DE102016213174A1 - Creation of a semantic card - Google Patents

Abstract

A device is adapted to move automatically within a predetermined area. A method of creating a semantic map using the device includes steps of moving the device in the area; detecting a wireless network device in the area of the device; determining a type of device; determining semantic information about the area based on the particular type; and assigning the semantic information to the area.

Description

The invention relates to the creation of a semantic card. In particular, the invention relates to the creation of a semantic card by means of an automatically moving device.

State of the art

A household appliance, for example in the form of an automatic vacuum cleaner, is adapted to move automatically within a building. To fulfill its purpose, the home appliance may require a topographical map. On the basis of this map, for example, a route from a current position to a charging station can then be planned or controlled. Further, for example, a first bottom portion to be cleaned and a second bottom portion which is not to be cleaned can be discriminated from each other by the map. Even obstacles such as a step or a door can be noted in the map.

The utility of the household appliance can be increased if additional semantic information is stored in the topographical map. In particular, semantic information may relate to the nature of a fixed object, area, or space in which the home appliance is traveling. A semantic information may, for example, relate to a predominant usage of a room. In a kitchen there is usually a different need for cleaning than in a living room, so that on the basis of the semantic information of the room type or its preferred mode of use, the operation of the household appliance can be improved.

An object of the present invention is to provide a technique for improved creation of a semantic card. The invention solves this problem by means of the subjects of the independent claims. Subclaims give preferred embodiments again.

Disclosure of the invention

A device is adapted to move automatically within a predetermined area. A method of creating a semantic map using the device includes steps of moving the device in the area; detecting a wireless network device in the area of the device; determining a type of device; determining semantic information about the area based on the particular type; and associating the semantic information with the domain. In particular, the apparatus may comprise a robot for performing a predetermined operation, such as an automated home appliance or a household robot such as an automatic lawnmower or an automatic dry or wet floor cleaner.

In cartography, the accuracy of a map is usually distinguished by geometric and semantic (thematic) accuracy. A map is geometrically accurate if the used map symbols (signatures) depict the space-related objects corresponding to them in the correct position and in the correct geometrical context within the context of the scaling and generalization-related representation possibility. It is semantically accurate if the map symbols used in the map correctly reflect the meaning of their corresponding spatial objects. Consequently, in the context of the present disclosure, a semantic card may, in particular, be understood as a card which comprises card characters having a meaning.

The invention is based on the idea that wirelessly networked devices, as they are generally assigned to the Internet of Things (IoT), by virtue of their type, permit inference to semantic information of the area in which they are respectively arranged. In other words, due to its proximity to an IoT device-and possibly further device information-the device may generate semantic information about the area that may be noted in a particular topographic map, later for improved positioning of the device or other device to be used. Exemplary types of apparatus include a door, elevator, home appliance such as a dishwasher or stove, or a fixed or mobile sensor.

The method may not require active, in particular manual, provision of semantic information about a region of the region. The device may automatically collect and interpret information to generate or validate the sematic information, for example, as part of its dedicated activity or orienteering to scan the topography of the area. The collected semantic information can later be used to position or control the device or other device. The positioning may relate to the determination of the position of the device, a movement or the control of a function of the device, such as a transport of an object in the area or a cleaning of a portion of the area. In particular, in a sprawling area such as a hospital, a parking lot, a sports complex, an industrial plant or a catering facility can by the semantic information associated with the area allows for improved operation of the other device. The semantic information map can also be used to better perform an activity of the device. For example, a task that has been set can be specified using the semantic information.

The semantic card may be used to position a device, where the positioning may include determining or approaching a position. To move to a position, in particular a movement path, a movement type or a movement time of the device can be adapted. For example, a subtask of the device may include disposing of previously collected debris or waste. Based on the semantic information, it can be determined where in the area the previously collected debris or waste can be disposed of preferentially. The movement type may, for example, relate to a movement speed. If the semantic information indicates, for example, that the device is in a bathroom, a slower movement type can be selected on the ground due to the risk of slipping on the floor. In addition, the bathroom can be shunned at certain times of the day, for example between 6 and 8 o'clock in the morning, when it is likely to be frequented by people.

The area may include a building, the area may be in a room of the building, and based on the particular type, semantic information may be determined as a type of use of the space. The type of use often indicates a dedication of the room, for example, to a bathroom, a toilet, a living room, a bedroom, a hall, a kitchen or a garage. With the type of use or dedication of the room are usually temporally determined patterns of use by people, so that the device can better decide at what times in the room are people and whether they should be avoided or targeted search. In addition, the usage type semantic information may determine when or how the device performs a task in the room. For example, a cleaning robot may preferentially wet clean a kitchen or bathroom, while dry cleaning may be preferred in a hallway or garage.

It is particularly preferable that the types of a plurality of wirelessly networked devices are detected in space and a usage mode of the room is determined on the basis of the particular types. For example, if one of the devices includes a refrigerator and another a toaster, the space may be identified as a kitchen. A combination of a refrigerator and a screen, on the other hand, may indicate a laboratory.

It is further preferred that communication with the networked device is performed and the type of device is determined based on networked device identification data. In order to be able to conclude from the identification data on the type of the device, a database can be provided. For example, the identification may be "Bosch Series 6 Avantixx" and the database may determine the type of device as a washing machine.

In one embodiment, determining the type includes querying a central entity (server), particularly based on the identification data. On the part of the central entity, a corresponding database can be kept up-to-date in an up-to-date manner, so that, for example, it is also possible to determine the type of device that was produced after the production of the device. In addition, for example, an update of the networked device can be reflected improved in the central database.

In another embodiment, communication of the device with the networked device is performed, wherein the semantic information is determined based on user data of the networked device. For example, a temperature sensor may provide a certain temperature or a humidity sensor may provide certain humidity as information. From such information semantic information can be determined, which notes in the map and later contribute to the improved positioning or navigation of another automatically movable device. If, for example, the sensor information points to open windows and doors, so that a draft is to be expected, an airworthy domestic appliance (drone) can incorporate the danger of a gust of wind into its movement planning.

In yet another embodiment, a moving object in the area of the device is determined and the semantic information is validated on the basis of the moving object. For example, if it has previously been determined that a predetermined area of the building is a aisle, tracking a trajectory of a person or determining traffic within the area may validate or falsify that estimate.

In yet another embodiment, the semantic information includes a determination time. It is preferred that provisions type, payload, or moving object information at different times of the day or week to improve the particular semantic information or to validate certain semantic information. In one embodiment, a particular traffic volume may be time bounded by persons in an area of the area, such that the area may be avoided, for example by a device, when increased traffic is to be expected. For example, in a commercial building, this may be the case at the beginning of a working day.

A navigation device for an automated mobile device in a predetermined area includes an interface to a motion controller of the device; a detection device for a wireless network device in the area of the device; and a processing device configured to determine a type of device to determine semantic information about the region based on the determined type and to associate it with the region. The semantic information may later be used to position a device, such as the same device, based on the particular semantic information. The processing device may in particular comprise a programmable microcomputer or microcontroller. In particular, the processing device is preferably configured to carry out at least parts of the method described above.

The features disclosed with respect to the method can be applied analogously to the navigation device. In particular, embodiments of the navigation device include means and means for carrying out the disclosed steps.

Brief description of the figures

Embodiments of the invention will now be described in more detail with reference to the accompanying figures, in which:

1 a system with an automatically moving device;

2 a flowchart of a method for navigating a device; and

3 an exemplary map of a building with the moving device of 1
represents.

Detailed description of an embodiment

1 shows a system 100 that a device 105 , a device 110 and optionally a central instance 115 includes. The device 105 and the device 110 are in a predetermined area 120 which may include, for example, a building, a remodeled or open facility or a parking lot. Usually the device is located 105 and the device 110 in relative proximity to each other, for example in a common room 125 of the building or in a common area 130 which has a predetermined extent in the range of, for example, about 1-10 meters around the device 110 can have. The central instance 115 can be anywhere on the system 100 be provided and is usually outside the area 120 ,

The device 105 is preferably set up within the area 120 to be automatically mobile. For example, the device may 105 a mobile floor cleaning robot or a flying device (a drone) include. Particularly preferably, the device comprises 105 a home appliance, which is usually set up to a predetermined purpose within the area 120 for example, to monitor or control, transport an object, clean an object or similar purpose. This includes the device 105 a motion control, not shown, which is a position of the device s 105 in the area of 120 determine and / or a movement of the device 105 can effect by means of one or more actuators. For the navigation of the device 105 For example, a map is typically used that provides a topographical representation of relevant traffic routes, objects or obstacles within the area 120 may include. This card can be scanned by the device 105 be created or improved and includes, above all, spatial information. Usually, the card is stored in digital format.

The positioning or operation of another device 105 that with the above device 105 It can often be improved when semantic information about places or objects within the area can be identical 120 are provided. The semantic information may be, for example, a useable or avoidable device or device 110 , a purpose of an area 130 of the area 120 or point out a landmark. In particular, the semantic information may be the essence of an object or area 130 to characterize improved. For example, doors, windows, elevators, stairs, furniture or other equipment or facilities in the area can be improved in an improved way 120 be determined.

It is proposed to board the first device 105 a navigation device 135 which is set up to provide semantic information about an area 130 or an object in the area 120 and in particular store in a semantic card. The navigation device detects this 135 one or more wirelessly networked devices 110 in the area 130 of the device 105 , determines their types, provides semantic information based on the specific types, and maps the semantic information to the scope 130 to. The area 130 is stationary when the device 110 is fixed, otherwise the area 130 also to the device 110 be coupled. The navigation device 135 comprises a processing device 140 , an interface 145 for connection to a motion control of the device 105 and a first wireless communication device 150 used to capture the wireless networked device 110 is set up. Preferably, by means of the first communication device 150 a communication with the device 110 be performed. For this purpose, a known protocol such as CoAP (Constrained Application Protocol), XMPP (Extensible Messaging and Presence Protocol) or MQTT (Message Queue Telemetry Transport) can be used.

In one embodiment, there is further a second wireless communication device 155 for communication with the central instance 115 intended. In addition, a data memory is preferred 160 intended. In a further embodiment, a sensor 165 to scan an environment of the device 105 be provided. The processing device 140 is preferably set up by means of the first communication device 150 to determine that a device 110 in the area 130 located. In a similar way, the absence of the device 110 in the area 130 be determined. Further preferred is then a wireless communication with the device 110 established and at least one identification of the device 110 queried. On the basis of identification or other information, one type of device may be 110 be determined. For example, the type can be another device 105 , a sensor or a fixed device in the area 120 affect. In one embodiment, the type is determined based on the identification by means of a table, an algorithm or a database. For this purpose, appropriate information in the data memory 160 be provided. Alternatively, by means of the second wireless communication device 155 a request to the central authority 115 be put to the type of the device 110 based on the information available. In a preferred embodiment, in the context of communication with the device 110 queried further information, in particular configuration or user data of the device 110 , The payload may be for the purpose of the device 110 correspond and the configuration data, for example, an indication of other devices 110 or devices 105 in the area of 120 , a position of the device 110 or other information. The processing device 140 then determines based on the type of device 110 and possibly further specific information semantic information pertaining to the field 130 Respectively. For example, the semantic information may be a preferred usage of the room 125 affect. For example, it is possible to differentiate between the device itself 105 located in a kitchen, a bathroom or a bedroom. This semantic information can be improved for navigation of the device 105 be used.

2 shows a flowchart of a method 200 to navigate a device 105 within an area 120 , The procedure 200 is in particular for implementation by means of the navigation device 135 of the device 105 from 1 set up.

In an optional step 205 becomes a type of area 120 in which the device is 105 located, determined. Building types may include, for example, an apartment, a dwelling house, an office, an industrial plant, a parking garage or a hall. It is preferred that in a further step 210 a topographical map of the area 120 is determined. The card can in particular by the device 105 even created by itself within the area 120 moves and scans its surroundings, for example by means of the sensor 165 performs. This step can also be done integrated with the following steps.

In one step 215 drives the device 105 a predetermined range 130 at. In one embodiment, the area becomes 130 initially determined and then approached, in another embodiment, the position of the moving device 105 currently as area 130 certainly. In a following step 220 become one or more wirelessly networked devices 110 in the area 130 of the device 105 detected. Preferably also takes place in one step 225 a communication with the detected devices 110 instead of. For detection or communication, the first wireless communication device is preferred 150 used. The detection or communication can for example be based on WLAN, Bluetooth, GPRS, NFC, RFID, ZigBee or other wireless communication technology. In the context of comprehension 220 or communication 225 At least one type of device is preferred 110 certainly. The type may be for a purpose of the device 110 , one with the contraption 110 linked device or device that interfaced with the device 110 connected is. The device 110 In particular, it may be part of a network known as the Internet of Things (IoT). In this case, the device 110 exploit a wide range of complexity and functional diversity. For example, the device 110 on a door, a lift, a device 105 , indicate an escape route or a passage. Based on the type of device 110 For example, it can be determined in which kind of room 125 the device 105 located. Includes the device 110 For example, a washing machine, so the room 125 For example, include a kitchen or a bathroom, a bedroom can be excluded. Includes the device 110 a sensor 165 for determining the opening degree of a window, as it is used for the control of an extractor hood, so includes the room 125 with high probability a kitchen. The specific semantic information about the space 125 , the area 130 or the area 120 can in the step 210 certain topographic map are stored.

In an optional step 235 The semantic information can be validated or made plausible before going to the field 130 be assigned, in which the device 105 located. Was for example by means of the sensor 165 an object recognized by the device 110 is assigned, such as a door or a staircase, the semantic information can also directly the object instead of the area 130 be assigned to the object.

In one embodiment, verification of the semantic information is performed by comparing the acquired information or the particular semantic information with each other. If the information changes, for example, in the daily or weekly rhythm, the specific semantic information can be extended by the rhythm. This can result in an improved validation or falsification of the semantic information. If necessary, a predetermined semantic information of the area 130 taken back or deleted from the map. In a further embodiment, the validation or plausibility check by means of the sensor 165 be performed. The sensor 165 In particular, it can be used to scan a moving object such as a human or an animal or another mobile device 105 be used. The frequenting of an area 130 , in particular by humans, can also be regarded as independent semantic information and stored.

In one step 240 who is also independent of the other steps of the procedure 200 can be performed, a positioning of the device 105 on the basis of the area 130 associated topographic information. Positioning may be either determining or approaching a position by the device 105 include. The position of the device 105 may be semantic, for example "at a staircase", "first floor" or "in the bathroom", or absolutely, in particular with respect to a card system such as the WGS84. It is advantageous if an absolute position of a device 110 when determining the semantic information as well.

The procedure 200 can be carried out continuously in the manner of a program loop or in the manner of an endless loop.

3 shows a map 300 an exemplary building 120 This is an area 120 represents, with a first room 125.1 a second room 125.2 and a third room 125.3 , Inside the building 120 are several exemplary devices 110 distributed, the reference numbers from here 305 wear. The following is a purely exemplary determination of semantic information regarding areas 130 of the building 120 through a device 105 described.

The device 105 is exemplary in the first room 125.1 , Here are a smoke detector 305 , a cooker 310 and a window sensor 315 intended. It is also located at a transition to the third room 125.3 a door 320 , On the basis of the devices 305 . 310 and possibly 315 may be the first room 125.1 to be identified as a kitchen. Also, a position of the device 105 near the smoke detector 305 as approximately in the middle of the first room 125.1 be determined. A window or a door may be based on the position of the sensor 315 respectively. 320 be determined. On the basis of the stove 310 or the smoke detector 305 can be concluded that in the first room 125.1 flammable objects can be located.

The second room 125.2 includes a washing machine 325 , a humidity sensor 330 , which may be part of a ventilation control in particular, and no smoke detector 305 , On the basis of these devices 325 . 330 can be concluded that the second room 125.2 includes a bathroom or a toilet. At the transition to the third room 125.3 is another door 335 intended. Based on the shape of the third room 125.3 as well as the number and position of the doors 320 and 335 can the third room 125.3 be identified as a gait. By scanning a trajectory of a human in the building 120 This assessment can be supported.

The explanations made with reference to the figures are purely illustrative and not restrictive.

LIST OF REFERENCE NUMBERS

100

 system

105

 device

110

 contraption

115

 central instance

120

 area

125

 room

130

 Area

135

 navigation device

140

 processing device

145

 interface

150

 first wireless communication device

155

 second wireless communication device

160

 data storage

165

 sensor

200

 method

205

 Determine the building type

210

 determine topographical map

215

 Approaching predetermined area

220

 Capture devices

225

 Communication with devices

230

 Determine semantic information

235

 Validate / validate / map semantic information

240

 positioning

300

 map

305

 smoke detector

310

 stove

315

 window sensor

320

 door

325

 Washing machine

330

 humidity sensor

335

 door

Claims (10)

Procedure ( 200 ) to create a semantic map ( 300 ) of a predetermined area ( 120 ) by means of an automated mobile device ( 105 ), the process ( 200 ) includes the following steps: - Move ( 215 ) of the device ( 105 ) in the area of ( 120 ); - To capture ( 220 ) a wirelessly networked device ( 110 ) in the area ( 130 ) of the device ( 105 ); - Determine ( 225 ) of a type of device ( 110 ); - Determine ( 230 ) semantic information about the area ( 130 ) based on the specific type; and assigning the semantic information to the area ( 130 ); Procedure ( 200 ) according to claim 1, with the further step determining or starting ( 240 ) of a position. Procedure ( 200 ) according to claim 1 or 2, wherein the area is a building ( 120 ), the area ( 130 ) in a room ( 125 ) of the building ( 120 ) and, based on the particular type, a type of use of the space ( 125 ) is determined. Procedure ( 200 ) according to claim 3, wherein the types of multiple wirelessly networked devices ( 110 ) in the room ( 125 ) and, based on the types determined, a type of use of the space ( 125 ) is determined. Procedure ( 200 ) according to one of the preceding claims, wherein a communication with the networked device ( 110 ) carried out ( 225 ) and determining the type of device ( 110 ) based on identification data of the networked device ( 110 ) he follows. Procedure ( 200 ) according to claim 5, wherein said determining ( 225 ) of the type querying a central instance ( 115 ). Procedure ( 200 ) according to one of the preceding claims, wherein a communication with the networked device ( 110 ) carried out ( 225 ) and the semantic information on the basis of user data of the networked device ( 110 ) certainly ( 230 ) become. Procedure ( 200 ) according to one of the preceding claims, wherein a movable object in the region ( 130 ) of the device ( 105 ) and validates the semantic information on the basis of the mobile object ( 235 ) become. Procedure ( 200 ) according to one of the preceding claims, wherein the semantic information comprises a determination time. Navigation device ( 135 ) for an automated mobile device ( 105 ) in an area ( 120 ), the navigation device ( 135 ) comprises: an interface ( 145 ) to a motion control of the device ( 105 ); A detection device ( 150 ) for a wireless networked device ( 110 ) in the area ( 130 ) of the device ( 105 ); and a processing device ( 140 ), which is adapted to a type of device ( 110 ) to determine semantic information about the area ( 130 ) based on the particular type determine and the semantic information the area ( 130 ).

Images