeled, the grid that barrier may be not present is not labeled, thus shape
In contrast with distinct warehouse map.
It is more effective that inventor it is also found that: for big scene, using cartographer builds figure building laser map, because
Its bring error can will not thus be added up independent of the data of odometer by building figure using cartographer.Therefore,
The warehouse big for area can build the laser map that drawing method constructs warehouse offline using cartographer.Fig. 2 B is shown
The flow chart of other embodiments of drawing method building laser map is built using gmapping.
Cartographer builds the data that figure only needs laser radar, and the data of the data of inertial sensor and odometer are all
It is not required.As shown in Figure 2 B, it includes: step 211B that cartographer, which builds drawing method, is estimated using figure Optimized model
The position of robot and posture;Step 212B constructs local map;Step 213B splices local map to construct global map;
With step 214B, the format of global map is converted, constructs grid laser map.
In step 211B, according to the data of laser radar, position and the appearance of robot are estimated using figure Optimized model
State.
In step 212B, using the frame data of laser scanning, local map is constructed by adding up.For example, using
The matching of Ceres laser constructs multiple local maps.
In step 213B, global map is constructed by splicing the multiple local maps constructed.Here it is possible to pass through
Closed loop (Loop closing) eliminates the cumulative errors generated in splicing.
In step 214B, global map is converted to the grid laser map adapted to localization method.In this way
The grating map that format is converted to can improve the accuracy of positioning.
It, can be with positioning robot, to carry out path planning after having been built up the laser map in Long Position library.Return to ginseng
Fig. 1 is examined to describe the step 120 of positioning robot and carry out the step 130 of path planning.
In the step 120, believed by merging position acquired in the odometer and inertial sensor that are arranged in robot
Breath and posture information and environmental information, using adaptability Monte Carlo (AMCL) localization method come positioning robot.AMCL is fixed
Position method cooperates laser scanning feature, obtains optimal anchor point using particle filter according to the laser map built.
For not only including shelf but also including the warehouse of cargo, shelf feature can be made big in the weight of particle filter
In cargo feature particle filter weight, to mitigate influence of the variation to robot localization of cargo on shelf.Namely
It says, increases it in the weight of particle filter by extracting shelf feature, and when matching the feature that laser scanning goes out, so that
Positioning relatively relies on shelf rather than the cargo on shelf, to mitigate influence of the variation to robot localization of cargo.
In step 130, on laser map, machine is dynamically planned according to specified inspection target point and inspection sequence
The polling path of device people.
One complete path generally includes starting point, each target point and terminal.The beginning and end of polling path is general
It can be selected as automatic charging point.Inspection target point can be selected according to the security needs in warehouse.The planning in path be often referred to according to
According to certain Optimality Criterias (such as walking path is most short, travel time is most short etc.), one is found on map from starting point to target
The optimal path of point energy avoiding obstacles.
In some embodiments, the planning of polling path includes receiving specified inspection target point and each target point
Between inspection sequence after, plan the shortest path between each point, the shortest path between especially each target point.Example
It such as, can be based on A* algorithm and dynamic window algorithm come the polling path of dynamically planning robot.
In the case of inspection target point for warehouse is fewer, such as at 20 or so, specify inspection sequence than using machine
It is higher that device people finds shortest path efficiency between multiple inspection target points.Also, due to the most of all positions of target inspection point
In on the major trunk roads in warehouse, only on the road of very small part between shelves, and walk for robot in major trunk roads and
Say that shelf feature is obvious, it is easier to position.It is therefore intended that inspection sequence can allow robot to be walked in major trunk roads as far as possible, thus
Reduce the position error of robot.
It, can be automatic after robot receives the instruction of increase and decrease inspection target point or modification polling path during inspection
New path planning is carried out, i.e., dynamically plans polling path.In the case where the not big change of the environment in warehouse, work as planning
Without new map when new polling path.Which reduce the troubles for building figure again.
In some embodiments, paths planning method can also include: step 140, control robot patrolling according to planning
It examines path and carries out inspection.
It includes mobile by inspection sequence from starting point (such as charge point) that robot, which carries out inspection according to the polling path of planning,
To specified each inspection target point.Robot is in moving process, by calculating in laser map (such as grating map)
The distance between coordinate and the coordinate of robot of each inspection target point, to judge whether robot reaches target point.Away from
When from meeting certain requirements, when being, for example, less than 0.01m, judge that robot reaches target point.
The influence of barrier is had been contemplated that when although planning polling path, robot still can be real-time in practical inspection
Barrier around detecting, and avoidance is realized by real-time path planning.For example, what robot will test by laser
Barrier is mapped in grating map, the barrier in front of robot within certain distance (such as several meters) is considered, by real-time
Path planning realize avoidance.That is, robot is during inspection, although whole inspection sequence will not change,
It is that path between inspection target point can be according to change the case where barrier.
In some embodiments, step 140 can also include that control robot is specific in the execution of specific inspection target point
Movement.
To check at inspection target point A for the cargo of shelf top layer, corresponding inspection instruction includes at least following information:
The position and course of robot, the motor of camera holder rotation angle and direction.Robot is real-time during inspection
The position for comparing oneself current position and target point when the position for having determined target point A and reaches the course of target point A
After it is required that, corresponding inspection requirement is inquired, control camera holder moves the cargo to check shelf top layer.
Fig. 4 shows the block diagram of some embodiments of the path planning system 4 according to the disclosure.
As shown in figure 4, path planning system 4 includes: laser radar 41, odometer 42 and inertial sensor 43.Laser thunder
It may be located in robot, be respectively configured as where acquisition robot up to 41, odometer 42 and inertial sensor 43
The location information and posture information of environmental information, robot.
As shown, path planning system 4 further includes controller 44.Controller 44 also can be set in robot, use
In the data that processing is obtained from laser radar 41, odometer 42 and inertial sensor 43, to control the action of robot.
In some embodiments, controller 44 is configured as: being based on environmental information, is used according to storage area and built accordingly
Drawing method to construct the laser map in warehouse offline;By fusion location information, posture information and the environmental information, using suitable
Answering property Monte Carlo localization method carrys out positioning robot;With on laser map, it is suitable according to specified inspection target point and inspection
Sequence carrys out the polling path of dynamically planning robot.
In some embodiments, controller 44 be additionally configured to control robot patrolled according to the polling path of planning
Inspection, and specific movement is executed in specific inspection target point.
Fig. 5 shows the structure chart of the other embodiment of the path planning system of the disclosure.
As shown in figure 5, the device 5 of the embodiment includes: memory 51 and the processor 52 for being coupled to the memory 51,
Processor 52 is configured as executing the path in the disclosure in any some embodiments based on the instruction being stored in memory 51
Planing method.
Memory 51 is such as may include system storage, fixed non-volatile memory medium.System storage is for example
It is stored with operating system, application program, Boot loader (Boot Loader), database and other programs etc..
Fig. 6 shows the block diagram of some embodiments of the robot 6 according to the disclosure.
As shown in fig. 6, robot 6 includes path planning system 61 and basic machine 62.Path planning system 61 can be
Any one path planning system in above-described embodiment.In some embodiments, basic machine includes motor, driving wheel and deceleration
Qi Deng mechanism.Robot use differential motion principle when, basic machine may include two driving wheels, two it is driven universal
The structures such as wheel, two direct current generators realize the control of movement.
Those skilled in the art should be understood that embodiment of the disclosure can provide as method, system or computer journey
Sequence product.Therefore, complete hardware embodiment, complete software embodiment or combining software and hardware aspects can be used in the disclosure
The form of embodiment.Moreover, it wherein includes the calculating of computer usable program code that the disclosure, which can be used in one or more,
Machine can use the meter implemented in non-transient storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of calculation machine program product.
So far, the paths planning method, device and computer readable storage medium according to the disclosure is described in detail.
In order to avoid covering the design of the disclosure, some details known in the field are not described.Those skilled in the art are according to upper
The description in face, completely it can be appreciated how implementing technical solution disclosed herein.
Disclosed method and system may be achieved in many ways.For example, can by software, hardware, firmware or
Software, hardware, firmware any combination realize disclosed method and system.The said sequence of the step of for the method
Merely to be illustrated, the step of disclosed method, is not limited to sequence described in detail above, special unless otherwise
It does not mentionlet alone bright.In addition, in some embodiments, also the disclosure can be embodied as to record program in the recording medium, these programs
Including for realizing according to the machine readable instructions of disclosed method.Thus, the disclosure also covers storage for executing basis
The recording medium of the program of disclosed method.
Although being described in detail by some specific embodiments of the example to the disclosure, the skill of this field
Art personnel it should be understood that above example merely to be illustrated, rather than in order to limit the scope of the present disclosure.The skill of this field
Art personnel are it should be understood that can modify to above embodiments in the case where not departing from the scope of the present disclosure and spirit.This public affairs
The range opened is defined by the following claims.