CN107144285A

CN107144285A - Posture information determines method, device and movable equipment

Info

Publication number: CN107144285A
Application number: CN201710317975.0A
Authority: CN
Inventors: 杨德刚
Original assignee: Shenzhen Horizon Robotics Science and Technology Co Ltd
Current assignee: Shenzhen Horizon Robotics Science and Technology Co Ltd
Priority date: 2017-05-08
Filing date: 2017-05-08
Publication date: 2017-09-08
Anticipated expiration: 2037-05-08
Also published as: CN107144285B

Abstract

Disclose a kind of posture information and determine method, device and movable equipment.This method be applied to movable equipment and including：The sample data of the movable equipment obtained by environmental sensor just current mobile environment movable within is received, the sample data includes position data and view data；The local posture information of the movable equipment is determined according at least to the position data；The local semantic map of mobile environment before deserving is built according to the local posture information of the movable equipment and the view data, the local semantic map includes the semantic entity and its attribute information before deserving in mobile environment；Obtain the semantic map of the overall situation of mobile environment before deserving；The semantic map of the part is matched in the semantic map of the overall situation；And in response to matching the semantic map of the part in the semantic map of the overall situation, the local posture information is updated according to the result of matching.It therefore, it can efficiently and accurately obtain the posture information of movable equipment.

Description

Posture information determines method, device and movable equipment

Technical field

The application is related to movable equipment field, and determines method, device and can more particularly, to a kind of posture information Mobile device.

Background technology

Movable equipment (for example, robot) needs to position the position and orientation information of oneself in real time in the task of execution (that is, posture information), and then perform destination decision rule, the perception of real-time barrier etc..

Existing high fine positioning mainly has two methods：One kind is high-precision global positioning system (GPS)/real time dynamic differential (RTK) coordinate high-end mobile receiving device, influence is waited ideally not blocking, it is smart that it can obtain centimeter-level positioning Degree；Another is to coordinate high-precision sensor (to be typically laser radar and height costly based on traditional high-precision numerical map Hold integrated navigation system), real time data is gathered by high-precision sensor and matched with high-precision map, so as to obtain high-precision fixed Position.

However, above-mentioned first method needs extra earth station, the special map collecting device of second of needs is real It is relatively low with property.That is, positioning function is more placed on server apparatus end (high in the clouds) by traditional method, significantly increase The expense of high in the clouds infrastructure.

The content of the invention

In order to solve the above-mentioned technical problem, it is proposed that the application.It is true that embodiments herein provides a kind of posture information Determine method, device, movable equipment, computer program product and computer-readable recording medium, it can efficiently and accurately be obtained Obtain the posture information of movable equipment.

Method is determined there is provided a kind of posture information according to the one side of the application, it is described applied to movable equipment Method includes：Receive the sample of the movable equipment obtained by environmental sensor just current mobile environment movable within Data, the sample data includes position data and view data；Determined according at least to the position data described removable The local posture information of equipment；Work as described in being built according to the local posture information and described image data of the movable equipment The local semantic map of preceding mobile environment, the local semantic map include semantic entity in the current mobile environment and its Attribute information, the semantic entity is possible to influence mobile entity, and the attribute information indicates the thing of the semantic entity Manage characteristic；Obtain the semantic map of the overall situation of the current mobile environment；The local language is matched in the global semantic map Free burial ground for the destitute figure；And in response to matching the local semantic map in the global semantic map, according to the result of matching come Update the local posture information.

It is described applied to movable equipment according to the another aspect of the application there is provided a kind of posture information determining device Device includes：Sample data receiving unit, is just being moved wherein for receiving the movable equipment obtained by environmental sensor The sample data of dynamic current mobile environment, the sample data includes position data and view data；Posture information determines single Member, the local posture information for determining the movable equipment according at least to the position data；Semantic map structuring list Member, the current mobile environment is built for the local posture information and described image data according to the movable equipment Local semanteme map, the local semantic map includes the semantic entity and its attribute information in the current mobile environment, institute It is that possible influence mobile entity to state semantic entity, and the attribute information indicates the physical characteristic of the semantic entity；It is semantic Map acquiring unit, the semantic map of the overall situation for obtaining the current mobile environment；Semantic map matching unit, in institute State and the local semantic map is matched in global semantic map；And posture information updating block, in response to described complete The local semantic map is matched in the semantic map of office, the local posture information is updated according to the result of matching.

According to the another aspect of the application there is provided a kind of movable equipment, including：Processor；Memory；And storage Computer program instructions in the memory, the computer program instructions cause described when being run by the processor The above-mentioned posture information of computing device determines method.

System is determined there is provided a kind of posture information according to the another aspect of the application, including：It is above-mentioned removable to set It is standby；And server apparatus, the semantic map of the overall situation for storing current mobile environment.

It is described according to the another aspect of the application there is provided a kind of computer program product, including computer program instructions Computer program instructions by processor when being run so that the above-mentioned posture information of the computing device determines method.

According to the another aspect of the application there is provided a kind of computer-readable recording medium, computer journey is stored thereon with Sequence is instructed, and the computer program instructions by processor when being run so that the above-mentioned posture information of the computing device is determined Method.

Compared with prior art, using determining method, device according to the posture information of the embodiment of the present application, removable set Standby, computer program product and computer-readable recording medium, can receive described may move obtained by environmental sensor and set For the sample data of current mobile environment just movable within, the sample data includes position data and view data；Extremely Few local posture information that the movable equipment is determined according to the position data；According to the part of the movable equipment Posture information and described image data are come the semantic map of part for building the current mobile environment, the local semantic map bag The semantic entity and its attribute information in the current mobile environment are included, the semantic entity is possible to influence mobile entity, The attribute information indicates the physical characteristic of the semantic entity；Obtain the semantic map of the overall situation of the current mobile environment； The local semantic map is matched in the global semantic map；And in response to being matched in the global semantic map The local semantic map, the local posture information is updated according to the result of matching.It therefore, it can efficiently and accurately obtain The posture information of movable equipment.

Brief description of the drawings

By the way that the embodiment of the present application is described in more detail with reference to accompanying drawing, the above-mentioned and other purposes of the application, Feature and advantage will be apparent.Accompanying drawing is used for providing further understanding the embodiment of the present application, and constitutes explanation A part for book, is used to explain the application together with the embodiment of the present application, does not constitute the limitation to the application.In the accompanying drawings, Identical reference number typically represents same parts or step.

Fig. 1 illustrates the block diagram that system is determined according to the posture information of the embodiment of the present application.

Fig. 2 illustrates the flow chart that method is determined according to the posture information of the embodiment of the present application.

Fig. 3 illustrates the flow chart that step is determined according to the local posture information of the embodiment of the present application.

Fig. 4 illustrates the flow chart of the local semantic map construction step according to the embodiment of the present application.

Fig. 5 illustrates the flow chart of the semantic map matching step according to the embodiment of the present application.

Fig. 6 illustrates the flow chart that sub-step is matched based on category attribute according to the embodiment of the present application.

Fig. 7 illustrates the flow chart that step is updated according to the local posture information of the embodiment of the present application.

Fig. 8 illustrates the flow chart for the additional step that method is determined according to the posture information of the embodiment of the present application.

Fig. 9 illustrates the schematic diagram that scene is determined according to the posture information of the embodiment of the present application first specific example.

Figure 10 illustrates the schematic diagram that scene is determined according to the posture information of the embodiment of the present application second specific example.

Figure 11 illustrates the block diagram of the posture information determining device according to the embodiment of the present application.

Figure 12 illustrates the block diagram of the movable equipment according to the embodiment of the present application.

Embodiment

Below, the example embodiment according to the application will be described in detail by referring to the drawings.Obviously, described embodiment is only Only be a part of embodiment of the application, rather than the application whole embodiments, it should be appreciated that the application is not by described herein The limitation of example embodiment.

Application general introduction

As described above, there is problems with existing high-precision localization method：

1) support equipment complex and expensive：Differential positioning method needs base station covering, and need high in the clouds on a large scale comprehensively Integrate communication correction；And traditional high-precision Orientation on map method is also required to high-end integrated navigation system and expensive laser radar is contour Capital equipment；

2) support system complexity, map maintenance cost are high：Large-scale Differential positioning needs the complicated resolving of high in the clouds operation Algorithm, to merge the correction data reduction noise of a large amount of base stations and provide the real time access of a large amount of mobile terminals；The high-precision map of tradition Making usually require using expensive professional sensor multi collect data, cost of manufacture is very high, it is difficult to accomplish a large amount of collection terminals Collection updates in real time, while the mass data collected needs the processed offline of complicated and time consumption (to be likely to require artificial) ability Obtain high-precision map.

For problems of the prior art, the basic conception of the application be propose a kind of posture information determine method, Posture information determining device, movable equipment, computer program product and computer-readable recording medium, it can pass through lifting The intelligent infrastructure overhead to mitigate server apparatus of movable equipment.Specifically, movable equipment (for example, vehicle) Local semantic map and positioning result are obtained according to common hardware first, for the Decision Control of Vehicular automatic driving, when entering one Walk from server apparatus (for example, high in the clouds) download to the high-precision overall situation semantic map when, semantic cartography can also be passed through and practised With obtaining global high fine positioning.

In other words, by present inventive concept, movable equipment can add structuring semantic marker to enter line position using common GPS Appearance is calculated, and special differential GPS base station is not essential.Due to making full use of existing mobile environment (for example, road) traffic base Infrastructure, therefore, with the characteristic that cost bottom is applied widely.Traditional localization method relies primarily on high in the clouds facility and calculating, Real-time and reliability are relatively low.On the contrary, in embodiments herein, by more intelligent movable equipment, greatly The stationkeeping ability of its own is enhanced, Real-time Decision rapidly intelligently more can be carried out according to scene, and in structuring Global localization result, therefore, the more fast and reliable intelligence of system on the whole are obtained in the case that semantic information is enough.

After the general principle of the application is described, carry out specifically to introduce the various non-limits of the application below with reference to the accompanying drawings Property embodiment processed.

Example system

As shown in figure 1, determining that system includes movable equipment 100 and server according to the posture information of the embodiment of the present application Equipment 200.

The movable equipment 100 can be moved in known or unknown mobile environment.When semantic in no priori When being moved in the unknown mobile environment of map, it can be carried out local pose and calculated, newly according to the environmental samples data collected The local semantic map of unknown mobile environment is built, and moves according to local positioning result control.When there is priori language When being moved in the known mobile environment of free burial ground for the destitute figure, it can carry out semantic using the semantic map in the semantic map drawn game portion of the priori Match somebody with somebody, so that the local positioning precision of oneself is improved, to generate more accurate efficient mobile control strategy.In addition, this is removable Equipment 100 can also upload onto the server the part semantic map in equipment 200.

The server apparatus 200 can receive the download request of movable equipment 100, detect whether there is movable equipment The semantic map of priori of 100 just current mobile environment movable within, if it is present the semantic map of the priori is supplied to The movable equipment 100.In addition, the server apparatus 200 can also receive the part of the movable equipment 100 upload semantically Figure, is merged to the semantic map of the overall situation of the semantic map of the part and priori, to realize the dynamic learning process of semantic map.

For example, the movable equipment 100 can be any kind of electronic equipment that can be moved in mobile environment.Example Such as, the mobile environment for example can be indoor environment and/or outdoor environment.Also, the movable equipment 100 can be used for respectively The mobile robot of kind of purposes, for example, it may be in such as vehicle, aircraft, spacecraft, water delivery vehicle etc traffic Instrument etc..Certainly, the application not limited to this.For example, it can also be sweeping robot, window wiping robot, air cleaning machine People, security robot, household electrical appliances management robot, prompting robot, patrol robot etc..

The server apparatus 200 can be cloud server, and its calculation processing power is stronger, and can draw comprising multiple processing Hold up and semantic map is merged.Certainly, the application not limited to this.For example, the server apparatus 200 can also be all or part of Ground is located at the local side of movable equipment 100, and forms the framework of distributed server.

It should be noted that the posture information shown in Fig. 1 determine system be for only for ease of understand spirit herein and Principle and show, embodiments herein not limited to this.For example, the movable equipment 100 and/or the server apparatus 200 can be with It is one or more.

After the general principle of the application is described, carry out specifically to introduce the various non-limits of the application below with reference to the accompanying drawings Property embodiment processed.For convenience, by using the vehicle and cloud server that are travelled on road as movable equipment 100 and the example of server apparatus 200 illustrate.

The illustrative methods of movable equipment

First, description is determined into method according to the posture information applied to movable equipment 100 of the embodiment of the present application.

As shown in Fig. 2 determine that method can apply to movable equipment 100 according to the posture information of the embodiment of the present application, And including：

In step s 110, the movable equipment obtained by environmental sensor is received just movable within when reach The sample data in rotating ring border, the sample data includes position data and view data.

Movable equipment 100 (for example, vehicle) can be moved in mobile environment (for example, road), while it can To catch the sample data of the mobile environment using the environmental sensor equipped thereon.

For example, the environmental sensor can be used for obtaining the movable equipment just current mobile environment movable within Sample data, it can be various types of sensors.For example, it can include：Image for catching view data is passed Sensor, it can be camera or camera array；Laser sensor for catching scan data；For obtaining removable set The GPS devices of standby 100 real-time location coordinates；For based on carrier phase observation data carry out real-time dynamic positioning RTK devices, For Inertial Measurement Unit (IMU) device for being positioned based on object three-axis attitude angle (or angular speed) and acceleration etc.. Certainly, the environmental sensor can also be other various devices, as long as it can produce the sample number for describing mobile environment According to.It should be noted that in embodiments herein, the environmental sensor need not use high end sensor, and can be with It is the sampler of low cost.

For example, basically, the sample data that environmental sensor is got can include the position data and the car of vehicle The view data of surrounding environment.For example, the position data can be the movable equipment 100 absolute location coordinates (for example, The longitude and latitude that GPS devices are directly obtained) or relative position coordinates (for example, by the rotation number of turns, the speed of wheel of vehicle Etc. obtained cumulative movement parameter, distance etc.).The view data can be that camera visual pattern or laser are swept Trace designs picture.Further, the sample data can also include attitude data, and it can be obtained by differential GPS exhausted To towards angle or relative orientation angle (for example, the cumulative movement ginseng that rotational angle by wheel of vehicle etc. is obtained Number, direction etc.).For example, the attitude data can be combined with position data and form pose data.

In the step s 120, the local posture information of the movable equipment is determined according at least to the position data.

In one example, position data (and attitude data) can be based only on to determine that the movable equipment is being worked as Local posture information in preceding mobile environment.

Simply, the movable equipment can be determined in current mobile environment only according to the position data gathered In local posture information.For example, can be directly using the GPS longitudes and latitudes parameter of vehicle as its local location coordinate, and lead to The Difference Calculation for crossing front and rear two frames GPS parameters goes out partially toward angle, and combines the local posture information of generation.Due to this pose Information relies solely on GPS acquisitions, so it is a kind of absolute local posture information.

Alternatively, it can also be determined according to the position data and attitude data (pose data) gathered described removable Local posture information of the equipment in current mobile environment.For example, can be according to the initial posture information and IMU during vehicle launch The parameter such as resulting cumulative movement distance and direction determines the local posture information of the vehicle.Due to this posture information only Motion change acquisition is only relied on, so it is a kind of relatively local posture information.

Blocked because GPS there may be, the factor such as multipath effect influences its setting accuracy, and cumulative movement parameter also may be used There can be the factors such as deviation accumulation and produce deviations, it is further possible to be merged to both results, to obtain more Plus accurately and reliably local posture information.

In another example, can also be further it is determined that during the local posture information of the movable equipment Introduce view data.Below, it will be illustrated with reference to Fig. 3.

As shown in figure 3, step S120 can include：

In sub-step S121, the absolute pose letter of the movable equipment is determined according at least to the position data Breath.

In sub-step S122, the relative pose letter of the movable equipment is determined according at least to described image data Breath.

In sub-step S123, the local position is generated according to the absolute posture information and the relative pose information Appearance information.

In addition to this traditional positioning devices of GPS, IMU, the camera and/or laser sensing equipped in vehicle The real-time vision image and/or scan image that device is gathered can also obtain local odometer after by algorithm process.Office Portion's odometer can for example embody the local motion difference of vehicle between two field pictures, and it can calculate vehicle in this two frame Between local relative displacement and turn to etc..In other words, by the way that IMU, GPS and the local odometer of camera are merged, it can obtain The more more accurate local positioning result of robust.For example, this fusion process can correct noise than larger local position Appearance information.

In step s 130, described in being built according to the local posture information and described image data of the movable equipment The local semantic map of current mobile environment, the local semantic map include semantic entity in the current mobile environment and Its attribute information, the semantic entity is possible to influence mobile entity, and the attribute information indicates the semantic entity Physical characteristic.

, can be further according in it and sample data it is determined that after the local posture information of the movable equipment View data builds local semantic map.

As shown in figure 4, step S130 can include：

In sub-step S131, the semantic entity in the current mobile environment is detected according to described image data.

For example, can detect there is which semantic entity in current mobile environment according to acquired image data.Example Such as, the semantic entity can be the entity that possible influence movable equipment 100 itself to move.Certainly, the application not limited to this. For example, in a broad sense, it can also be possible the influence entity that other concern objects are moved in addition to movable equipment 100. Because for example, although movable equipment 100 is vehicle, its during map is built, can with it is also contemplated that Other traffic entities (for example, pedestrian, bicycle etc.) that may occur on the road use the potential of the map as future Main body.

For example, in the case where movable equipment 100 is vehicle, the semantic entity can be wheeled road, curb, friendship Logical mark (for example, the dimensional mark such as signal lamp, camera, guideboard, the road surface such as lane line, stop line, crossing Mark), isolation strip, greenbelt etc..

As a rule, the semantic entity follows certain specification and with specific meaning.For example, it may have specifically Geometry (for example, circle, square, triangle, strip etc.), or may have specific signature identification (for example, two Tie up code etc.).In addition, it may be decorated with stop flag, mark of going slowly, front falling rocks mark etc. above, so as to embody containing for it Justice.

Specifically, sub-step S131 can include：Detecting and tracking identification is carried out to described image data；And according to inspection The result of Tracking Recognition is surveyed to determine the semantic entity in the current mobile environment.

For example, the detecting and tracking that each car can carry out semantic entity according to local computing is recognized.Specifically, it can pass through Machine learning model that great amount of samples trains is in advance based on to extract the character representation in described image data, and according to institute Character representation is stated to carry out semantic entity detection.For example, the machine learning model can use such as convolutional neural networks, depth The various machine learning models of confidence network etc are realized.

In sub-step S132, the semantic entity is determined according to local posture information and the described image data Attribute information.

Next, the semantic entity two that can be detected by the local posture information of movable equipment and based on view data Person is integrated to determine which attribute information each semantic entity includes.

For example, the attribute information may indicate that the physical characteristic of the semantic entity, such as described semantic entity may Influence the attribute of movable equipment 100 itself movement.Similarly, in a broad sense, it may also indicate that the semantic entity may Influence the attribute of other concern object movements outside movable equipment 100.

For example, basically, the attribute information can also be the spaces such as position, shape, size, the direction of each semantic entity Attribute information.In addition, the attribute information can be that (such as, each semantic entity is studied carefully for the category attribute information of each semantic entity Unexpectedly it is connecting way, curb, track and lane line, traffic sign, pavement marker, traffic lights, stop line, crossing, roadside It is any in trees or pillar etc.).

For example, sub-step S132 can include：Determined according to described image data the semantic entity with it is described can Relative position relation between mobile device；And according to the local posture information and the relative position relation to determine State the space attribute information of semantic entity.

For example, each car is after the detecting and tracking identification of semantic entity is carried out according to local computing, then according to tracking Sequence and Ben Che fusion GPS/RTK and semantic position location direction, calculate the space attribute of semantic marker.For example, the space Attribute can include the size of semantic marker, shape, direction, height, the various attributes related to spatial character such as occupy.

In addition to space attribute information, for example, it is also possible to according to view data (for example, the detecting and tracking of image is recognized Result), further determine that out the classification of each semantic entity.

In sub-step S133, the local semantic map is built according to the semantic entity and its attribute information.

Once it is determined that each semantic entity and its attribute information that current mobile environment includes, it is possible to these letters Breath is integrated, to build the local semantic map based on present frame sample data.That is, each frame semantic marker result is weighed The simultaneously attribute such as coal addition position size is built, the semantic marker map with absolute attribute is obtained.

In step S140, the semantic map of the overall situation of the current mobile environment is obtained.

Before step S110, afterwards or concurrently, the semantic map of the overall situation being generated in advance can be obtained, with root Determine there is which semantic entity in current mobile environment according to prior information.

For example, overall situation semanteme map can be stored in movable equipment 100, for example, it is probably in advance from service What device equipment 200 was downloaded, it is also possible to be stored in after being built before the movable equipment 100 local.Alternatively, the overall situation What semantic map can also be downloaded from server apparatus 200 in real time.

As a rule, for the semantic map of the high-precision overall situation, in systems can by the way of mass-rent on-line study come The semantic map of generation high accuracy.That is, each movable equipment 100 in system can be by the local semantic map locally generated Equipment of uploading onto the server 200 (for example, high in the clouds).Once server apparatus 200 acquires office from movable equipment 100 (vehicle) Portion's semanteme map, then can parse the vehicle and currently be in which bar road, have which semantic entity (can on the road Trade road, curb, track and lane line, traffic sign, pavement marker, traffic lights, stop line, crossing, roadside trees post Son etc.) and its corresponding attribute (position, size, direction, classification etc.), and these semantic entities and its attribute are constantly merged, To cause map more complete, and improve its precision.

That is, when a road is also without high-precision map, the mass-rent vehicle equipped with relevant device and algorithm passes through The high-precision map in this road part can be generated and (be limited to the consideration of camera perspective and semantic marker confidence level, this possible locomotive Only upload part map), as this road vehicles increases by number of times, the integrality of one side map is become better and better, separately The precision (attribute accuracy of semantic marker, such as position coordinates, classification, size, direction etc.) of one side map can also be improved.

In one example, step S140 can include：Downloaded according to the position data from server apparatus The global semantic map.

For example, the movable equipment 100 (vehicle) can be communicated and be tasted with server apparatus 200 (cloud server) Examination obtains the semantic map of the overall situation for the priori for deserving preceding mobile environment.

For example, the priori can be obtained from server apparatus 200 according to the thick gps coordinate of the vehicle determined before Global semanteme map.It is of course also possible to semantic based on the priori is downloaded by the revised local posture information of view data Map.Alternatively, the map etc. can also be obtained according to the movement locus of the vehicle.

If there is no the semantic map of the overall situation to priori, illustrate that current road did not had any mass-rent vehicle once Pass by.It is possible to directly according to the non-semantic numerical map (traditional numerical map) of the current mobile environment Mobile control strategy is generated with thick gps coordinate (or revised local posture information etc.), the mobile control strategy is used for The movable equipment is controlled to allow it to realize the predetermined mobile purpose in the current mobile environment.For example, relying on this The sensor on ground combines existing numerical map and carries out local slow-path planning and control, such as determines the target track to be walked Road, the area of feasible solutions obtained according to local mobile terminal, connecting way carry out planning control, according to obtained real-time accurate track Line carries out lane-change or cruise.

If the semantic map of the overall situation for obtaining priori, this method enters step S150.

In step S150, the local semantic map is matched in the global semantic map.

If the movable equipment 100 (vehicle) obtains the overall situation of priori from server apparatus 200 (cloud server) Semantic map, then illustrate had mass-rent vehicle once to pass by before current road.It is possible to global semantic map and Local semanteme map is matched, to obtain Orientation on map result.

In step S160, in response to matching the local semantic map in the global semantic map, according to The result matched somebody with somebody updates the local posture information.

When a road is schemed in whole or in part, the semantic marker result of camera video sequence can be with high in the clouds Figure carries out matching search, obtains Orientation on map result, then carries out merge the more excellent pose of acquisition defeated with IMU, GPS, odometer Go out.

Below, it will be explained in detail step S150 and S160 implementation procedure.

As shown in figure 5, step S150 can include：

In sub-step S151, the global semantic map is parsed, to determine semantic entity therein and its category Property information.

Once movable equipment 100 (vehicle) obtains global semantic map from server apparatus 200 (cloud server), then Can parse current mobile environment all include which road, have on each road which semantic entity (connecting way, curb, Track and lane line, traffic sign, pavement marker, traffic lights, stop line, crossing, roadside trees pillar etc.) and its phase The attribute (position, size, direction, classification etc.) answered.

In sub-step S152, found according to attribute information in the global semantic map and the local semantic map Semantic entity matching pair.

For example, simply, it can come according to the position coordinates of each semantic entity to the complete of local semantically figure and high in the clouds The semantic map of office is matched.That is, above-mentioned matching operation is performed using the thick gps coordinate of vehicle.

However, blocked because GPS there may be, the factor such as multipath effect influences its setting accuracy, and cumulative movement is joined Number also likely to be present the factors such as deviation accumulation and produce deviations, can not be real so the positioning for being based only on position coordinates is possibly Now preferably match.

In the first example, measurement scaling matching operation can be performed based on the category attribute of semantic entity.Measurement contracting It is that local map may have scale difference with true map to put matching, so can ignore when matching in yardstick influence either Face adds a scale factor.Wherein, measurement just refers to relative position, shape size, the height obtained according to local posture information The attribute informations such as size.

As shown in fig. 6, sub-step S152 can include：

In action S1521, effective semantic entity in the local semantic map is determined, effective semantic entity is Its confidence level is more than or equal to the semantic entity of predetermined threshold.

In action S1522, effective semantic entity is determined according to the attribute information of effective semantic entity Effective orientation and non-effective orientation.

In action S1523, measurement scaling is carried out to the local semantic map in the non-effective orientation Match somebody with somebody.

In action S1524, the result of matching is scaled according to measurement to find the semantic entity matching pair.

, can be according to the high-precision language of acquisition for example, in order to obtain more accurate matching result, improve matching confidence level Free burial ground for the destitute figure and local semantically figure (can be the semantic map of single frames generated based on single frames sample data or will be many The semantic map of multiframe of the semantic map conflation of individual single frames) inner maximally effective semantic marker (semantic entity) classification, take not Same matching scaling algorithm, current local pose accuracy is corrected according to match flag classification.

For example, semantic marker different classes of in means of transportation has a positioning action of different directions, wherein pavement marker, The dimensional marks such as sign board, horizontal crossing can correct vehicle, and in longitudinal direction, (that is, track bearing of trend, it is typically parallel to vehicle Travel direction) on positioning precision, the pavement marker such as lane line, curb can correct vehicle in laterally (that is, deviation side To it is typically off in vehicle heading, such as vertical with vehicle heading or intersect at a certain angle) on positioning accurate Degree.

Such as, in the case of lane line and curb are reliable, measurement scaling can be carried out to the longitudinal direction of local semantically figure Matching；And can semantically scheme laterally to carry out measurement scaling matching to local in the case of traffic lights, guideboard are reliable.So, So that the matching process of local semanteme map and global semantic map has certain yardstick robustness.

As shown in fig. 7, sub-step S160 can include：

In sub-step S161, in the non-effective orientation, according to semantic entity matching to it described in The attribute information of semantic entity in global semanteme map matches the part to it described in semantically to correct the semantic entity The attribute information of semantic entity in figure.

In sub-step S162, the attribute information of the semantic entity in the revised local semantic map is come school Just described local posture information.

Because effective semantic entity in local semantic map has very high confidence level in effective orientation, so The amendment in non-effective orientation can be carried out only according to the attribute information of the matching semantic entity in global semantic map should The attribute information (for example, position, distance, shape, height etc.) of effective semantic entity, so as to update the local position of movable equipment Appearance information.It is of course also possible to carry out updating in effective orientation and non-effective orientation, to obtain more simultaneously simultaneously Comprehensive effect.

Alternatively or cumulatively, in the second example, topology metric can be performed based on the category attribute of semantic entity Semantic matches are operated.Topology metric semantic matches are due to if only based on the semantic map of single frames or the semantic map of several frames Matched, several matchings may be simultaneously scanned in nigh map, but it is true to be actually only possible to one Match somebody with somebody, therefore, by the semantic map match in track being spliced into by the semantic map of most several frames, those can be mismatched and removed. Wherein, measurement just refers to the attribute informations such as relative position, shape size, the height size obtained according to local posture information.Open up Flutter refer to each semantic entity appearance order, for example, around, upper inferior attribute of a relation (without considering occurrence).

As shown in fig. 6, sub-step S152 can also include：

In action S1525, topology metric is carried out using the local semantic map and the global semantic map semantic Matching.

In action S1526, the semantic entity matching pair is found according to the result of topology metric semantic matches.

When nearby there is close repetition scene in local semantically figure, it is possible to use the part having built up is semantically Figure carries out overall topology metric semantic matches with the semantic map of the high accuracy obtained, to correct current according to match flag classification Local pose accuracy.

Correspondingly, as shown in fig. 7, sub-step S160 can also include：

In sub-step S163, the semantic entity to it described in global semanteme map is matched according to the semantic entity Attribute information come the semantic entity effective orientation.

In sub-step S164, in effective orientation, matched according to the semantic entity complete to it described in The attribute information of semantic entity in the semantic map of office matches the local semanteme map to it described in correct the semantic entity In semantic entity attribute information.

In sub-step S165, the attribute information of the semantic entity in the revised local semantic map is come school Just described local posture information.

With the first example similarly, because the positioning that semantic marker different classes of in means of transportation has different directions is made With, it is possible to effective orientation of the matching semantic entity in global semantic map is determined first, and it is effectively fixed at this The attribute information for matching semantic entity in local semantic map is corrected in the direction of position (for example, position, distance, shape, height Deng), so as to update the local posture information of movable equipment.

It should be noted that first example and the second example can be implemented separately, realization can also be combined, in combination feelings Under condition, the execution sequence of the first example and the second example does not constitute the restriction to the application.But it is preferred that show due to first Example is based only on single frames or the local semantic map of several frames to be matched, and the amount of calculation that it needs is smaller, and the second example To be matched based on the local semantic map in track, the amount of calculation that it needs is larger, so it is preferred that can first carry out first The matching process of example, then perform the matching process of the second example.

So, when the matching confidence level obtained by based on the local semantic map of single frames or several frames is not high, rail is utilized Mark semanteme map carries out overall topology metric semantic matches, can weed out and mismatch noise, further improve positioning precision, school Just previous resulting local posture information.After more accurate local positioning is obtained, movable equipment 100 can also enter Control is moved to one step based on the result.

In addition, it is other one or more additional to determine that method can also include according to the posture information of the embodiment of the present application Step.

As shown in figure 8, determining that method can also include according to the posture information of the embodiment of the present application：

In step S170, mobile control is generated according to the local posture information after the global semantic map and renewal Strategy, the mobile control strategy is realized in the current mobile environment for controlling the movable equipment to allow it to Predetermined mobile purpose.

, can be according to renewal in the case of the confidence level sufficiently high (positioning successfully) of the semantic map match of global and local Local posture information move accordingly decision rule, such as to realize purpose that automatic Pilot/auxiliary drives, example Such as shift to an earlier date path planning, reduce oil consumption, hide accident etc..

In step S180, in response to being not matched to the local semantic map, root in the global semantic map Mobile control strategy is generated according to the non-semantic numerical map and original local posture information of the current mobile environment, it is described The predetermined movement that mobile control strategy is used to controlling the movable equipment to allow it to realizing in the current mobile environment Purpose.

Do not matched with global semantic map and in the case of high accuracy positioning can not being obtained, movable equipment 100 Local sensory perceptual system combination rough grade positioning map (for example, existing numerical map) progress can simply be relied on local low Fast path planning and control, such as determine the target road, the area of feasible solutions obtained according to local mobile terminal, connecting way to be walked Carry out planning control, lane-change or cruise are carried out according to obtained real-time accurate lane line.

Further, while the mobile control strategy of movable equipment 100 is generated, if it is possible to ensureing removable While dynamic equipment 100 safely realizes mobile purpose, semantic map match is further taken into account, this is clearly desired.

Therefore, after step S170 or step S180, in step S182, according to the position data and the overall situation Semantic map determines the auxiliary control strategy of the movable equipment, and the auxiliary control strategy is used to controlling described removable Equipment allows it to the semantic entity that will appear from current mobile environment described in active obtaining.

In step S184, the mobile control strategy and the auxiliary control strategy are integrated, to generate synthesis Control strategy.

It is not high enough in the confidence level of the semantic map match of global and local, that is, confidence level is positioned still than relatively low situation Under, the semantic map of high accuracy that can be positioned and obtain according to rough grade is combined, and safety, comfortableness and destination are not influenceed Active decision (such as, vehicle lane-changing, towards adjustment, in-vehicle camera focal length towards adjustment etc.), may have to obtain on one's own initiative The video sequence of crucial semantic marker (at this position, for the seeking semantics mark that laterally longitudinal direction plays a decisive role) is examined Survey Tracking Recognition and matched with the semantic map of high accuracy, improve pose accuracy.

In step S190, the local semantic map is uploaded onto the server equipment.

After local semantic map is obtained, in order to realize mass-rent pattern, each movable equipment can also be by the map Equipment of uploading onto the server 200 (for example, high in the clouds), to realize the technique effect of map dynamic renewal.

For example, the map after local integrate can be uploaded to high in the clouds by vehicle, overall transmitted data amount is very little, and And be that figure process is built in automation, it is not necessary to artificial mark.For example, at once by it after the local semantic map of a frame can be generated High in the clouds is uploaded to, can also be the semantic map in track by the local semantic Map Generalization of multiframe, upload operation is then performed again.Example Such as, the semantic map in a track can be formed based on the time (for example, at regular intervals), other conditions (example can also be based on Such as, since being driven into a road untill the road is rolled away from) the semantic map in triggering generation track.

As can be seen here, method is determined using the posture information according to the embodiment of the present application, can received by environmental sensor The sample data of the movable equipment obtained just current mobile environment movable within, the sample data includes position Data and view data；The local posture information of the movable equipment is determined according at least to the position data；According to institute The local posture information and described image data of movable equipment is stated to build the local semantic map of the current mobile environment, The local semantic map includes the semantic entity and its attribute information in the current mobile environment, and the semantic entity is can Mobile entity can be influenceed, the attribute information indicates the physical characteristic of the semantic entity；Obtain the current shift(ing) ring The semantic map of the overall situation in border；The local semantic map is matched in the global semantic map；And in response to described complete The local semantic map is matched in the semantic map of office, the local posture information is updated according to the result of matching.Therefore, The posture information of movable equipment can efficiently and accurately be obtained.

Specifically, embodiments herein has advantages below：

1) positioning precision is improved using semantic marker characteristic, such as lane line, the curb on one road etc. can improve transverse direction Positioning precision, pole, trees, sign board etc. can improve longitudinal register precision, and stop line, pavement, traffic lights can be improved Crossing positioning precision；

2) the local semantic marker map match obtained using track (high-precision local odometer) obtains the high-precision overall situation Since positioning result, such as one car i.e. can intactly detecting and tracking lane line entering road, then on this road Any moment vehicle is skew of the track where can obtaining it relative to original lane, and this information can reduce single frames car What diatom was positioned mismatches.The local semantic marker map of vehicle is (with global map, being simply that this is collected, also There is no world coordinates) with server end map match (being the matching of sequence mark-on will), global map matching positioning can be improved Precision.

Specific example

Next, description to be determined to the specific example of method according to the posture information of the embodiment of the present application.Wherein assuming can Mobile device is that vehicle, server apparatus are cloud server.

In the specific example of the embodiment of the present application, local ability is first depending on after the start of mobile terminal vehicle and carries out local build Figure and perception, including road pavement, lane line, curb, area of feasible solutions and land marking, sign board, traffic lights, stop line, people's row The detecting and tracking identification and reconstruction of lateral road etc., and the Local Phase obtained according to GPS, IMU, visual odometry etc. is to positioning pose note Volume obtains local semantic marker map into local map, and vehicle can enter according to this local semantic map and real-time perception Row Decision Control.Meanwhile, according to GPS location, vehicle can be matched local semantic map with the semantic map of the overall situation in high in the clouds, After confidence level is matched more than certain threshold value, local semanteme map can carry out global position correction, obtain accurately global pose, And then vehicle can carry out more complicated controlling with consistent Global motion planning.Local semanteme map is matched with global semantic map It is a dynamic process, different classes of semantic marker can improve the pose confidence level of different directions, overall local path Semantic map match can further elite position cancelling noise.

In following first specific example, will stress be how based on the semantic marker in the semantic map of single frames come Improve positioning precision.

Different classes of semantic marker has different positioning actions, wherein pavement marker, sign board, horizontal stroke in means of transportation The longitudinal register of vehicle can be corrected to crossing etc., lane line (and classification), curb etc. can correct the located lateral of vehicle, such as Shown in Fig. 9, when vehicle is travelled to A points, car can only obtain a position location substantially, because the semantic marker now run into It is not obvious enough；When vehicle drives to B points, because mobile terminal has recognized accurate sign board mark and with the semantic map in high in the clouds Matched, therefore longitudinal direction of car positioning precision is improved, but transverse precision is still not enough；When vehicle drives to C points Wait, road occurs in that lane line and curb and vehicle can accurately identify modeling, therefore lateral direction of car positioning precision is also obtained Correction, overall precision is further improved, and more preferable foundation is provided for follow-up decision.

In following second specific example, will stress it is how positioning to be improved based on the semantic map of local path Precision.

Single structure semantic marker is easy to repeat (such as, have multiple speed(-)limit signs near gps coordinate in map Board), it is thus possible to positioning can be caused deviation occur, as shown in Figure 10, when vehicle is recognized near GPS location and travel During one similar round sign board, due to nearby having multiple same category of sign boards (while being present in position A, position B and position C same positioning confidence level), therefore at several sign boards is obtained.With moving ahead for vehicle, when running into another kind of triangle mark Will, now have in local path has a similar round and a class triangle sign board in order respectively, due to nearby there is such distribution The number of the sign board of form is tailed off and (is merely present in position B and position C), and so basis is further picked with semantic map match Except interference, the confidence level of possible position is improved.Moved ahead with the continuation of vehicle, when driving at the square sign board of a class, by (uniquely exist in the distribution form for there was only the classification at one with this same three classes sign board and sequence of positions on neighbouring road In position C), therefore obtain the final positioning posture of very high confidence level.

Therefore, the positioning of automatic driving vehicle can be realized using the equipment of more cheap and simple using this programme, most directly The equipment such as camera, GPS, IMU that is such as provided with there is lane line, traffic sign and Pedestrians and vehicles detecting and tracking identification function The senior drive assist system of vision (ADAS) system, while this programme compatible high differential GPS or binocular, low line number laser thunder Up to etc. system.Determine in addition, this programme is perceived existing semantic marker and realized based on the common device for currently having had extensive application Position, it is not necessary to which substantial amounts of infrastructure rebuilding and cooperation, support system and cloud system are simple, the making of positioning map, storage, Processing, transmission cost are low.Finally, that mobile terminal is had is stronger intelligent for this programme, can only rely only on the energy of mobile terminal Power carry out it is local builds figure and perceives decision-making, and beyond the clouds global map can with confidence level it is higher in the case of to carry out global pose strong Just, therefore, this programme has more preferable adaptability, robustness and real-time.

It should be noted that because this programme greatly strengthens and make use of the intelligent of mobile vehicle end, therefore complete When office's absolute fix has interference or confidence level not high enough, vehicle can still be carried out according to this local map and real-time perception The real-time positioning and decision-making of vehicle, therefore, this programme has more preferable adaptability and robustness.

Exemplary pose information determining means

As shown in figure 11, it can apply to may move according to the posture information determining device 300 of the embodiment of the present application Equipment 100, and can include：Sample data receiving unit 310, for receive by environmental sensor obtain it is described removable The sample data of equipment just current mobile environment movable within, the sample data includes position data and view data； Posture information determining unit 320, for determining that the local pose of the movable equipment is believed according at least to the position data Breath；Semantic map constructing unit 330, for the local posture information and described image data according to the movable equipment come structure The local semantic map of the current mobile environment is built, the local semantic map includes the semanteme in the current mobile environment Entity and its attribute information, the semantic entity are possible to influence mobile entity, and the attribute information indicates the semanteme The physical characteristic of entity；Semantic map acquiring unit 340, the semantic map of the overall situation for obtaining the current mobile environment；Language Free burial ground for the destitute figure matching unit 350, for matching the local semantic map in the global semantic map；And posture information is more New unit 360, in response to matching the local semantic map in the global semantic map, according to the result of matching To update the local posture information.

In one example, the posture information determining unit 320 can determine described according at least to the position data The absolute posture information of movable equipment；The relative pose letter of the movable equipment is determined according at least to described image data Breath；And the local posture information is generated according to the absolute posture information and the relative pose information.

In one example, the semantic map constructing unit 330 can detect described current according to described image data Semantic entity in mobile environment；The category of the semantic entity is determined according to local posture information and the described image data Property information；And the local semantic map is built according to the semantic entity and its attribute information.

In one example, the semantic map constructing unit 330 can carry out detecting and tracking identification to described image data； And the result recognized according to detecting and tracking determines the semantic entity in the current mobile environment.

In one example, the semantic map constructing unit 330 can determine the semanteme according to described image data Relative position relation between entity and the movable equipment；And according to the local posture information and the relative position Relation determines the space attribute information of the semantic entity.

In one example, the semantic map acquiring unit 340 can be according to the position data come from server apparatus It is middle to download the global semantic map.

In one example, the semantic map matching unit 350 can be parsed to the global semantic map, with true Fixed semantic entity and its attribute information therein；And the global semantic map and the part are found according to attribute information Semantic entity matching pair in semantic map.

In one example, the semantic map matching unit 350 can determine effective language in the local semantic map Adopted entity, effective semantic entity is the semantic entity that its confidence level is more than or equal to predetermined threshold；According to effective language Adopted entity attributes information determines the effective orientation and non-effective orientation of effective semantic entity；Described non- Measurement scaling matching is carried out to the local semantic map in effective orientation；And according to measurement scale matching result come Find the semantic entity matching pair.

In one example, the posture information updating block 360 can be in the non-effective orientation, according to described Semantic entity matches the attribute information of the semantic entity in global semanteme map to it described in correct the semantic entity The attribute information of semantic entity described in pairing in local semanteme map；And according to the revised part semantically The attribute information of semantic entity in figure corrects the local posture information.

In one example, the semantic map matching unit 350 can utilize the local semantic map and the overall situation Semantic map carries out topology metric semantic matches；And the semantic entity is found according to the result of topology metric semantic matches Matching pair.

In one example, the posture information updating block 360 can be according to semantic entity matching to it described in The attribute information of semantic entity in global semanteme map carrys out effective orientation of the semantic entity；In effective positioning In direction, match the attribute information of the semantic entity to it described in global semanteme map to correct according to the semantic entity The attribute information of semantic entity of the semantic entity matching to it described in local semanteme map；And according to revised The attribute information of semantic entity in the local semantic map corrects the local posture information.

In one example, it can also be included according to the posture information determining device 300 of the embodiment of the present application：Control Strategy generating unit, for generating mobile control strategy.

In one example, the control strategy generation unit can be according to the part after the global semantic map and renewal Posture information generates mobile control strategy, and the mobile control strategy is used to control the movable equipment to allow it to reality Predetermined mobile purpose in the existing current mobile environment.

In one example, the control strategy generation unit can be in response to not matching in the global semantic map To the local semantic map, according to the non-semantic numerical map of the current mobile environment and original local posture information come The mobile control strategy of generation, the mobile control strategy is used to controlling the movable equipment to allow it to realizing described current Predetermined mobile purpose in mobile environment.

In one example, the control strategy generation unit can be according to the position data and the global semantic map To determine the auxiliary control strategy of the movable equipment, the auxiliary control strategy be used to control the movable equipment so that It can be the semantic entity that will appear from current mobile environment described in active obtaining；And to the mobile control strategy and institute State auxiliary control strategy to be integrated, to generate integrated control strategy.

In one example, it can also be included according to the posture information determining device 300 of the embodiment of the present application：It is semantic Map uploading unit, for the local semantic map to be uploaded onto the server equipment.

Unit and the concrete function of module and operation in above-mentioned posture information determining device 300 have been described above ginseng The posture information for examining Fig. 1 to Figure 10 descriptions determines to be discussed in detail in method, and therefore, will omit its repeated description.

It should be noted that a software mould can be used as according to the posture information determining device 300 of the embodiment of the present application Block and/or hardware module and be integrated into movable equipment 100, in other words, the movable equipment 100 can include the pose letter Cease determining device 300.For example, the posture information determining device 300 can be one in the operating system of the movable equipment 100 Individual software module, or can be directed to its application program developed；Certainly, the posture information determining device 300 Equally can be one of numerous hardware modules of movable equipment 100.

Alternatively, in another example, the posture information determining device 300 and the movable equipment 100 can also be point Vertical equipment (for example, server), and the posture information determining device 300 can be connected by wired and or wireless network Interactive information is transmitted to the movable equipment 100, and according to the data format of agreement.

Exemplary movable equipment

Below, it is described with reference to Figure 12 the movable equipment according to the embodiment of the present application.

As shown in figure 12, movable equipment 100 includes one or more processors 11 and memory 12.

Processor 11 can be CPU (CPU) or with data-handling capacity and/or instruction execution capability Other forms processing unit, and the other assemblies in movable equipment 100 can be controlled to perform desired function.

Memory 12 can include one or more computer program products, and the computer program product can include each The computer-readable recording medium of the form of kind, such as volatile memory and/or nonvolatile memory.The volatile storage Device is such as can include random access memory (RAM) and/or cache memory (cache).It is described non-volatile to deposit Reservoir is such as can include read-only storage (ROM), hard disk, flash memory.It can be deposited on the computer-readable recording medium One or more computer program instructions are stored up, processor 11 can run described program instruction, to realize this Shen described above Posture information in the movable equipment 100 of each embodiment please determines method and/or other desired functions.

In one example, movable equipment 100 can also include：Input unit 13 and output device 14.

For example, the input unit 13 can include such as keyboard, mouse and communication network and its connected it is long-range defeated Enter equipment etc..

For example, the input unit 13 can include environmental sensor, just moved wherein for obtaining the movable equipment The sample data of dynamic current mobile environment.For example, the environmental sensor can be the image sensing for catching view data Device, it can be camera or camera array.And for example, the environmental sensor can be the laser for catching scan data Sensor, it can be laser or laser array.And for example, the environmental sensor can also be motion sensor, be configured to obtain Take the exercise data of the movable equipment 100.For example, the motion sensor can be built-in inertia in movable equipment Measuring unit and motion encoder (including accelerometer with gyroscope etc.), the kinematic parameter for measuring movable equipment, example Such as, speed, acceleration, displacement etc., to determine position of the movable equipment in mobile environment with towards (posture), and may be used also To be built-in magnetometer etc., with the cumulative errors of real time calibration attitude transducer.So, more accurate pose can be obtained Estimation.Certainly, the application not limited to this.The environmental sensor can also be other various devices of radar etc.In addition, also may be used To gather the sample data using other discrete environmental sensors, and it is sent to movable equipment 100.

Output device 14 can export various information etc. to outside (for example, user).The output equipment 14 can include example Such as loudspeaker, display, printer and communication network and its remote output devices connected etc..

Certainly, to put it more simply, illustrate only one in component relevant with the application in the movable equipment 100 in Figure 12 A bit, the component of such as bus, input/output interface etc. is eliminated.It should be noted that movable equipment 100 shown in Figure 12 Component and structure are illustrative, and not restrictive, and as needed, movable equipment 100 can also have other assemblies And structure.

Although for example, not shown, movable equipment 100 can also be including communicator etc., and communicator can pass through net Network or other technologies communicate with other equipment (for example, personal computer, server, mobile station, base station etc.), and the network can be with It is internet, WLAN, mobile communications network etc., the other technologies can for example include Bluetooth communication, infrared communication Deng.

Illustrative computer program product and computer-readable recording medium

In addition to the above method and equipment, embodiments herein can also be computer program product, and it includes meter Calculation machine programmed instruction, the computer program instructions by processor when being run so that described computing device this specification is above-mentioned The posture information according to the various embodiments of the application described in " illustrative methods " part determines the step in method.

The computer program product can be write with any combination of one or more programming languages for holding The program code of row the embodiment of the present application operation, described program design language includes object oriented program language, such as Java, C++ etc., in addition to conventional procedural programming language, such as " C " language or similar programming language.Journey Sequence code can perform fully on the user computing device, partly perform on a user device, independent soft as one Part bag is performed, part is performed or completely in remote computing device on a remote computing on the user computing device for part Or performed on server.

In addition, embodiments herein can also be computer-readable recording medium, it is stored thereon with computer program and refers to Order, the computer program instructions by processor when being run so that above-mentioned " the exemplary side of described computing device this specification The posture information according to the various embodiments of the application described in method " part determines the step in method.

The computer-readable recording medium can use any combination of one or more computer-readable recording mediums.Computer-readable recording medium can To be readable signal medium or readable storage medium storing program for executing.Readable storage medium storing program for executing can for example include but is not limited to electricity, magnetic, light, electricity Magnetic, the system of infrared ray or semiconductor, device or device, or any combination above.Readable storage medium storing program for executing is more specifically Example (non exhaustive list) includes：Electrical connection, portable disc with one or more wires, hard disk, random access memory Device (RAM), read-only storage (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc Read-only storage (CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.

The general principle of the application is described above in association with specific embodiment, however, it is desirable to, it is noted that in this application Advantage, advantage, effect referred to etc. is only exemplary rather than limitation, it is impossible to which it is the application to think these advantages, advantage, effect etc. Each embodiment is prerequisite.In addition, detail disclosed above is merely to the effect of example and the work readily appreciated With, and it is unrestricted, above-mentioned details is not intended to limit the application to realize using above-mentioned concrete details.

The method that is related in the application, device, device, equipment, system block diagram only illustratively the example of property and It is not intended to require or implies to be performed in the way of square frame is illustrated, connect, arrange, configuring.Such as art technology What personnel will be recognized that, it can connect, arrange by any-mode, configuring these devices, device, equipment, system.Such as " bag Include ", "comprising", the word of " having " etc. be open vocabulary, refer to " including but is not limited to ", and can be with its used interchangeably.This In used in vocabulary "or" and " and " refer to vocabulary "and/or", and can be with its used interchangeably, unless context is explicitly indicated is not So.Vocabulary " such as " used herein above refers to phrase " such as, but not limited to ", and can be with its used interchangeably.

It may also be noted that in device, apparatus and method in the application, each part or each step are to decompose And/or reconfigure.These decompose and/or reconfigured the equivalents that should be regarded as the application.

The above description of disclosed aspect is provided so that any person skilled in the art can make or use this Application.Various modifications in terms of these are readily apparent to those skilled in the art, and defined herein General Principle can apply to other aspect without departing from scope of the present application.Therefore, the application is not intended to be limited to Aspect shown in this, but according to the widest range consistent with the feature of principle disclosed herein and novelty.

In order to which purpose of illustration and description has been presented for above description.In addition, this description is not intended to the reality of the application Apply example and be restricted to form disclosed herein.Although already discussed above multiple exemplary aspects and embodiment, this area skill Art personnel will be recognized that its some modifications, modification, change, addition and sub-portfolio.

Claims

1. a kind of posture information determines method, applied to movable equipment, methods described includes：

The sample data of the movable equipment obtained by environmental sensor just current mobile environment movable within is received, The sample data includes position data and view data；

The local posture information of the movable equipment is determined according at least to the position data；

The office of the current mobile environment is built according to the local posture information and described image data of the movable equipment Portion's semanteme map, the local semantic map includes the semantic entity and its attribute information in the current mobile environment, described Semantic entity is possible to influence mobile entity, and the attribute information indicates the physical characteristic of the semantic entity；

Obtain the semantic map of the overall situation of the current mobile environment；

The local semantic map is matched in the global semantic map；And

In response to matching the local semantic map in the global semantic map, updated according to the result of matching described in Local posture information.

2. the office of the movable equipment is the method for claim 1, wherein determined according at least to the position data Position appearance information includes：

The absolute posture information of the movable equipment is determined according at least to the position data；

The relative pose information of the movable equipment is determined according at least to described image data；And

The local posture information is generated according to the absolute posture information and the relative pose information.

3. the method for claim 1, wherein according to the local posture information and described image number of the movable equipment Include according to come the local semantic map that builds the current mobile environment：

The semantic entity in the current mobile environment is detected according to described image data；

The attribute information of the semantic entity is determined according to local posture information and the described image data；And

The local semantic map is built according to the semantic entity and its attribute information.

4. method as claimed in claim 3, wherein, the language in the current mobile environment is detected according to described image data Adopted entity includes：

Detecting and tracking identification is carried out to described image data；And

The result recognized according to detecting and tracking determines the semantic entity in the current mobile environment.

5. method as claimed in claim 3, wherein, determined according to local posture information and the described image data described in The attribute information of semantic entity includes：

The relative position relation between the semantic entity and the movable equipment is determined according to described image data；And

The space attribute information of the semantic entity is determined according to the local posture information and the relative position relation.

6. the semantic map of the overall situation for the method for claim 1, wherein obtaining the current mobile environment；

The global semantic map is downloaded from server apparatus according to the position data.

7. the local semantic map bag is the method for claim 1, wherein matched in the global semantic map Include：

The global semantic map is parsed, to determine semantic entity therein and its attribute information；And

The semantic entity matching pair in the global semantic map and the local semantic map is found according to attribute information.

8. method as claimed in claim 7, wherein, the global semantic map and the part are found according to attribute information In semantic map semantic entity matching to including：

Effective semantic entity in the local semantic map is determined, effective semantic entity is that its confidence level is more than or equal to The semantic entity of predetermined threshold；

Effective orientation of effective semantic entity is determined according to the attribute information of effective semantic entity and non-is had Imitate orientation；

Measurement scaling matching is carried out to the local semantic map in the non-effective orientation；And

The semantic entity matching pair is found according to the result of measurement scaling matching.

9. method as claimed in claim 8, wherein, the local posture information is updated according to the result of matching to be included：

In the non-effective orientation, the semanteme to it described in global semanteme map is matched according to the semantic entity Entity attributes information corrects the attribute of semantic entity of the semantic entity matching to it described in local semanteme map Information；And

The attribute information of semantic entity in the revised local semantic map corrects the local posture information.

10. method as claimed in claim 7, wherein, the global semantic map and the office are found according to attribute information Portion semanteme map in semantic entity matching to including：

Using the local semantic map topology metric semantic matches are carried out with the global semantic map；And

The semantic entity matching pair is found according to the result of topology metric semantic matches.

11. method as claimed in claim 10, wherein, the local posture information is updated according to the result of matching to be included：

The attribute information of semantic entity to it described in global semanteme map is matched come institute's predicate according to the semantic entity Effective orientation of adopted entity；

In effective orientation, the semanteme matched according to the semantic entity to it described in global semanteme map is real The attribute information of body is believed to correct the attribute of semantic entity of the semantic entity matching to it described in local semanteme map Breath；And

12. the method as described in claim 1, methods described also includes：

Mobile control strategy, the mobile control are generated according to the local posture information after the global semantic map and renewal Strategy is used to control the movable equipment to allow it to realize the predetermined mobile purpose in the current mobile environment.

13. the method as described in claim 1, methods described also includes：

In response to being not matched to the local semantic map in the global semantic map, according to the current mobile environment Non-semantic numerical map and original local posture information generate mobile control strategy, the mobile control strategy is used to control Make the movable equipment and allow it to realize predetermined mobile purpose in the current mobile environment.

14. method as claimed in claim 13, wherein, methods described also includes：

The auxiliary control strategy of the movable equipment is determined according to the position data and the global semantic map, it is described Auxiliary control strategy is used to control the movable equipment to allow it to go out in current mobile environment described in active obtaining Existing semantic entity；And

The mobile control strategy and the auxiliary control strategy are integrated, to generate integrated control strategy.

15. the method for claim 1, wherein methods described also includes：

The local semantic map is uploaded onto the server equipment.

16. a kind of posture information determining device, applied to movable equipment, described device includes：

Sample data receiving unit, for receiving, the movable equipment obtained by environmental sensor is just movable within to work as The sample data of preceding mobile environment, the sample data includes position data and view data；

Posture information determining unit, for determining that the local pose of the movable equipment is believed according at least to the position data Breath；

Semantic map constructing unit, builds for the local posture information and described image data according to the movable equipment The local semantic map of the current mobile environment, the semanteme that the local semantic map is included in the current mobile environment is real Body and its attribute information, the semantic entity are possible to influence mobile entity, and the attribute information indicates described semantic real The physical characteristic of body；

Semantic map acquiring unit, the semantic map of the overall situation for obtaining the current mobile environment；

Semantic map matching unit, for matching the local semantic map in the global semantic map；And

Posture information updating block, in response to matching the local semantic map, root in the global semantic map The local posture information is updated according to the result of matching.

17. a kind of movable equipment, including：

Processor；

Memory；And

The computer program instructions in the memory are stored in, the computer program instructions by the processor when being run So that method of the computing device as any one of claim 1-15.

18. a kind of computer-readable recording medium, is stored thereon with computer program instructions, the computer program instructions are in quilt Processor causes method of the computing device as any one of claim 1-15 when running.