WO2019034115A1 - 结合标识的同时定位和地图创建导航方法、装置及系统 - Google Patents

结合标识的同时定位和地图创建导航方法、装置及系统 Download PDF

Info

Publication number
WO2019034115A1
WO2019034115A1 PCT/CN2018/100869 CN2018100869W WO2019034115A1 WO 2019034115 A1 WO2019034115 A1 WO 2019034115A1 CN 2018100869 W CN2018100869 W CN 2018100869W WO 2019034115 A1 WO2019034115 A1 WO 2019034115A1
Authority
WO
WIPO (PCT)
Prior art keywords
identifier
positioning object
positioning
pose
information
Prior art date
Application number
PCT/CN2018/100869
Other languages
English (en)
French (fr)
Inventor
俞毓锋
李洪波
刘凯
王政
Original Assignee
北京极智嘉科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京极智嘉科技有限公司 filed Critical 北京极智嘉科技有限公司
Priority to JP2018558212A priority Critical patent/JP6781771B2/ja
Priority to EP18832973.4A priority patent/EP3598072A4/en
Priority to US16/315,574 priority patent/US11340628B2/en
Priority to SG11202001328PA priority patent/SG11202001328PA/en
Priority to KR1020207007437A priority patent/KR102367438B1/ko
Publication of WO2019034115A1 publication Critical patent/WO2019034115A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0234Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using optical markers or beacons
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/28Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
    • G01C21/30Map- or contour-matching
    • G01C21/32Structuring or formatting of map data
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/28Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
    • G01C21/30Map- or contour-matching
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34Route searching; Route guidance
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34Route searching; Route guidance
    • G01C21/36Input/output arrangements for on-board computers
    • G01C21/3626Details of the output of route guidance instructions
    • G01C21/3644Landmark guidance, e.g. using POIs or conspicuous other objects
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/38Electronic maps specially adapted for navigation; Updating thereof
    • G01C21/3804Creation or updating of map data
    • G01C21/3807Creation or updating of map data characterised by the type of data
    • G01C21/3811Point data, e.g. Point of Interest [POI]
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/38Electronic maps specially adapted for navigation; Updating thereof
    • G01C21/3804Creation or updating of map data
    • G01C21/3859Differential updating map data
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S17/00Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
    • G01S17/02Systems using the reflection of electromagnetic waves other than radio waves
    • G01S17/06Systems determining position data of a target
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0212Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
    • G05D1/0225Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving docking at a fixed facility, e.g. base station or loading bay

Definitions

  • the present application relates to the field of automatic navigation, for example, to a simultaneous positioning and map creation navigation method, apparatus and system incorporating identification.
  • SLAM Vision Simultaneous Localization and Mapping
  • the SLAM system continuously locates and updates the map during the walking of the positioning object based on the initialized pose, and generates a map based on the initial pose.
  • the SLAM system searches and matches the map in the initial pose region, obtains the initial pose information, and continuously locates and updates the map while the positioning object is walking.
  • SLAM navigation does not require the placement of markers in the scene, and its convenience has become an important branch of the indoor navigation scheme.
  • the accuracy of SLAM navigation mainly depends on the closing precision. When the scene becomes larger, the map closure cannot be effectively performed, which will greatly affect the positioning and map effects. In addition, the accuracy of the SLAM system is affected by the accuracy of the sensor. The low-cost SLAM system cannot meet the precise positioning required for the conveyor belt docking and robot arm operation.
  • the present application provides a novel and improved simultaneous positioning and map creation navigation method, apparatus and system.
  • the present application provides a simultaneous positioning and map creation navigation method, including: setting a positioning object initialization area; adding a first identifier in the positioning object initialization area, the first identifier includes pose information or identification information; And causing the positioning object to be activated from the positioning object initialization area, and determining an initial posture of the positioning object based on the first identifier; adding a second identifier on the travel path of the positioning object; and, when the positioning object passes the second identifier And updating the current pose of the positioning object based on the second identifier.
  • the positioning object is one of a robot that travels in a simultaneous positioning and map creation navigation and a mobile device that carries a conveyor belt, a shelf, a cage, and a robotic arm.
  • the first identifier includes pose information
  • determining an initial pose of the positioning object based on the first identifier includes: comparing the first identifier with respect to the positioning object
  • the pose of the first pose is directly extracted from the pose information of the first identifier via coordinate transformation.
  • the first identifier includes the identification information
  • determining the initial pose of the positioning object based on the first identifier includes: matching, by using the identifier information of the first identifier, Recording a plurality of pose information and an error covariance matrix of the positioning object, and converting the plurality of pose information and the error covariance matrix of the first identifier; and integrating the pose information of the plurality of first identifiers by the Gaussian model to estimate An optimal pose information and an error covariance matrix of the first identifier; and based on the pose of the first identifier relative to the positioning object, directly extracting the optimal pose information from the first identifier via coordinate transformation Position the initial pose information of the object.
  • the second identifier includes pose information, or contains identification information, or contains non-identification type graphic information.
  • the second identifier includes pose information
  • updating the current pose of the positioning object based on the second identifier includes: based on the second identifier relative to the positioning object
  • the pose of the second pose directly extracts the current pose information of the positioning object from the pose information of the second identifier.
  • the second identifier includes the identifier information
  • updating the current pose of the positioning object based on the second identifier includes: matching, by using the identifier information of the second identifier, Recording a plurality of pose information and an error covariance matrix of the positioning object, and converting into a plurality of pose information and an error covariance matrix of the second identifier; fusing the plurality of pose information by a Gaussian model to estimate the second Identifying the optimal pose information and the error covariance matrix; and based on the pose of the second identifier relative to the positioning object, directly extracting the current position of the positioning object from the optimal pose information of the second identifier via coordinate transformation Pose information.
  • the second identifier includes non-identification type graphic information
  • updating the current pose of the positioning object based on the second identifier includes: using the same identifier when the positioning object passes The similarity of the postures is determined whether the identifier passed by the positioning object is the second identifier; the plurality of pose information and the error covariance matrix of the positioning object that is subjected to the second identifier are recorded, and converted into the plurality of pose information of the second identifier And an error covariance matrix; the plurality of pose information of the second identifier is fused by the Gaussian model to estimate the optimal pose information and the error covariance matrix of the second identifier; and the bit based on the second identifier relative to the positioning object Position, directly extracting current pose information of the positioning object from the optimal pose information of the second identifier via coordinate conversion
  • adding a second identifier to the travel path of the positioning object includes: Adding a plurality of second identifiers on the travel path; and, wherein, if the positioning object passes the second identifier, updating a current pose of the positioning object based on the second identifier, further comprising: positioning When the object passes through the plurality of second identifiers, the current pose of the positioning object is updated based on the plurality of second identifiers.
  • the auxiliary positioning belt connecting the second identifier is included between two adjacent second identifiers of the plurality of second identifiers, so that when the positioning object passes
  • the auxiliary positioning belt corrects an angular deviation of the positioning object relative to the auxiliary positioning belt such that the positioning object travels along the auxiliary positioning belt, ensuring that the positioning object can pass and identify the second identification.
  • the method further includes: adding a third identifier to the docking device on the travel path of the positioning object; and in the case that the positioning object passes the third identifier, Performing a docking operation with the docking device based on the third identifier.
  • the present application provides a simultaneous positioning and map creation navigation device combined with an identifier, comprising: an initialization area setting unit configured to set a positioning object initialization area; a first identification adding unit configured to add a first in the positioning object initialization area
  • the first identifier includes posture information or identification information;
  • the positioning object activation unit is configured to enable the positioning object to be activated from the positioning object initialization area, and determine an initial posture of the positioning object based on the first identifier;
  • a second identifier adding unit configured to add a second identifier on the travel path of the positioning object; and a pose updating unit configured to update the current bit of the positioning object based on the second identifier if the positioning object passes the second identifier posture.
  • the positioning object includes one of a robot traveling in the navigation of the simultaneous positioning and map creation and a mobile device carrying the conveyor belt or the robot arm.
  • the first identifier includes pose information
  • the positioning object activation unit is further configured to: based on the pose of the first identifier relative to the positioning object, via The coordinate conversion directly extracts initial pose information of the positioning object from the pose information of the first identifier.
  • the first identifier includes the identification information
  • the positioning object activation unit is further configured to: use the identification information of the first identifier to match, and record the plurality of positioning objects. a pose information and an error covariance matrix, and converted into a plurality of pose information and an error covariance matrix of the first identifier; and merging the plurality of pose information of the first identifier by a Gaussian model to estimate the first identifier Optimal pose information and error covariance matrix; based on the pose of the first identifier relative to the positioning object, directly extracting the initial position of the positioning object from the optimal pose information of the first identifier via coordinate transformation Position information.
  • the second identifier includes pose information, identification information, or non-identification type graphic information.
  • the second identifier includes pose information
  • the pose update unit is further configured to: based on the pose of the second identifier relative to the positioning object, via The coordinate conversion directly extracts the current pose information of the positioning object from the pose information of the second identifier.
  • the second identifier includes the identification information
  • the pose updating unit is further configured to: use the identification information of the second identifier to match, and record the plurality of positioning objects. a pose information and an error covariance matrix, and converted into a plurality of pose information and an error covariance matrix of the second identifier; fusing the plurality of pose information by a Gaussian model to estimate an optimal bit of the second identifier The pose information and the error covariance matrix; based on the pose of the second identifier relative to the positioning object, the current pose information of the positioning object is directly extracted from the optimal pose information of the second identifier via coordinate transformation.
  • the second identifier includes non-identification type graphic information
  • the pose updating unit is configured to: determine the similarity of the poses when the positioning object passes the same identification. Whether the identifier passed by the positioning object is a second identifier; recording a plurality of pose information and an error covariance matrix of the positioning object that is subjected to the second identifier, and converting the plurality of pose information and the error covariance matrix of the second identifier; Integrating a plurality of pose information of the second identifier by a Gaussian model to estimate an optimal pose information and an error covariance matrix of the second identifier; and based on the pose of the second identifier relative to the positioning object, The optimal pose information of the second identifier directly extracts current pose information of the positioning object.
  • the method includes: the first identifier adding unit is further configured to add on a travel path of the positioning object a plurality of second identifiers; and the pose updating unit is configured to update a current pose of the positioning object based on the plurality of second identifiers when the positioning object passes the plurality of second identifiers.
  • the method further includes: an auxiliary positioning unit, configured to add a connection between the two adjacent second identifiers between the two adjacent second identifiers of the plurality of second identifiers Auxiliary positioning belt and correcting an angular deviation of the positioning object relative to the auxiliary positioning belt when the positioning object passes the auxiliary positioning belt, so that the positioning object travels along the auxiliary positioning belt.
  • an auxiliary positioning unit configured to add a connection between the two adjacent second identifiers between the two adjacent second identifiers of the plurality of second identifiers Auxiliary positioning belt and correcting an angular deviation of the positioning object relative to the auxiliary positioning belt when the positioning object passes the auxiliary positioning belt, so that the positioning object travels along the auxiliary positioning belt.
  • the method further includes: a third identifier adding unit, configured to add a third identifier on the docking device on the travel path of the positioning object; and the docking unit is set to be in the In the case that the positioning object passes the third identifier, the docking operation is completed with the docking device based on the third identifier.
  • the present application provides a simultaneous navigation and map creation navigation system, including: a positioning object, which is used for real-time positioning and constructing a map in a scene; a first identifier, including pose information or identification information; a second identifier;
  • the positioning and map creating device comprises: an initializing area setting unit, configured to set a positioning object initializing area; a first identifier adding unit, configured to add the first identifier to the positioning object initializing area; positioning an object starting unit, setting In order to enable the positioning object to be activated from the positioning object initialization area, and determine an initial pose of the positioning object based on the first identifier; the second identifier adding unit is configured to add the second identifier to the travel path of the positioning object And a pose updating unit configured to update the current pose of the positioning object based on the second identifier if the positioning object passes the second identifier.
  • the positioning object is one of a robot that travels in the simultaneous positioning and map creation navigation and a mobile device that carries a conveyor belt or a robotic arm.
  • the first identifier includes pose information
  • the positioning object activation unit is configured to: based on the pose of the first identifier relative to the positioning object, via coordinates Converting the initial pose information of the positioning object directly from the pose information of the first identifier.
  • the first identifier includes identification information
  • the positioning object activation unit is configured to: use the identification information of the first identifier to match, and record multiple locations of the positioning object.
  • the pose information and the error covariance matrix are converted into a plurality of pose information and an error covariance matrix of the first identifier; the plurality of pose information of the first identifier are fused by the Gaussian model to estimate the first identifier
  • An optimal pose information and an error covariance matrix based on the pose of the first identifier relative to the positioning object, directly extracting an initial pose of the positioning object from the optimal pose information of the first identifier via coordinate transformation information.
  • the second identifier includes pose information identification information or non-identification type graphic information.
  • the second identifier includes pose information
  • the pose update unit is further configured to: based on the pose of the second identifier relative to the positioning object, via The coordinate conversion directly extracts the current pose information of the positioning object from the pose information of the second identifier.
  • the second identifier includes the identification information
  • the pose updating unit is further configured to: use the identification information of the second identifier to match, and record the plurality of positioning objects. a pose information and an error covariance matrix, and converted into a plurality of pose information and an error covariance matrix of the second identifier; fusing the plurality of pose information by a Gaussian model to estimate an optimal bit of the second identifier The pose information and the error covariance matrix; based on the pose of the second identifier relative to the positioning object, the current pose information of the positioning object is directly extracted from the optimal pose information of the second identifier via coordinate transformation.
  • the second identifier includes non-identification type graphic information
  • the pose updating unit is further configured to: use the similarity of the pose when the positioning object passes the same logo, Determining whether the identifier passed by the positioning object is a second identifier; recording a plurality of pose information and an error covariance matrix of the positioning object that is subjected to the second identifier, and converting the plurality of pose information and the error covariance matrix of the second identifier Encoding a plurality of pose information of the second identifier by a Gaussian model to estimate an optimal pose information and an error covariance matrix of the second identifier; and converting the pose based on the pose of the second identifier relative to the positioning object Extracting current pose information of the positioning object directly from the optimal pose information of the second identifier.
  • the method includes: the first identifier adding unit, further configured to add the plurality of second identifiers Going to the travel path of the positioning object; and the pose update unit is configured to update the current pose of the positioning object based on the plurality of second identifiers if the positioning object passes the plurality of second identifiers.
  • the auxiliary positioning belt is included between two adjacent second identifiers of the plurality of second identifiers
  • the simultaneous positioning and map creating device further includes: an auxiliary positioning unit And arranging to correct an angular deviation of the positioning object relative to the auxiliary positioning belt when the positioning object passes the auxiliary positioning belt, so that the positioning object travels along the auxiliary positioning belt.
  • the method further includes: a third identifier adding unit, configured to add a third identifier on the docking device on the travel path of the positioning object; and the docking unit is set to be in the In the case that the positioning object passes the third identifier, the docking operation is completed with the docking device based on the third identifier.
  • An embodiment of the present application provides a method for simultaneously positioning and mapping a navigation object, including: setting a positioning object initialization area; wherein the positioning object initializing area, the positioning path of the positioning object, and the positioning object are docked At least one of the devices is provided with an identifier, the identifier including at least one of pose information, identification information, and non-identification type graphic information; controlling the positioning object to perform at least one of: from the positioning object Initializing an area to start, and determining an initial pose of the positioning object based on the identifier; if the identifier is passed on the travel path, updating a current pose of the positioning object based on the identifier; when docking with the docking device And adjusting a relative posture of the positioning object and the docking device based on the identifier.
  • the positioning object includes a robot that travels in a simultaneous positioning and map creation navigation and a mobile device that needs to interface with the docking device.
  • the docking device comprises a conveyor belt or a robotic arm.
  • the identifier of the positioning object initialization area setting is a first identifier
  • the first identifier includes pose information
  • determining an initial pose of the positioning object based on the first identifier includes: The first identifier is directly extracted from the pose information of the first identifier with respect to the pose of the positioning object, and the initial pose information of the positioning object is directly extracted.
  • the identifier of the positioning object initialization area is a first identifier
  • the first identifier includes identifier information
  • determining an initial pose of the positioning object based on the first identifier includes:
  • the identifier is activated, the identifier information included in the first identifier is recorded, and the identifier information is used as a coordinate origin to perform positioning and a map is created;
  • the identification information of the first identifier is matched to obtain the pose information of the positioning object, and the plurality of pose information and the error of the positioning object are recorded, and converted into a plurality of pose information and an error of the first identifier;
  • the Gaussian model fuses the plurality of pose information of the first identifier to estimate an optimal pose information and an error of the first identifier; and based on the pose of the first identifier relative to the positioning object, The optimal pose information of the first identifier directly determines the initial pose information of the positioning object.
  • the identifier set on the travel path of the positioning object is a second identifier
  • the second identifier includes pose information
  • updating the current pose of the positioning object based on the second identifier includes: And describing the pose of the second identifier relative to the positioning object, and directly extracting current pose information of the positioning object from the pose information of the second identifier via coordinate conversion.
  • the identifier set on the travel path of the positioning object is a second identifier
  • the second identifier includes the identifier information
  • updating the current pose of the positioning object based on the second identifier includes:
  • the positioning information of the second identifier is used to obtain the pose information of the positioning object, and the plurality of pose information and the error of the positioning object are recorded, and Converting into a plurality of pose information and an error of the second identifier; fusing a plurality of pose information of the second identifier by a Gaussian model to estimate an optimal pose information and an error of the second identifier; Determining, according to the pose of the second identifier, the current pose information of the positioning object is directly determined from the optimal pose information of the second identifier via coordinate transformation.
  • the identifier set on the travel path of the positioning object is a second identifier
  • the second identifier includes non-identification type graphic information
  • updating a current pose of the positioning object based on the second identifier includes And determining, by using the similarity of the poses when the positioning object passes through the same identifier, determining whether the identifier that the positioning object repeatedly passes is a second identifier; if the second identifier is, recording the second identifier Locating a plurality of pose information and an error of the object, and converting the plurality of pose information and the error of the second identifier; and integrating the plurality of pose information of the second identifier by a Gaussian model to estimate the first The optimal pose information and error of the second identifier; and determining the current bit of the positioning object directly from the optimal pose information of the second identifier via coordinate transformation based on the pose of the second identifier relative to the positioning object Position information.
  • the identifier set on the travel path of the positioning object is a second identifier, and a plurality of second identifiers are disposed.
  • an auxiliary positioning belt connecting the two adjacent second identifiers is included between two adjacent second identifiers of the plurality of second identifiers, so that the positioning object passes the auxiliary
  • the angular deviation of the positioning object relative to the auxiliary positioning belt is corrected when the belt is positioned such that the positioning object travels along the auxiliary positioning belt.
  • the embodiment of the present application provides a simultaneous positioning and map creation navigation device combined with an identifier, including: an initialization area setting unit, a positioning object activation unit, a pose updating unit, and a docking unit.
  • the initialization area setting unit is configured to set a positioning object initialization area, wherein at least one of the positioning object initialization area, the traveling path of the positioning object, and the docking device of the positioning object is provided with an identifier,
  • the identifier includes at least one of pose information, identification information, and non-identification type graphic information; and the positioning object activation unit is configured to control the positioning object to perform at least one of: starting from the positioning object initialization area, And determining, according to the identifier, an initial pose of the positioning object;
  • the pose updating unit is configured to update the current pose of the positioning object based on the identifier if the identifier is on the travel path; and the docking unit is set to be docked When the device is docked, the relative position of the positioning object and the docking device
  • the positioning object comprises a robot traveling in a navigation that is simultaneously positioned and map created and a mobile device that needs to interface with the docking device.
  • the docking device comprises a carrying belt or a robotic arm.
  • the identifier of the positioning object initialization area is a first identifier
  • the first identifier includes pose information
  • the positioning object activation unit is further configured to: based on the first identifier, relative to Positioning the pose of the object, and directly extracting the initial pose information of the positioning object from the pose information of the first identifier via coordinate conversion.
  • the identifier of the positioning object initialization area is a first identifier
  • the first identifier includes identifier information
  • the positioning object activation unit is further configured to: when the first identifier is started Recording the identification information included in the first identifier, using the identifier information as a coordinate origin to locate and create a map; and when the positioning object repeatedly passes the first identifier repeatedly, using the identifier of the first identifier Information is matched to obtain pose information of the positioning object, and a plurality of pose information and an error of the positioning object are recorded, and converted into a plurality of pose information and an error of the first identifier; a plurality of pose information of the first identifier to estimate an optimal pose information and an error of the first identifier; and based on a pose of the first identifier relative to the positioning object, from the first identifier via coordinate conversion
  • the optimal pose information directly determines the initial pose information of the positioning object.
  • the identifier set on the travel path of the positioning object is a second identifier
  • the second identifier includes pose information
  • the pose update unit is further configured to: relative to the second identifier And positioning the current pose information of the positioning object directly from the pose information of the second identifier via coordinate transformation.
  • the identifier set on the travel path of the positioning object is a second identifier
  • the second identifier includes identifier information
  • the pose update unit is further configured to: repeat When the second identifier is passed, the posture information of the positioning object is obtained by matching the identification information of the second identifier, and a plurality of pose information and an error of the positioning object are recorded, and converted into the first a plurality of pose information and an error of the second identifier; a plurality of pose information of the second identifier is fused by a Gaussian model to estimate an optimal pose information and an error of the second identifier; and based on the second identifier
  • the current pose information of the positioning object is directly determined from the optimal pose information of the second identifier via a coordinate transformation.
  • the identifier set on the travel path of the positioning object is a second identifier
  • the second identifier includes non-identification type graphic information
  • the pose update unit is configured to: utilize the positioning object And determining, by the similarity of the poses of the same identifier, whether the identifier that the positioning object repeatedly passes is a second identifier; if the second identifier is, recording the multiple digits of the positioning object that passes the second identifier Position information and error, and converted into a plurality of pose information and an error of the second identifier; and merging a plurality of pose information of the second identifier by a Gaussian model to estimate an optimal pose information of the second identifier And an error; and based on the pose of the second identifier relative to the positioning object, the current pose information of the positioning object is directly determined from the optimal pose information of the second identifier via coordinate transformation.
  • the identifier set on the travel path of the positioning object is a second identifier, and a plurality of second identifiers are disposed.
  • the method further includes: an auxiliary positioning unit, configured to add an auxiliary positioning belt connecting the two adjacent second identifiers between two adjacent second identifiers of the plurality of second identifiers, and The positioning object corrects an angular deviation of the positioning object relative to the auxiliary positioning belt when the auxiliary positioning belt passes, so that the positioning object travels along the auxiliary positioning belt.
  • an auxiliary positioning unit configured to add an auxiliary positioning belt connecting the two adjacent second identifiers between two adjacent second identifiers of the plurality of second identifiers, and The positioning object corrects an angular deviation of the positioning object relative to the auxiliary positioning belt when the auxiliary positioning belt passes, so that the positioning object travels along the auxiliary positioning belt.
  • the embodiment of the present application provides a simultaneous navigation and map creation navigation system, including: positioning objects, which are set to locate and construct a map in real time in the scene; the identifier includes posture information, identification information, and non-identification type graphic information.
  • At least one of the simultaneous positioning and map creation means comprising: an initialization area setting unit configured to set a positioning object initialization area; wherein the positioning object initialization area, the travel path of the positioning object, and the positioning object
  • At least one of the docking device is provided with an identifier;
  • the positioning object activation unit is configured to control the positioning object to perform at least one of: starting from the positioning object initialization region, and determining an initial of the positioning object based on the identifier a pose updating unit configured to update a current pose of the positioning object based on the identifier if the identifier is set on the travel path;
  • the docking unit is configured to adjust based on the identifier when docking with the docking device The relative position of the positioning object and the docking device.
  • the positioning object comprises a robot traveling in a navigation that is simultaneously positioned and map created and a mobile device that needs to interface with the docking device.
  • the docking device comprises a carrying belt or a robotic arm.
  • the identifier of the positioning object initialization area is a first identifier
  • the first identifier includes pose information
  • the positioning object activation unit is configured to: relative to the first identifier based on the positioning The pose of the object directly extracts the initial pose information of the positioning object from the pose information of the first identifier via coordinate transformation.
  • the identifier of the positioning object initialization area is a first identifier
  • the first identifier includes identifier information
  • the positioning object activation unit is configured to: when the first identifier is activated, Recording the identification information included in the first identifier, using the identifier information as a coordinate origin to locate and create a map; and when the positioning object repeatedly passes the first identifier repeatedly, using the identifier information of the first identifier Performing matching to obtain pose information of the positioning object, recording a plurality of pose information and an error of the positioning object, and converting into a plurality of pose information and an error of the first identifier; and fusing the first identifier by a Gaussian model a plurality of pose information to estimate an optimal pose information and an error of the first identifier; and based on a pose of the first identifier relative to the positioning object, from an optimal position of the first identifier via coordinate transformation
  • the pose information directly determines the initial pose information of the positioning object.
  • the identifier set on the travel path of the positioning object is a second identifier
  • the second identifier includes pose information
  • the pose update unit is further configured to: relative to the second identifier And positioning the current pose information of the positioning object directly from the pose information of the second identifier via coordinate transformation.
  • the identifier set on the travel path of the positioning object is a second identifier
  • the second identifier includes identifier information
  • the pose update unit is further configured to: repeat When the second identifier is passed, the posture information of the positioning object is obtained by matching the identification information of the second identifier, and a plurality of pose information and an error of the positioning object are recorded, and converted into the first Determining a plurality of pose information and an error of the second identifier; merging the plurality of pose information of the second identifier by a Gaussian model to estimate an optimal pose information and an error of the second identifier; and comparing the second identifier based on the second identifier
  • the current pose information of the positioning object is directly determined from the optimal pose information of the second identifier via a coordinate transformation.
  • the identifier set on the travel path of the positioning object is a second identifier
  • the second identifier includes non-identification type graphic information
  • the pose update unit is further configured to: utilize the positioning object And determining, by the similarity of the poses of the same identifier, whether the identifier that the positioning object repeatedly passes is a second identifier; if the second identifier is, recording the multiple digits of the positioning object that passes the second identifier Position information and error, and converted into a plurality of pose information and an error of the second identifier; and merging a plurality of pose information of the second identifier by a Gaussian model to estimate an optimal pose information of the second identifier And an error; based on the pose of the second identifier relative to the positioning object, the current pose information of the positioning object is directly determined from the optimal pose information of the second identifier via coordinate transformation.
  • the identifier set on the travel path of the positioning object is a second identifier, and a plurality of second identifiers are disposed.
  • an auxiliary positioning belt is included between two adjacent second identifiers of the plurality of second identifiers, the simultaneous positioning and map creating device further comprising: an auxiliary positioning unit, configured to be in the positioning object An angular deviation of the positioning object relative to the auxiliary positioning belt is corrected as the auxiliary positioning belt passes, such that the positioning object travels along the auxiliary positioning belt.
  • FIG. 1 is a schematic flowchart of a method for simultaneous positioning and map creation navigation in combination with an identifier according to an embodiment of the present application
  • FIG. 2 is a schematic diagram showing an example of an identification according to an embodiment of the present application.
  • FIG. 3 is a schematic diagram showing a projective transformation of an identification in accordance with an embodiment of the present application.
  • FIG. 4 is a schematic diagram of coordinate conversion of a logo image according to an embodiment of the present application.
  • FIG. 5 is a schematic diagram of pose orientation information fusion according to an embodiment of the present application.
  • 6A and 6B are schematic views of a visual identification and a laser marking according to an embodiment of the present application.
  • FIGS. 7A and 7B are schematic diagrams of coordinate systems defined for visual identification and laser identification, in accordance with an embodiment of the present application.
  • FIG. 8 is a schematic diagram of merging similar gesture poses according to an embodiment of the present application.
  • FIGS. 9A and 9B are schematic diagrams showing a manner of connection between a logo and a ribbon according to an embodiment of the present application.
  • FIG. 10 is a schematic view showing calculation of a deviation between a robot and a ribbon according to an embodiment of the present application.
  • FIG. 11 is a schematic block diagram of a simultaneous positioning and map creation navigation device incorporating an identification according to an embodiment of the present application
  • FIG. 12 is a schematic block diagram of a simultaneous positioning and map creation navigation system incorporating identifications in accordance with an embodiment of the present application.
  • the term “a” is understood to mean “at least one” or “one or more”, that is, in one embodiment, the number of one element may be one, and in other embodiments, the element The number can be multiple, and the term “a” cannot be construed as limiting the quantity.
  • ordinal numbers such as “first”, “second”, etc. will be used to describe various components, those components are not limited herein. This term is only used to distinguish one component from another. For example, a first component could be termed a second component, and as such, a second component could also be termed a first component without departing from the teachings of the application.
  • the term “and/or” used herein includes any and all combinations of one or more of the associated listed items.
  • the embodiment of the present application provides a method for simultaneously positioning and mapping a navigation in combination with an identifier, including: setting a positioning object initialization area; adding a first identifier in the positioning object initialization area, where the first identifier includes pose information or identification information; And causing the positioning object to be activated from the positioning object initialization area, and determining an initial posture of the positioning object based on the first identifier; adding a second identifier on the traveling path of the positioning object; and, in the case that the positioning object passes the second identifier, Updating the current pose of the positioning object based on the second identifier.
  • FIG. 1 is a schematic flowchart of a simultaneous positioning and map creation navigation method combined with an identifier according to an embodiment of the present application.
  • the simultaneous positioning and map creation navigation method according to the embodiment of the present application includes: Step S1 to Step S5.
  • step S1 a positioning object initialization area is set.
  • step S2 a first identifier is added to the positioning object initialization area, and the first identifier includes pose information or identification information.
  • step S3 the positioning object is caused to be activated from the positioning object initialization area, and an initial posture of the positioning object is determined based on the first identification.
  • step S4 a second identification is added on the travel path of the positioning object.
  • step S5 if the positioning object passes the second identifier, the current pose of the positioning object is updated based on the second identifier.
  • the problem that the positioning object is not ambiguously initialized and the pose shift during the traveling process in the related art is solved.
  • the high precision and robustness of the SLAM navigation method can be adapted to industrial environments that require precise positioning.
  • the initialization of the positioning object can be resolved.
  • the problem of unclear posture In this way, by performing simultaneous positioning and map creation based on explicit initialization pose information, the navigation result of the SLAM navigation system can be matched with the real environment, thereby facilitating the combination of the navigation system and the real application.
  • the positioning is in the case of a known map, the initial pose of the positioning object can be constrained to achieve a successful match with the map.
  • the positioning deviation generated by the SLAM navigation system during the traveling process can be corrected, thereby facilitating map closure and improving map creation and positioning effects.
  • the method further includes: adding a plurality of second identifiers on a travel path of the positioning object; and, in the case that the positioning object passes the plurality of second identifiers, based on the plurality of The second identifier updates the current pose of the positioning object.
  • the traveling path of the positioning object A plurality of identifiers are added, so that when the positioning object passes each identifier, the current position of the positioning object when the positioning object passes through the identifier is updated in real time, thereby reducing the cumulative offset of the positioning object when moving in a large scene, thereby avoiding This caused the SLAM navigation system to be unable to perform closed detection.
  • the positioning object in the combined SLAM navigation method according to the embodiment of the present application may be multiple objects based on the application scenario.
  • the conventional SLAM navigation method and navigation system it is a robot for real-time positioning and map construction, and when used for precise positioning of conveyor belt docking, robot arm operation, etc., it is a conveyor belt and a robot arm that need precise positioning. Wait.
  • the positioning object may be a robot traveling in the SLAM navigation, or a device or device in a scene such as a belt docking or a robot arm operation.
  • the identification may have various forms depending on the method of determining the pose of the positioning object based on the identification.
  • the identifier may be a two-dimensional code (QR Code, Data Matrix, etc.), a barcode, and other visual identifiers, as shown in FIG. 2 .
  • FIG. 2 is a schematic diagram showing an example of an identification according to an embodiment of the present application.
  • the identifiers can be divided into multiple types depending on the stored information. For example, one type of identifier contains pose information (x, y, ⁇ ), another type of identifier contains only identification (ID) information, and no pose information, and another type of identifier may contain non-identification type graphic information.
  • the corresponding application method based on the identification to determine the pose of the positioning object is also different, which will be described in detail below.
  • the information stored in the identified codeword contains at least the actual location (x, y) of the identifier in the scene.
  • the angle ⁇ information needs to be provided in the codeword.
  • the positioning object is not limited to a robot, and may be a mobile device suitable for a scene such as a belt docking or a robot arm operation as described above.
  • FIG. 3 is a schematic diagram showing a projective transformation of an identification in accordance with an embodiment of the present application.
  • the image is facing the two-dimensional code, and each pixel in the image corresponds to the actual ground distance.
  • the pixel points (u, v) in the image correspond to the coordinates (x, y) of the unique robot coordinate system. As shown in the following formula (2):
  • f is the focal length, which is the actual distance corresponding to one pixel.
  • u 0 and v 0 refer to the abscissa and ordinate of the camera focal length center in the image, which is part of the parameters within the camera.
  • H i is the transformation matrix of the image coordinate system to the robot coordinate system.
  • FIG. 4 is a schematic diagram of coordinate conversion of a logo image according to an embodiment of the present application.
  • (u 0 , v 0 ) is the coordinate of the origin of the robot coordinate system in the image.
  • the image coordinates are right u and v, and the robot coordinate system is right x up y.
  • the coordinates (u, v, ⁇ ) of the two-dimensional code in the camera coordinate system are obtained by scanning the two-dimensional code through the downward facing camera. According to the transformation matrix H i , the position of the two-dimensional code in the robot coordinate system can be obtained. Through the inverse transformation, the coordinates of the robot under the two-dimensional code can be obtained, as shown in the following formula (5):
  • the actual angle of the robot in the two-dimensional code coordinate system is - ⁇ .
  • the code word stores the actual coordinate information (x b , y b , ⁇ b ) of the two-dimensional code in the scene, and obtains the two-dimensional code in the scene by decoding the code word on the two-dimensional code
  • the actual pose in the middle Therefore, the coordinates of the robot under the scene can be expressed by the following formula (6):
  • the heading angle of the robot in the scene is the same as the heading angle "- ⁇ " of the robot at the two-dimensional code plus the heading angle ⁇ b of the two-dimensional code itself, as shown in the following formula (7):
  • the initial pose (x w , y w , ⁇ w ) of the robot can be determined by the position of the two-dimensional code in the camera and the code word of the two-dimensional code. . Starting the robot in this initial pose ensures the consistency of the overall system coordinate system.
  • the identifier includes the pose information
  • the imaging device equipped with it such as a camera photographing the logo and acquiring the pose information of the logo
  • the pose information of the positioning object and the pose posture information of the marker There is a definite spatial relationship between them. Therefore, by performing coordinate conversion based on the pose information imaged in the imaging device of the positioning object, the initial pose information of the positioning object can be directly extracted from the identified pose information.
  • the relative position of the robot and the logo can be determined by identifying the relative position of the identifier and the positioning object, such as the above-mentioned robot, by identifying the imaging in the imaging device of the positioning object, thereby The pose information of the positioning object is extracted from the identified pose information via coordinate transformation.
  • other forms of coordinate transformation may be used to directly extract the pose information of the positioned object from the identified pose information.
  • the first identifier includes pose information
  • determining an initial pose of the positioning object based on the first identifier may be: based on the pose of the first identifier relative to the positioning object,
  • the initial pose information of the positioning object is directly extracted from the pose information of the first identifier via coordinate conversion.
  • the pose of the first identifier relative to the positioning object is determined based on the pose of the image formed by the first identifier in the imaging device prepared by the positioning object.
  • the identification may include an ID without including pose information.
  • the ID included in the identifier is recorded, and in the SLAM process, the ID is used as the coordinate origin to locate and create a map.
  • the ID included in the identifier is used for matching to obtain the pose of each robot, and the pose mean ⁇ w of each robot and the covariance matrix ⁇ w are recorded, and the opposite is performed as described above.
  • the coordinate transformation process is converted to the corresponding pose's pose mean and covariance matrix.
  • the pose information of the plurality of obtained markers is merged, and the optimal pose estimation and covariance matrix of the identifier are estimated, and converted into corresponding robots by the opposite coordinate conversion process as described above.
  • the pose is used as the initial pose of the robot.
  • FIG. 5 is a schematic diagram of pose information fusion according to an embodiment of the present application.
  • five different poses and errors eg, a covariance matrix
  • the pose and error ie, the optimal pose estimation and error
  • the optimal pose estimate of the logo is used as the initial pose information of the logo, and the initial pose and error of the robot are calculated according to the relative pose of the robot and the logo.
  • the optimal estimate can be transformed into the product of multiple Gaussian distributions, namely:
  • the first identifier includes the identification information
  • determining the initial pose of the positioning object based on the first identifier may be: using the identification information of the first identifier to perform matching Recording a plurality of pose information and errors of the positioning object, and converting the plurality of pose information and the error of the first identifier; and integrating the plurality of pose information of the first identifier by the Gaussian model to estimate the maximum of the first identifier
  • the orientation information and the error based on the pose of the first identifier relative to the positioning object, the initial pose information of the positioning object is directly determined from the optimal pose information of the first identifier via coordinate transformation.
  • the initial pose of the positioning object is determined based on the first identifier, thereby solving the positioning object. Initialize the problem of ambiguity.
  • one or more identifiers are added on the traveling path of the positioning object to perform positioning deviation correction.
  • the initial time identification with respect to the positioning object needs to include pose information or identification information, and the identifier added on the travel path of the positioning object may also include non-identification type graphic information.
  • the second identifier includes pose information, or contains identification information, or contains non-identification type graphic information.
  • the third identifier includes pose information, or includes identification information, or includes non-identification type graphic information.
  • the codeword information may be extracted from the pose information of the second identifier by using the method as described above, and based on the second identifier relative to the positioning object The pose, thereby obtaining the pose information of the positioned object, to directly correct the positioning result.
  • the estimated pose and error of the positioning object eg, a covariance matrix
  • the pose and error of the second identifier For example, covariance matrix
  • a second identifier containing pose information x b , y b , ⁇ b
  • the current pose of the positioning object can be directly updated. This method has also been described above, and therefore will not be described again.
  • the second identifier includes pose information
  • updating the current pose of the positioning object based on the second identifier may be: based on the pose of the second identifier relative to the positioning object And determining the current pose information of the positioning object directly from the pose information of the second identifier via coordinate conversion.
  • the second identifier includes the identifier information
  • updating the current pose of the positioning object based on the second identifier may be: using the identifier information of the second identifier to perform matching, obtaining and recording Positioning a plurality of pose information and an error of the object, and converting into a plurality of pose information and an error of the second identifier; fusing the plurality of pose information by a Gaussian model to estimate an optimal pose information and an error of the second identifier And determining, according to the pose of the second identifier relative to the positioning object, the current pose information of the positioning object from the optimal pose information of the second identifier via coordinate conversion.
  • the second identifier may also be an identifier including non-identification type graphic information.
  • the non-identifying type graphic information may not contain any meaningful information.
  • the non-identification type graphic information identifier may be a regular graphic class identifier, and a plurality of regular graphic class identifiers as described above may be added in the positioning object initialization area.
  • the second identification may be a visual identification (regular graphic) or a laser identification (rule block).
  • 6A and 6B are schematic diagrams of a visual identification and a laser identification in accordance with an embodiment of the present application. As shown in FIG. 6A, the visual identification is a detectable pattern of the orientation of the belt, and as shown in FIG. 6B, the laser is identified as a regular block.
  • the x-axis orientation is p 3 -p 1 and the y-axis orientation is p 2 -O.
  • the coordinates of the marker in the robot coordinate system are finally determined.
  • a coordinate system as shown in Fig. 7B is defined.
  • the identification block is discretized into a plurality of laser points.
  • the point cloud matching scheme is used to detect the coordinates of the marker in the lidar coordinate system, and then the calibration parameters are used to solve the identifier in the robot coordinates.
  • the coordinates of the system. 7A and 7B are schematic diagrams of coordinate systems defined for visual identification and laser identification, in accordance with an embodiment of the present application.
  • the robot can be used to optimize the position of the mark by repeatedly passing or observing the pose of the mark. Different from the identifier containing the ID, the identifier that does not contain the ID does not have a specific ID when it matches, and the pose information needs to be used for correspondence. As shown in FIG. 8 , there is a certain positional interval between the logo and the logo that the robot passes or observes multiple times.
  • the robot detects the logo, the pose of the robot is obtained, and the pose of the robot is recorded. For the same logo, the recorded robot's multiple poses are similar. By merging similar pose information, the optimal location estimate of the marker can be solved.
  • FIG. 8 is a schematic diagram of merging similar gesture poses according to an embodiment of the present application.
  • the identification when the identification itself contains non-identification type graphic information, it is determined whether the same identification is by the similarity of the pose of the robot when passing the identification. Alternatively, it can be said that the similar pose of the robot when it is identified is used as the corresponding ID of the logo. Subsequent processes are essentially the same as in the case where the identification described above includes ID information. In an embodiment, after determining the same identifier, a similar pose of the robot passing through the identifier may be obtained, and after the conversion, the identified poses are merged to obtain an optimal positioning estimate of different identifiers. Therefore, in the case where the identification does not contain any information, the identification ID is substantially given by the similarity of the pose of the robot when it is identified, and the subsequent procedure is the same as in the case where the previously described identification includes the identification information.
  • the second identifier includes non-identification type graphic information
  • updating the current pose of the positioning object based on the second identifier may be: using the positioning object to pass the same logo when the pose is similar And determining whether the identifier passed by the positioning object is a second identifier; if yes, recording a plurality of pose information and an error of the positioning object that is subjected to the second identifier, and converting into a plurality of pose information and an error of the second identifier; Merging the plurality of pose information of the second identifier by the Gaussian model to estimate the optimal pose information and the error of the second identifier; and based on the pose of the second identifier relative to the positioning object, converting from the second identifier via the coordinate
  • the optimal pose information directly extracts the current pose information of the positioning object.
  • the positioning object in the case that a plurality of identifiers are set on the travel path of the positioning object, for each of the identifiers, the positioning object may be subjected to the identifier in the above manner.
  • the pose is updated to reduce the cumulative offset and enable closed detection of the SLAM system.
  • the method of the embodiment of the present application further includes: adding a third identifier at a docking device on a travel path of the positioning object; and in a case where the positioning object passes the third identifier, The positioning object completes the docking operation with the docking device based on the third identifier. For example, when the positioning object carries the carrying item, if the positioning object passes the third identification, the relative position of the positioning object and the docking device can be adjusted based on the third identifier to achieve accurate docking.
  • SLAM navigation method in order to solve a high-precision positioning requirement such as a precise docking of a conveyor belt or a robot arm, one or more markers are added in the docking area to perform precise positioning of the positioning object.
  • a high-precision positioning requirement such as a precise docking of a conveyor belt or a robot arm
  • one or more markers are added in the docking area to perform precise positioning of the positioning object.
  • the specific implementation is the same as that described above for the robot, and will not be described again in order to avoid redundancy.
  • a ribbon may be added between the markers.
  • the angular deviation of the robot relative to the ribbon is corrected, and the robot is allowed to walk along the ribbon, thereby facilitating accurate positioning when the robot encounters the logo.
  • adding a second identifier on a travel path of the positioning object includes: adding a plurality of second identifiers on a travel path of the positioning object; And, wherein, in a case that the positioning object passes the second identifier, updating a current pose of the positioning object based on the second identifier, further comprising: after the positioning object passes the plurality of second identifiers In the case, the current pose of the positioning object is updated based on the plurality of second identifiers.
  • the plurality of second identifiers can be two types of identifiers.
  • FIGS. 9A and 9B are schematic views of a manner of connection between an identification and a ribbon according to an embodiment of the present application.
  • FIGS. 9A and 9B typical two types of identification and ribbon connection methods are shown (identification is based on a two-dimensional code).
  • FIG. 9A shows the manner of connecting straight lines, that is, when the two marks are aligned in the x or y direction, the two marks are directly connected using the straight line ribbon.
  • FIG. 9B shows a curved connection, that is, when the two marks are inconsistent in the x and y directions, or the angles are not uniform, a smooth curve ribbon connection can be employed.
  • the connection mode of the ribbon is not limited to the two connection modes, and the ribbon curve is smooth, and the robot can walk along the ribbon.
  • ⁇ 1 and ⁇ 2 are two parameters indicating the control weights caused by controlling the angular deviation and the positional deviation.
  • FIG. 10 is a schematic view showing calculation of a deviation between a robot and a color ribbon according to an embodiment of the present application.
  • the auxiliary positioning belts connecting the plurality of second identifiers are included between two adjacent second identifiers of the plurality of second identifiers, so that The positioning object corrects the angular deviation of the positioning object relative to the auxiliary positioning belt when the auxiliary positioning belt passes, so that the positioning object travels along the auxiliary positioning belt, ensuring that the positioning object can pass and identify the plurality of second identifiers.
  • the plurality of second identifiers may include not only two types of identifiers but also at least two types of identifiers.
  • the set positions of the at least two types of marks may be different shapes such as a matrix, a rectangle, and the like, and an auxiliary positioning band is set between any two types of marks.
  • a navigation device for simultaneous positioning and map creation including: an initialization area setting unit configured to set a positioning object initialization area; and a first identifier adding unit configured to Adding a first identifier to the positioning object initialization area, the first identifier includes pose information or identification information, and the positioning object starting unit is configured to start the positioning object from the positioning object initialization area, and determine an initial of the positioning object based on the first identifier a second identifier adding unit configured to add a second identifier on the travel path of the positioning object; and a pose updating unit configured to update the positioning object based on the second identifier if the positioning object passes the second identifier The current pose.
  • the simultaneous positioning and map creation navigation apparatus 200 includes: an initialization area setting unit 210 configured to set a positioning object initialization area; and a first identifier adding unit 220 configured to The positioning object initialization area set by the initialization area setting unit 210 adds a first identifier, the first identifier includes pose information or identification information, and the positioning object activation unit 230 is set to a positioning object set by the positioning object from the initialization area setting unit 210.
  • the second identifier adding unit 240 is configured to add a second identifier on the traveling path of the positioning object; and the pose The updating unit 250 is configured to update the current pose of the positioning object based on the second identifier if the positioning object passes the second identifier added by the second identifier adding unit 240.
  • the positioning object is a robot that travels in simultaneous positioning and map creation navigation and a mobile device in a carrying conveyor or robotic arm.
  • the first identifier includes pose information
  • the positioning object activation unit is configured to: based on the pose of the first identifier relative to the positioning object, The pose information of the first identifier directly extracts initial pose information of the positioning object.
  • the first identifier includes identification information
  • the positioning object activation unit is configured to: use the identification information of the first identifier to match, and record multiple pose information of the positioning object. And the error, and converted into the plurality of pose information and the error of the first identifier; the plurality of pose information of the first identifier is fused by the Gaussian model to estimate the optimal pose information and the error of the first identifier; An identifier is positioned relative to the position of the positioning object, and the initial pose information of the positioning object is directly extracted from the optimal pose information of the first identifier via coordinate conversion.
  • the second identifier includes pose information, identification information, or non-identification type graphic information.
  • the second identifier includes pose information
  • the pose update unit is configured to: based on the pose of the second identifier relative to the positioning object, via coordinate conversion
  • the pose information of the second identifier directly extracts current pose information of the positioning object.
  • the second identifier includes identification information
  • the pose update unit is configured to: use the identification information of the second identifier to match, and record multiple pose information of the positioning object. And the error, and converted into the plurality of pose information and the error of the second identifier; the plurality of pose information is fused by the Gaussian model to estimate the optimal pose information and the error of the second identifier; based on the second identifier Positioning the pose of the object, and directly extracting the current pose information of the positioning object from the optimal pose information of the second identifier via coordinate conversion.
  • the second identifier includes non-identification type graphic information
  • the pose updating unit is configured to: determine the positioning object by using the similarity of the posture when the positioning object passes the same identification Whether the traversed identifier is a second identifier; recording a plurality of pose information and an error of the locating object passing the second identifier, and converting into a plurality of pose information and an error of the second identifier; and merging the second identifier by the Gaussian model a plurality of pose information to estimate an optimal pose information and an error of the second identifier; and based on the pose of the second identifier relative to the positioning object, directly extracting the optimal pose information from the second identifier via coordinate transformation Position the current pose information of the object.
  • the first identifier adding unit is configured to add a plurality of paths on the travel path of the positioning object.
  • the pose update unit is further configured to update the current pose of the positioning object based on the plurality of second identifiers when the positioning object passes the plurality of second identifiers.
  • the method further includes: an auxiliary positioning unit, configured to add an auxiliary positioning belt connecting two adjacent second identifiers of the plurality of second identifiers, and in the positioning object The angular deviation of the positioning object relative to the auxiliary positioning belt is corrected when the auxiliary positioning belt is passed, so that the positioning object travels along the auxiliary positioning belt, ensuring that the positioning object can pass and recognize the second identification.
  • an auxiliary positioning unit configured to add an auxiliary positioning belt connecting two adjacent second identifiers of the plurality of second identifiers, and in the positioning object The angular deviation of the positioning object relative to the auxiliary positioning belt is corrected when the auxiliary positioning belt is passed, so that the positioning object travels along the auxiliary positioning belt, ensuring that the positioning object can pass and recognize the second identification.
  • the method further includes: a third identifier adding unit, configured to add a third identifier on the docking device on the travel path of the positioning object; and the docking unit is set to be in the In the case that the positioning object passes the third identifier, the docking operation is completed with the docking device based on the third identifier.
  • the first identifier adding unit and the second identifier adding unit can be integrated as the same identifier for adding the identifier.
  • the unit, and a part of the functions in the positioning object activation unit may also be included in the pose updating unit for updating the position information of the positioning object based on the identification.
  • a simultaneous navigation and map creation navigation system including: positioning an object, configured to locate and construct a map in real time in the scene; the first identifier, including pose information or identification information a second identifier; a simultaneous positioning and map creation device, comprising: an initialization area setting unit, configured to set a positioning object initialization area; a first identifier adding unit, configured to add the first identifier to the positioning object initialization area; Activating unit, configured to enable the positioning object to be activated from the positioning object initialization area, and determine an initial pose of the positioning object based on the first identifier; the second identifier adding unit is configured to add the second identifier to the traveling path of the positioning object And a pose updating unit configured to update the current pose of the positioning object based on the second identifier if the positioning object passes the second identifier.
  • the simultaneous positioning and map creation navigation system 300 includes: a positioning object 301 configured to locate and construct a map in real time in the scene; the first identifier 302 includes pose information or The identification information; the second identifier 303; the simultaneous positioning and map creation device 310, comprising: an initialization area setting unit 311, configured to set a positioning object initialization area; a first identification adding unit 312, configured to add the first identifier 302 to the The positioning object initialization area set by the initialization area setting unit 311; the positioning object activation unit 313 is set such that the positioning object 301 is activated from the positioning object initialization area set by the initialization area setting unit 311, and is added based on the first identification adding unit 312.
  • the first identifier 302 determines an initial pose of the positioning object 301; the second identifier adding unit 314 is configured to add the second identifier 303 to the travel path of the positioning object; and the pose updating unit 315 is configured to be located at the positioning object 301.
  • the base In the case of the second identifier 303 added by the second identifier adding unit 314, the base The second update identifier 303 to locate objects 301 current pose.
  • the positioning object is a robot that travels in the simultaneous positioning and map creation navigation and a mobile device that carries a conveyor belt or a robotic arm.
  • the first identifier includes pose information
  • the positioning object activation unit is further configured to: based on the pose of the first identifier relative to the positioning object, via coordinate conversion
  • the pose information of the first identifier directly extracts initial pose information of the positioning object.
  • the first identifier includes identification information
  • the positioning object activation unit is further configured to: use the identification information of the first identifier to match, and record multiple poses of the positioning object.
  • Information and error and converted into a plurality of pose information and error of the first identifier; merging the plurality of pose information of the first identifier by a Gaussian model to estimate an optimal pose information and an error of the first identifier;
  • the first marker is directly extracted from the pose posture of the first marker with respect to the pose of the positioning object, and the initial pose information of the positioning object is directly extracted from the optimal pose information of the first marker.
  • the second identifier includes pose information, identification information, or non-identification type graphic information.
  • the second identifier includes pose information
  • the pose update unit is further configured to: based on the pose of the second identifier relative to the positioning object, via coordinate conversion
  • the pose information of the second identifier directly extracts current pose information of the positioning object.
  • the second identifier includes identification information
  • the pose updating unit is further configured to: use the identification information of the second identifier to match, and record multiple poses of the positioning object.
  • the current pose information of the positioning object is directly extracted from the optimal pose information of the second identifier via the coordinate transformation.
  • the second identifier includes non-identification type graphic information
  • the pose updating unit is further configured to: determine the positioning by using the similarity of the posture when the positioning object passes the same identification. Whether the identifier passed by the object is a second identifier; recording a plurality of pose information and errors of the second identified positioning object, and converting into a plurality of pose information and an error of the second identifier; and merging the second identifier by the Gaussian model a plurality of pose information to estimate an optimal pose information and an error of the second identifier; and based on the pose of the second identifier relative to the positioning object, directly extracting from the optimal pose information of the second identifier via coordinate transformation The current pose information of the positioning object.
  • the first identifier adding unit is configured to add a plurality of second identifiers to the positioning object when the second identifier includes a plurality of second identifiers
  • the pose updating unit is further configured to update the current pose of the positioning object based on the plurality of second identifiers when the positioning object passes the plurality of second identifiers.
  • the auxiliary positioning belt is included between two adjacent second identifiers of the plurality of second identifiers
  • the simultaneous positioning and map creating device comprises: an auxiliary positioning unit, And being arranged to correct an angular deviation of the positioning object relative to the auxiliary positioning belt when the positioning object passes the auxiliary positioning belt, so that the positioning object travels along the auxiliary positioning belt.
  • the simultaneous positioning and map creation device further includes: a third identification adding unit configured to add a third to the docking device on the travel path of the positioning object And the docking unit is configured to complete the docking operation with the docking device based on the third identifier if the positioning object passes the third identifier.
  • the simultaneous positioning and map creation navigation method, apparatus and system provided by the present application can improve the positioning accuracy and robustness of the simultaneous positioning and map creation navigation system, and adapt it to an industrial environment requiring precise positioning.

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Electromagnetism (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Navigation (AREA)
  • Manipulator (AREA)

Abstract

提供了一种结合标识的同时定位和地图创建导航方法、装置和系统。该方法包括:设置定位对象初始化区域(S1);其中,定位对象初始化区域、定位对象的行进路径上和定位对象的对接装置处中的至少之一设置有标识,标识包括位姿信息、标识信息和非标识类图形信息中的至少一种;控制定位对象执行以下操作中的至少一种:从定位对象初始化区域启动,并基于标识确定定位对象的初始位姿(S3);在行进路径上经过标识的情况下,基于标识更新定位对象的当前位姿(S5);在与对接装置进行对接时,基于标识调整定位对象与对接装置的相对位姿。

Description

结合标识的同时定位和地图创建导航方法、装置及系统
本申请要求在2017年08月16日提交中国专利局、申请号为201710703818.3的中国专利申请的优先权,该申请的全部内容通过引用结合在本申请中。
技术领域
本申请涉及自动导航领域,例如涉及结合标识的同时定位和地图创建导航方法、装置及系统。
背景技术
基于激光雷达或视觉的同时定位与地图创建(Simultaneous localization and mapping,SLAM)导航方案,可在一个完全未知的场景进行实时定位并构建出场景的二维/三维地图。
例如,对于完全未知环境,SLAM系统以初始化的位姿为基础,在定位对象行走过程中不断定位并更新地图,生成以初始化位姿为基准的地图。另外,对于已建立地图的环境,SLAM系统在初始化位姿区域搜索并匹配地图,获取初始位姿信息,并在定位对象行走过程中不断定位并更新地图。
SLAM导航无需在场景中布设标志物,凭借其便捷性成为室内导航方案中重要的分支。SLAM导航的精度主要取决于闭合精度,当场景变大时,无法有效地进行地图闭合,将极大地影响定位与地图效果。另外SLAM系统精度受到传感器精度影响,低成本SLAM系统无法满足传送带对接、机械臂操作等所需的精准定位。
因此,基于上述问题,需要改进的同时定位和地图创建导航方法、装置及系统。
发明内容
本申请提供了一种新颖的和改进的结合标识的同时定位和地图创建导航方法、装置及系统。
本申请提供了一种结合标识的同时定位和地图创建导航方法,包括:设置定位对象初始化区域;在所述定位对象初始化区域添加第一标识,所述第一标识包括位姿信息或者标识信息;使得定位对象从所述定位对象初始化区域启动, 并基于所述第一标识确定定位对象的初始位姿;在定位对象的行进路径上添加第二标识;和,在定位对象经过第二标识的情况下,基于所述第二标识更新定位对象的当前位姿。
在上述结合标识的同时定位和地图创建导航方法中,所述定位对象是同时定位和地图创建导航中行进的机器人和携带传送带、货架、笼车以及机械臂的移动装置中的一个。
在上述结合标识的同时定位和地图创建导航方法中,所述第一标识包含位姿信息,且基于所述第一标识确定定位对象的初始位姿包括:基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的位姿信息直接提取所述定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航方法中,所述第一标识包含标识信息,且基于所述第一标识确定定位对象的初始位姿包括:利用所述第一标识的标识信息匹配,记录定位对象的多个位姿信息与误差协方差矩阵,并转换为第一标识的多个位姿信息与误差协方差矩阵;通过高斯模型融合所述多个第一标识的位姿信息以估计所述第一标识的最优位姿信息和误差协方差矩阵;基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的最优位姿信息直接提取所述定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航方法中,所述第二标识包含位姿信息,或者包含标识信息,或者包含非标识类图形信息。
在上述结合标识的同时定位和地图创建导航方法中,所述第二标识包含位姿信息,且基于所述第二标识更新定位对象的当前位姿包括:基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的位姿信息直接提取所述定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航方法中,所述第二标识包含标识信息,且基于所述第二标识更新定位对象的当前位姿包括:利用所述第二标识的标识信息匹配,记录定位对象的多个位姿信息与误差协方差矩阵,并转换为第二标识的多个位姿信息与误差协方差矩阵;通过高斯模型融合所述多个位姿信息以估计所述第二标识的最优位姿信息和误差协方差矩阵;基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接提取所述定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航方法中,所述第二标识包含非 标识类图形信息,且基于所述第二标识更新定位对象的当前位姿包括:利用定位对象经过同一标识时位姿的相似性,确定定位对象所经过的标识是否为第二标识;记录经过第二标识的定位对象的多个位姿信息与误差协方差矩阵,并转换为第二标识的多个位姿信息和误差协方差矩阵;通过高斯模型融合第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差协方差矩阵;基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接提取所述定位对象的当前位姿信息
在上述结合标识的同时定位和地图创建导航方法中,其中,当所述第二标识包括多个第二标识时,在所述定位对象的行进路径上添加第二标识,包括:在定位对象的行进路径上添加多个第二标识;和,其中,在所述定位对象经过所述第二标识的情况下,基于所述第二标识更新所述定位对象的当前位姿,还包括:在定位对象经过所述多个第二标识的情况下,基于所述多个第二标识更新定位对象的当前位姿。
在上述结合标识的同时定位和地图创建导航方法中,在多个第二标识的两个相邻第二标识之间包括连接所述第二标识的辅助定位带,以在当所述定位对象经过所述辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得所述定位对象沿着所述辅助定位带行进,确保所述定位对象可以通过并识别到第二标识。
在上述结合标识的同时定位和地图创建导航方法中,还包括:在所述定位对象的行进路径上的对接装置上添加第三标识;在所述定位对象经过所述第三标识的情况下,基于所述第三标识与所述对接装置完成对接操作。
本申请提供了一种结合标识的同时定位和地图创建导航装置,包括:初始化区域设置单元,设置为设置定位对象初始化区域;第一标识添加单元,设置为在所述定位对象初始化区域添加第一标识,所述第一标识包括位姿信息或者标识信息;定位对象启动单元,设置为使得定位对象从所述定位对象初始化区域启动,并基于所述第一标识确定定位对象的初始位姿;第二标识添加单元,设置为在定位对象的行进路径上添加第二标识;和位姿更新单元,设置为在定位对象经过第二标识的情况下,基于所述第二标识更新定位对象的当前位姿。
在上述结合标识的同时定位和地图创建导航装置中,所述定位对象包括同时定位和地图创建的导航中行进的机器人和携带传送带或者机械臂的移动装置中的一个。
在上述结合标识的同时定位和地图创建导航装置中,所述第一标识包含位姿信息,且所述定位对象启动单元还设置为:基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的位姿信息直接提取所述定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,所述第一标识包含标识信息,且所述定位对象启动单元还设置为:利用所述第一标识的标识信息匹配,记录定位对象的多个位姿信息与误差协方差矩阵,并转换为第一标识的多个位姿信息与误差协方差矩阵;通过高斯模型融合所述第一标识的多个位姿信息以估计所述第一标识的最优位姿信息和误差协方差矩阵;基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的最优位姿信息直接提取所述定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,所述第二标识包含位姿信息、标识信息,或者非标识类图形信息。
在上述结合标识的同时定位和地图创建导航装置中,所述第二标识包含位姿信息,且所述位姿更新单元还设置为:基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的位姿信息直接提取所述定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,所述第二标识包含标识信息,且所述位姿更新单元还设置为:利用所述第二标识的标识信息匹配,记录定位对象的多个位姿信息与误差协方差矩阵,并转换为第二标识的多个位姿信息与误差协方差矩阵;通过高斯模型融合所述多个位姿信息以估计所述第二标识的最优位姿信息和误差协方差矩阵;基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接提取所述定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,所述第二标识包含非标识类图形信息,且所述位姿更新单元设置为:利用定位对象经过同一标识时位姿的相似性,确定定位对象所经过的标识是否为第二标识;记录经过第二标识的定位对象的多个位姿信息与误差协方差矩阵,并转换为第二标识的多个位姿信息和误差协方差矩阵;通过高斯模型融合第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差协方差矩阵;基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接提取所述定 位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,其中,当所述第二标识包括多个第二标识时,包括:所述第一标识添加单元还设置为在定位对象的行进路径上添加多个第二标识;且所述位姿更新单元设置为在定位对象经过所述多个第二标识的情况下,基于所述多个第二标识更新定位对象的当前位姿。
在上述结合标识的同时定位和地图创建导航装置中,还包括:辅助定位单元,设置为在多个第二标识的两个相邻第二标识之间添加连接所述两个相邻第二标识的辅助定位带,并在所述定位对象经过所述辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得所述定位对象沿着所述辅助定位带行进。
在上述结合标识的同时定位和地图创建导航装置中,还包括:第三标识添加单元,设置为在所述定位对象的行进路径上的对接装置上添加第三标识;对接单元,设置为在所述定位对象经过所述第三标识的情况下,基于所述第三标识与所述对接装置完成对接操作。
本申请提供了一种结合标识的同时定位和地图创建导航系统,包括:定位对象,用于在场景中实时定位并构建地图;第一标识,包括位姿信息或者标识信息;第二标识;同时定位和地图创建装置,包括:初始化区域设置单元,设置为设置定位对象初始化区域;第一标识添加单元,设置为将所述第一标识添加到所述定位对象初始化区域;定位对象启动单元,设置为使得定位对象从所述定位对象初始化区域启动,并基于所述第一标识确定定位对象的初始位姿;第二标识添加单元,设置为将第二标识添加到所述定位对象的行进路径上;和位姿更新单元,设置为在定位对象经过第二标识的情况下,基于所述第二标识更新定位对象的当前位姿。
在上述结合标识的同时定位和地图创建导航系统中,所述定位对象是同时定位和地图创建导航中行进的机器人和携带传送带或者机械臂的移动装置中的一个。
在上述结合标识的同时定位和地图创建导航系统中,所述第一标识包含位姿信息,且所述定位对象启动单元设置为:基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的位姿信息直接提取所述定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,所述第一标识包含标识信息,且所述定位对象启动单元设置为:利用所述第一标识的标识信息匹配, 记录定位对象的多个位姿信息与误差协方差矩阵,并转换为第一标识的多个位姿信息与误差协方差矩阵;通过高斯模型融合所述第一标识的多个位姿信息以估计所述第一标识的最优位姿信息和误差协方差矩阵;基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的最优位姿信息直接提取所述定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,所述第二标识包含位姿信息标识信息,或者非标识类图形信息。
在上述结合标识的同时定位和地图创建导航系统中,所述第二标识包含位姿信息,且所述位姿更新单元还设置为:基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的位姿信息直接提取所述定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,所述第二标识包含标识信息,且所述位姿更新单元还设置为:利用所述第二标识的标识信息匹配,记录定位对象的多个位姿信息与误差协方差矩阵,并转换为第二标识的多个位姿信息与误差协方差矩阵;通过高斯模型融合所述多个位姿信息以估计所述第二标识的最优位姿信息和误差协方差矩阵;基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接提取所述定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,所述第二标识包含非标识类图形信息,且所述位姿更新单元还设置为:利用定位对象经过同一标识时位姿的相似性,确定定位对象所经过的标识是否为第二标识;记录经过第二标识的定位对象的多个位姿信息与误差协方差矩阵,并转换为第二标识的多个位姿信息和误差协方差矩阵;通过高斯模型融合第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差协方差矩阵;基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接提取所述定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,其中,当所述第二标识包括多个第二标识时,包括:所述第一标识添加单元,还设置为将多个第二标识添加到定位对象的行进路径上;且所述位姿更新单元设置为在定位对象经过所述多个第二标识的情况下,基于所述多个第二标识更新定位对象的当前位姿。
在上述结合标识的同时定位和地图创建导航系统中,在多个第二标识中的两个相邻第二标识之间包括辅助定位带,所述同时定位和地图创建装置还包括:辅助定位单元,设置为在所述定位对象经过所述辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得所述定位对象沿着所述辅助定位带行进。
在上述结合标识的同时定位和地图创建导航系统中,还包括:第三标识添加单元,设置为在所述定位对象的行进路径上的对接装置上添加第三标识;对接单元,设置为在所述定位对象经过所述第三标识的情况下,基于所述第三标识与所述对接装置完成对接操作。
本申请实施例提供一种结合标识的同时定位和地图创建导航方法,包括:设置定位对象初始化区域;其中,所述定位对象初始化区域、所述定位对象的行进路径上和所述定位对象的对接装置处中的至少之一设置有标识,所述标识包括位姿信息、标识信息和非标识类图形信息中的至少一种;控制定位对象执行以下操作中的至少一种:从所述定位对象初始化区域启动,并基于所述标识确定定位对象的初始位姿;在行进路径上经过所述标识的情况下,基于所述标识更新所述定位对象的当前位姿;在与对接装置进行对接时,基于所述标识调整所述定位对象与所述对接装置的相对位姿。
在一实施例中,所述定位对象包括同时定位和地图创建导航中行进的机器人和需要与对接装置进行对接的移动装置。
在一实施例中,所述对接装置包括传送带或者机械臂。
在一实施例中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含位姿信息,且,基于所述第一标识确定定位对象的初始位姿包括:基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的位姿信息直接提取所述定位对象的初始位姿信息。
在一实施例中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含标识信息,且,基于所述第一标识确定定位对象的初始位姿包括:在所述第一标识上启动时,记录所述第一标识包含的标识信息,以所述标识信息为坐标原点进行定位并制作地图;在所述定位对象重复多次经过所述第一标识时,利用所述第一标识的标识信息进行匹配获得所述定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第一标识的多个位姿信息和误差;通过高斯模型融合所述第一标识的多个位姿信息以估计所述第一标识的最优位姿信息和误差;和基于所述第一标识相对于定位对象的位姿,经 由坐标转换从所述第一标识的最优位姿信息直接确定所述定位对象的初始位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含位姿信息,且基于所述第二标识更新定位对象的当前位姿包括:基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的位姿信息直接提取所述定位对象的当前位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含标识信息,且,基于所述第二标识更新定位对象的当前位姿包括:在所述定位对象重复多次经过所述第二标识时,利用所述第二标识的标识信息进行匹配获得所述定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含非标识类图形信息,且,基于所述第二标识更新定位对象的当前位姿包括:利用所述定位对象经过同一标识时位姿的相似性,确定所述定位对象重复多次所经过的标识是否为第二标识;如果是所述第二标识,记录经过所述第二标识的所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,且设置有多个第二标识。
在一实施例中,在所述多个第二标识的两个相邻第二标识之间包括连接所述两个相邻第二标识的辅助定位带,以在所述定位对象经过所述辅助定位带时校正所述定位对象相对于所述辅助定位带的角度偏差,以使得所述定位对象沿着所述辅助定位带行进。
本申请实施例提供一种结合标识的同时定位和地图创建导航装置,包括:初始化区域设置单元、定位对象启动单元、位姿更新单元,以及对接单元。其 中,初始化区域设置单元,设置为设置定位对象初始化区域;其中,所述定位对象初始化区域、所述定位对象的行进路径上和所述定位对象的对接装置处中的至少之一设置有标识,所述标识包括位姿信息、标识信息和非标识类图形信息中的至少一种;定位对象启动单元,设置为控制定位对象执行以下操作中的至少一种:从所述定位对象初始化区域启动,并基于所述标识确定定位对象的初始位姿;位姿更新单元,设置为在行进路径上经过标识的情况下,基于所述标识更新定位对象的当前位姿;对接单元,设置为在与对接装置进行对接时,基于所述标识调整所述定位对象与所述对接装置的相对位姿。
在一实施例中,所述定位对象包括同时定位和地图创建的导航中行进的机器人和需要与对接装置进行对接的移动装置。
在一实施例中,所述对接装置包括携带传送带或者机械臂。
在一实施例中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含位姿信息,且,所述定位对象启动单元还设置为:基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的位姿信息直接提取所述定位对象的初始位姿信息。
在一实施例中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含标识信息,且,所述定位对象启动单元还设置为:在所述第一标识上启动时,记录所述第一标识包含的标识信息,以所述标识信息为坐标原点进行定位并制作地图;在所述定位对象重复多次经过所述第一标识时,利用所述第一标识的标识信息进行匹配获得所述定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第一标识的多个位姿信息与误差;通过高斯模型融合所述第一标识的多个位姿信息以估计所述第一标识的最优位姿信息和误差;和基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的最优位姿信息直接确定所述定位对象的初始位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含位姿信息,且所述位姿更新单元还设置为:基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的位姿信息直接提取所述定位对象的当前位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含标识信息,且,所述位姿更新单元还设置为:在所述定位对象重复多次经过所述第二标识时,利用所述第二标识的标识信息进行匹配获得所述 定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含非标识类图形信息,且,所述位姿更新单元设置为:利用所述定位对象经过同一标识时位姿的相似性,确定所述定位对象重复多次所经过的标识是否为第二标识;如果是所述第二标识,记录经过第二标识的所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,且设置有多个第二标识。
在一实施例中,还包括:辅助定位单元,设置为在多个第二标识的两个相邻第二标识之间添加连接所述两个相邻第二标识的辅助定位带,并在所述定位对象经过所述辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得所述定位对象沿着所述辅助定位带行进。
本申请实施例提供一种结合标识的同时定位和地图创建导航系统,包括:定位对象,设置为在场景中实时定位并构建地图;标识,包括位姿信息、标识信息和非标识类图形信息中的至少一种;同时定位和地图创建装置,包括:初始化区域设置单元,设置为设置定位对象初始化区域;其中,所述定位对象初始化区域、所述定位对象的行进路径上和所述定位对象的对接装置处中的至少之一设置有标识;定位对象启动单元,设置为控制定位对象执行以下操作中的至少一种:从所述定位对象初始化区域启动,并基于所述标识确定定位对象的初始位姿;位姿更新单元,设置为在行进路径上经过标识的情况下,基于所述标识更新定位对象的当前位姿;对接单元,设置为在与对接装置进行对接时,基于所述标识调整所述定位对象与所述对接装置的相对位姿。
在一实施例中,所述定位对象包括同时定位和地图创建的导航中行进的机器人和需要与对接装置进行对接的移动装置。
在一实施例中,所述对接装置包括携带传送带或者机械臂。
在一实施例中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含位姿信息,且,所述定位对象启动单元设置为:基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的位姿信息直接提取所述定位对象的初始位姿信息。
在一实施例中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含标识信息,且,所述定位对象启动单元设置为:在所述第一标识上启动时,记录所述第一标识包含的标识信息,以所述标识信息为坐标原点进行定位并制作地图;在所述定位对象重复多次经过所述第一标识时,利用所述第一标识的标识信息进行匹配获得所述定位对象的位姿信息,记录定位对象的多个位姿信息和误差,并转换为所述第一标识的多个位姿信息与误差;通过高斯模型融合所述第一标识的多个位姿信息以估计所述第一标识的最优位姿信息和误差;和基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的最优位姿信息直接确定所述定位对象的初始位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含位姿信息,且所述位姿更新单元还设置为:基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的位姿信息直接提取所述定位对象的当前位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含标识信息,且,所述位姿更新单元还设置为:在所述定位对象重复多次经过所述第二标识时,利用所述第二标识的标识信息进行匹配获得所述定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息与误差;通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含非标识类图形信息,且所述位姿更新单元还设置为:利用所述定位对象经过同一标识时位姿的相似性,确定所述定位对象重复多次所经过的标识是否为第二标识;如果是所述第二标识,记录经过第二标识的所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;通过 高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
在一实施例中,所述定位对象的行进路径上设置的标识为第二标识,且设置有多个第二标识。
在一实施例中,在多个第二标识中的两个相邻第二标识之间包括辅助定位带,所述同时定位和地图创建装置还包括:辅助定位单元,设置为在所述定位对象经过所述辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得所述定位对象沿着所述辅助定位带行进。
附图概述
图1是根据本申请实施例的结合标识的同时定位和地图创建导航方法的示意性流程图;
图2是示出了根据本申请实施例的标识的示例的示意图;
图3是示出根据本申请实施例的标识的射影变换的示意图;
图4是根据本申请实施例的标识图像的坐标转换示意图;
图5是根据本申请实施例的位姿信息融合的示意图;
图6A和图6B是根据本申请实施例的视觉标识和激光标识的示意图;
图7A和图7B是根据本申请实施例的对于视觉标识和激光标识所定义的坐标系的示意图;
图8是根据本申请实施例的对相近的标识位姿进行融合的示意图;
图9A和图9B是根据本申请实施例的标识与色带之间的连接方式的示意图;
图10是示出根据本申请实施例的计算机器人与色带之间的偏差的示意图;
图11是根据本申请实施例的结合标识的同时定位和地图创建导航装置的示意性框图;
图12是根据本申请实施例的结合标识的同时定位和地图创建导航系统的示意性框图。
具体实施方式
以下描述用于公开本发明以使本领域技术人员能够实现本发明。以下描述中的优选实施例只作为举例,本领域技术人员可以想到其他显而易见的变型。 在以下描述中界定的本发明的基本原理可以应用于其他实施方案、变形方案、改进方案、等同方案以及没有背离本发明的精神和范围的其他技术方案。
以下说明书和权利要求中使用的术语和词不限于字面的含义,而是仅由本发明人使用以使得能够清楚和一致地理解本发明。因此,对本领域技术人员很明显仅为了说明的目的而不是为了如所附权利要求和它们的等效物所定义的限制本发明的目的而提供本发明的各种实施例的以下描述。
可以理解的是,术语“一”应理解为“至少一”或“一个或多个”,即在一个实施例中,一个元件的数量可以为一个,而在另外的实施例中,该元件的数量可以为多个,术语“一”不能理解为对数量的限制。
虽然比如“第一”、“第二”等的序数将用于描述各种组件,但是在这里不限制那些组件。该术语仅用于区分一个组件与另一组件。例如,第一组件可以被称为第二组件,且同样地,第二组件也可以被称为第一组件,而不脱离申请构思的教导。在此使用的术语“和/或”包括一个或多个关联的列出的项目的任何和全部组合。
在这里使用的术语仅用于描述各种实施例的目的且不意在限制。如在此使用的,单数形式意在也包括复数形式,除非上下文清楚地指示例外。另外将理解术语“包括”和/或“具有”当在该说明书中使用时指定所述的特征、数目、步骤、操作、组件、元件或其组合的存在,而不排除一个或多个其它特征、数目、步骤、操作、组件、元件或其组的存在或者附加。
包括技术和科学术语的在这里使用的术语具有与本领域技术人员通常理解的术语相同的含义,只要不是不同地限定该术语。应当理解在通常使用的词典中限定的术语具有与相关技术中的术语的含义一致的含义。
下面结合附图和实施方式对本申请作详细的说明。
本申请实施例提供了一种结合标识的同时定位和地图创建导航方法,包括:设置定位对象初始化区域;在该定位对象初始化区域添加第一标识,该第一标识包括位姿信息或者标识信息;使得定位对象从该定位对象初始化区域启动,并基于该第一标识确定定位对象的初始位姿;在定位对象的行进路径上添加第二标识;和,在定位对象经过第二标识的情况下,基于该第二标识更新定位对象的当前位姿。
图1是根据本申请实施例的结合标识的同时定位和地图创建导航方法的示意性流程图。如图1所示,根据本申请实施例的结合标识的同时定位和地图创 建导航方法包括:步骤S1至步骤S5。
在步骤S1中,设置定位对象初始化区域。
在步骤S2中,在该定位对象初始化区域添加第一标识,该第一标识包括位姿信息或者标识信息。
在步骤S3中,使得定位对象从该定位对象初始化区域启动,并基于该第一标识确定定位对象的初始位姿。
在步骤S4中,在定位对象的行进路径上添加第二标识。
在步骤S5中,在定位对象经过第二标识的情况下,基于该第二标识更新定位对象的当前位姿。
因此,在根据本申请实施例的结合标识的同时定位和地图创建导航方法中,同时解决了相关技术中的定位对象初始化位姿不明确和在行进过程中的位姿偏移的问题,从而能够实现SLAM导航方法的高精度和高鲁棒性,可使其适应于需要精准定位的工业环境。
在一实施例中,通过设置定位对象初始化区域并添加第一标识,并使得定位对象从该定位对象初始化区域启动,以基于该第一标识确定定位对象的初始位姿,可以解决定位对象的初始化位姿不明确的问题。这样,通过基于明确的初始化位姿信息来进行同时定位和地图创建,可以实现SLAM导航系统的导航结果与真实环境的匹配,从而有助于该导航系统与真实应用结合。此外,如果是在已知地图情况下的再次定位,可以对定位对象的初始位姿做出约束,从而实现与地图的成功匹配。
另外,通过在定位对象的行进路径上添加标识,可以校正定位对象在行进过程中SLAM导航系统产生的定位偏差,从而有助于实现地图闭合,改进地图创建和定位效果。
在上述结合标识的同时定位和地图创建导航方法中,还包括:在定位对象的行进路径上添加多个第二标识;和,在定位对象经过该多个第二标识的情况下,基于该多个第二标识更新定位对象的当前位姿。
也就是说,在根据本申请实施例的结合标识的SLAM导航方法中,为了校正定位对象在行进过程中,SLAM导航系统的定位所产生的偏移,在大场景下,在定位对象的行进路径上添加多个标识,从而在定位对象经过每个标识时,基于标识对于定位对象在经过该标识时的当前位姿进行实时更新,从而减小定位对象在大场景中移动时的累积偏移,避免导致SLAM导航系统无法进行闭合检 测的情况。
这里,本领域技术人员可以理解,根据本申请实施例的结合标识的SLAM导航方法中的定位对象基于应用场景的不同可以是多种对象。例如,在通常的SLAM导航方法和导航系统中,是用于实现实时定位和地图构建的机器人,而在用于传送带对接、机械臂操作等精准定位时,是需要进行精准定位的传送带、机械臂等。
所以,在上述结合标识的SLAM导航方法中,该定位对象可以是SLAM导航中行进的机器人,或者适用于传送带对接或者机械臂操作等场景中的装置或设备。
在根据本申请实施例的结合标识的同时定位和地图创建的导航方法中,取决于基于标识确定定位对象的位姿的方法,该标识可以具有多种形式。
例如,该标识可以是二维码(QR Code,Data Matrix等编码方式)、条形码和其他视觉标识等人工标识物,如图2所示。图2是示出了根据本申请实施例的标识的示例的示意图。
另外,按照存储信息的不同,标识可以分为多个类型。例如,一类标识包含位姿信息(x,y,θ),另一类标识只包含标识(Identification,ID)信息,而不包含位姿信息,另一类标识可能包含非标识类图形信息。
基于标识类型的不同,相应的基于标识确定定位对象的位姿的应用方式也有不同,这将在下面进行详细说明。
首先,针对包含位姿信息的标识,标识的码字中存储的信息至少包含该标识在场景中的实际位置(x,y)。当标识的角度有偏差时,在码字中还需提供角度θ信息。
下面,将以定位对象是SLAM导航系统中的机器人为例,对基于标识确定定位对象的位置的具体方式进行说明。当然,本领域技术人员可以理解,定位对象并不限于机器人,也可以是如上所述适用于传送带对接或者机械臂操作等场景的移动装置。
以QR code二维码为例,在机器人启动前,将机器人移动至二维码上方。在机器人启动过程中,拍摄相机图像。由于相机在机器人上的位置固定,且地面平坦。因此,图像坐标系可通过射影变换,变换至俯视坐标系。即满足以下公式(1):
Figure PCTCN2018100869-appb-000001
其中,H为一个3×3矩阵,描述相机到真实坐标的射影变换。u和v是变换后的像素,u′和v′是变换前的像素,这里通过3×3矩阵H进行坐标变换。因此,通过该射影变换,可将原图像转化为俯视图,如图3所示。图3是示出根据本申请实施例的标识的射影变换的示意图。
在新图形中,图像正对二维码,且图像中每一个像素对应实际地面距离已知,图像中的像素点(u,v)对应唯一机器人坐标系下坐标(x,y),对应方式为如以下公式(2)所示:
Figure PCTCN2018100869-appb-000002
其中f为焦距,即一个像素对应的实际距离。u 0和v 0指的是相机焦距中心在图像中的横坐标和纵坐标,其是相机内参数的一部分。
上述公式(2)可以以矩阵的方式写为以下的公式(3):
Figure PCTCN2018100869-appb-000003
其中H i为图像坐标系到机器人坐标系的变换矩阵。
图4是根据本申请实施例的标识图像的坐标转换示意图。如图4所示,其中(u 0,v 0)为机器人坐标系原点在图像中的坐标。图像坐标为右u下v,机器人坐标系为右x上y。
接下来,检测图像中是否存在二维码。通过朝下的相机扫描二维码,获得二维码在相机坐标系下的坐标(u,v,θ)。根据变换矩阵H i,可得到二维码在机器人坐标系下的位置。通过反向变换,可得到机器人在二维码下的坐标,如以下公式(5)所示:
Figure PCTCN2018100869-appb-000004
其中,机器人在二维码坐标系下的实际的角度为-θ。
初始化所使用的二维码,码字中存储着二维码在场景中的实际坐标信息(x b,y b,θ b),通过解码二维码上的码字,获取二维码在场景中的实际位姿。因此, 机器人在场景下的坐标可由以下公式(6)表示:
Figure PCTCN2018100869-appb-000005
机器人在场景中的航向角同机器人在二维码处的航向角″-θ″加二维码自身的航向角θ b,如以下公式(7)所示:
θ w=θ b-θ       (7)
至此,当机器人在二维码可视的位置处启动时,通过二维码在相机中的位置和二维码的码字,可确定机器人的初始位姿(x w,y w,θ w)。以该初始位姿启动机器人,可以保证整体系统坐标系的一致性。
也就是说,在标识包含位姿信息的情况下,由于定位对象以其所配备的成像设备,比如相机拍摄标识并获取标识的位姿信息,在定位对象的位姿信息和标识的位姿信息之间存在确定的空间关系。因此,通过将标识包含的位姿信息基于标识在定位对象的成像设备中成像的位姿进行坐标转换,就可以从标识的位姿信息直接提取定位对象的初始位姿信息。
这里,本领域技术人员可以理解,可以通过标识在定位对象的成像设备中的成像确定标识与定位对象,例如上述机器人的相对位姿,也可以以其它方式确定机器人与标识的相对位姿,从而经由坐标转换从标识的位姿信息提取定位对象的位姿信息。并且相对地,当以其它方式确定机器人与标识的相对位姿时,也可以采用其它形式的坐标变换从标识的位姿信息直接提取定位对象的位姿信息的方法。
因此,在上述结合标识的SLAM导航方法中,该第一标识包含位姿信息,且基于该第一标识确定定位对象的初始位姿可以为:基于该第一标识相对于定位对象的位姿,经由坐标转换从该第一标识的位姿信息直接提取该定位对象的初始位姿信息。
并且,在上述结合标识的SLAM导航方法中,该第一标识相对于定位对象的位姿基于该第一标识在该定位对象准备的成像设备中所成的图像的位姿确定。
另外,在根据本申请实施例的结合标识的SLAM导航方法中,标识可以包含ID,而不包含位姿信息。针对该情况,机器人在该标识上启动时,记录标识包含的ID,并在SLAM过程中,以该ID为坐标原点进行定位并制作地图。当 机器人重复经过该标识时,利用该标识包含的ID进行匹配获得每一次机器人的位姿,记录每一次机器人的位姿均值μ w与协方差矩阵∑ w,并通过如上面所述的相反的坐标转换过程转换为相应的标识的位姿均值和协方差矩阵。然后,通过高斯模型,将多次得到的标识的位姿信息进行融合,估计出该标识的最优位姿估计与协方差矩阵,通过如上面所述的相反的坐标转换过程转换为相应的机器人的位姿作为机器人的初始位姿。
图5是根据本申请实施例的位姿信息融合的示意图。如图5所示,当机器人五次经过该标识时,通过SLAM导航定位获取五个不同的位姿与误差(如,协方差矩阵)。通过位姿的融合,得到右图优化后的位姿与误差(即最优位姿估计和误差)作为标识的真实位姿与误差。当机器人再次在该标识上启动时,使用该标识的最优位姿估计作为标识的初始化位姿信息,再根据机器人与该标识的相对位姿计算机器人的初始位姿与误差。
在一实施例中,假设机器人有N次经过或观察到该标识,定义第k次经过或观察到该标识的位姿满足二维高斯分布,即
Figure PCTCN2018100869-appb-000006
其中p(p k)为第k次的位姿概率分布,μ k为位姿均值,∑ k为协方差矩阵。当多次经过或观察到该标识时,该标识的最优估计即:
p(p|p 1,p 2,...,p N)
假设每次观测均为独立分布,则最优估计可转化为多个高斯分布的乘积,即:
Figure PCTCN2018100869-appb-000007
这里,基于机器人与标识的相对位姿,从标识包含的位姿信息直接确定机器人的初始位姿信息的方法已经在上面进行了描述,因此在这里便不再赘述。
因此,在上述结合标识的同时定位和地图创建导航方法中,该第一标识包含标识信息,且基于该第一标识确定定位对象的初始位姿可以为:利用该第一标识的标识信息进行匹配,记录定位对象的多个位姿信息与误差,并转换为第一标识的多个位姿信息与误差;通过高斯模型融合该第一标识的多个位姿信息以估计该第一标识的最优位姿信息和误差;基于该第一标识相对于定位对象的位姿,经由坐标转换从该第一标识的最优位姿信息直接确定该定位对象的初始 位姿信息。
这样,在第一标识包括位姿信息或者标识信息的情况下,都可以在定位对象从该定位对象的初始化区域启动时,基于该第一标识确定定位对象的初始位姿,从而解决定位对象的初始化位姿不明确的问题。
如上所述,为解决定位对象在行进过程中的定位偏差,在定位对象的行进路径上添加一个或多个标识进行定位偏差纠正。并且,相对于定位对象的初始化时标识需要包含位姿信息或者标识信息,在定位对象的行进路径上添加的标识也可以包含非标识类图形信息。
也即,在上述结合标识的SLAM导航方法中,该第二标识包含位姿信息,或者包含标识信息,或者包含非标识类图形信息。
并且,在上述结合标识的SLAM导航方法中,该第三标识包含位姿信息,或者包含标识信息,或者包含非标识类图形信息。
针对包含位姿信息的第二标识,当定位对象经过该第二标识时,可通过如上所述的方法,从第二标识的位姿信息提取码字信息,并基于第二标识相对于定位对象的位姿,从而得到定位对象的位姿信息,以对定位结果进行直接修正。
针对包含标识信息的第二标识,定位对象每次经过该第二标识时,都记录定位对象的估计位姿与误差(如,协方差矩阵),并转换为第二标识的位姿与误差(如,协方差矩阵)。经过多次融合,可以得到包含位姿信息(x b,y b,θ b)的第二标识。然后,通过得到的位姿信息和误差(如,协方差矩阵),可以直接更新定位对象的当前位姿。该方法也已经在上面进行了描述,因此不再赘述。
也就是,在上述结合标识的SLAM导航方法中,该第二标识包含位姿信息,且基于该第二标识更新定位对象的当前位姿可以为:基于该第二标识相对于定位对象的位姿,经由坐标转换从该第二标识的位姿信息直接确定该定位对象的当前位姿信息。
或者,在上述结合标识的SLAM导航方法中,该第二标识包含标识信息,且基于该第二标识更新定位对象的当前位姿可以为:利用该第二标识的标识信息进行匹配,获得并记录定位对象的多个位姿信息与误差,并转换为第二标识的多个位姿信息与误差;通过高斯模型融合该多个位姿信息以估计该第二标识的最优位姿信息和误差;基于该第二标识相对于定位对象的位姿,经由坐标转换从该第二标识的最优位姿信息直接确定该定位对象的当前位姿信息。
此外,在根据本申请实施例的结合标识的同时定位和地图创建导航方法中, 第二标识还可以是包含非标识类图形信息的标识。其中,非标识类图形信息可以不包含任何有含义的信息。例如,非标识类图形信息标识可以是有规则的图形类标识,在定位对象初始化区域可以添加多个如上述所述的有规则的图形类标识。还例如,第二标识可以是视觉标识(规则图形)或激光标识(规则物块)。图6A和图6B是根据本申请实施例的视觉标识和激光标识的示意图。如图6A所示,视觉标识为可检测的带朝向的图形,且如图6B所示,激光标识为规则物块。
在标识的检测上,对于如图6A所示的视觉标识,检测三角形与矩形相交的三个点(p 1,p 2,p 3),并以这三个点为基础,定义如图7A所示的坐标系。其中,坐标原点0如以下公式(8)表示:
Figure PCTCN2018100869-appb-000008
其中,x轴朝向为p 3-p 1,y轴朝向为p 2-O。最终确定该标识在机器人坐标系下的坐标。
对于激光标识,定义如图7B所示的坐标系。同时将标识块离散为多个激光点,当机器人的激光雷达监测到类似图案时,使用点云匹配的方案检测该标识在激光雷达坐标系下的坐标,再利用标定参数求解该标识在机器人坐标系下的坐标。图7A和图7B是根据本申请实施例的对于视觉标识和激光标识所定义的坐标系的示意图。
当机器人多次经过或观察到同一标识时,可使用机器人多次经过或观察到该标识的位姿进行标识位姿的优化。与包含ID的标识有所不同的是,不包含ID的标识在匹配时没有具体ID对应,需要使用位姿信息进行对应。如图8所示,机器人多次经过或观察到的标识与标识之间有一定的位置间隔,当机器人检测到该标识时,获得机器人的位姿,并记录机器人的位姿。对于同一标识,记录的机器人的多次位姿相近。对相近的位姿信息进行融合,即可求解出该标识的最优定位估计。当机器人再次检测到该标识时,匹配已记录标识中距离机器人最近的标识,当距离小于一定阈值时,使用该标识的位姿信息优化机器人的位姿信息;否则,创建一个新的标识。这里,图8是根据本申请实施例的对相近的标识位姿进行融合的示意图。
也就是说,当标识本身包含非标识类图形信息时,通过机器人在经过该标识时的位姿的相似性来确定是否为同一标识。或者,可以说将机器人在经过标识时的相似的位姿作为标识的相应ID。之后的过程实质上与上面所述的标识包 括ID信息的情况下相同。在一实施例中,在确定为同一标识之后,可以得到机器人每一次经过该标识的相似的位姿,并在转换后,将标识的位姿进行融合,求解出不同标识的最优定位估计。因此,在标识不包含任何信息的情况下,实质上是通过机器人经过标识时位姿的相似性来赋予标识ID,并且此后的过程就与之前说明的标识包括标识信息的情况下的相同了。
因此,在上述结合标识的SLAM导航方法中,该第二标识包含非标识类图形信息,且基于该第二标识更新定位对象的当前位姿可以为:利用定位对象经过同一标识时位姿的相似性,确定定位对象所经过的标识是否为第二标识;如果是,记录经过第二标识的定位对象的多个位姿信息与误差,并转换为第二标识的多个位姿信息和误差;通过高斯模型融合第二标识的多个位姿信息以估计该第二标识的最优位姿信息和误差;基于该第二标识相对于定位对象的位姿,经由坐标转换从该第二标识的最优位姿信息直接提取该定位对象的当前位姿信息。
并且,在根据本申请实施例的同时定位和地图创建导航方法中,在定位对象的行进路径上设置多个标识的情况下,对于每个标识都可以按照上述方式对定位对象在经过该标识时的位姿进行更新,从而减小累积偏移,实现SLAM系统的闭合检测。
另外,在一实施例中,本申请实施例的方法还包括:在所述定位对象的行进路径上的对接装置处添加第三标识;在所述定位对象经过所述第三标识的情况下,定位对象基于所述第三标识与所述对接装置完成对接操作。例如当定位对象携带搬运物品时,在定位对象经过第三标识的情况下,能够基于第三标识调整定位对象和对接装置的相对位姿,以达到精准对接目的。
在根据本申请实施例的SLAM导航方法中,为了解决诸如传送带或者机械臂的精准对接的高精度定位需求,在对接区域添加一个或多个标识以进行定位对象的精准定位。具体实现方式与上面针对机器人所述的相同,为了避免冗余便不再赘述。
此外,为了解决定位对象,比如机器人由于定位偏差过大可能在行进时无法准确经过标识的问题,可以在标识之间添加色带。这样,当机器人经过色带时,纠正机器人相对色带的角度偏差,让机器人沿色带行走,从而促进当机器人遇到标识时进行精准定位。
在一实施例中,当所述第二标识包括多个第二标识时,在所述定位对象的 行进路径上添加第二标识,包括:在定位对象的行进路径上添加多个第二标识;和,其中,在所述定位对象经过所述第二标识的情况下,基于所述第二标识更新所述定位对象的当前位姿,还包括:在定位对象经过所述多个第二标识的情况下,基于所述多个第二标识更新定位对象的当前位姿。例如,多个第二标识可以为两种标识。
图9A和图9B是根据本申请实施例的标识与色带之间的连接方式的示意图。如图9A和图9B所示,示出了典型的两种标识与色带连接方式(标识以二维码为例)。其中,图9A示出了直线连接的方式,即当两个标识在x或y方向一致时,直接使用直线色带连接两个标识。另外,图9B示出了曲线连接,即当两个标识在x和y方向均不一致,或角度不一致时,可采用平滑曲线色带连接。当然,本领域技术人员可以理解,色带的连接方式不限于这两种连接方式,只需色带曲线平滑,机器人可沿色带行走即可。
当机器人经过色带时,检测色带曲线。从曲线中搜索找到离机器人最近的点,并使用该点计算出机器人与色带之间的位置偏差Δd和角度偏差Δθ(如图10所示)。同时,检测色带的曲率C。若机器人位于色带正上方,即Δd=0,Δθ=0,则可通过跟踪色带的曲率达到沿色带行走的目的。以双轮差速机器人为例,控制参数为速度v和角速度w,开环跟踪方法为满足w=v·C。实际控制时,控制会产生偏差,且机器人初始位置与色带也有一定偏差。因此,在开环定位w=v·C的基础上,以位置偏差Δd和角度偏差Δθ为控制量进行PID控制。一个简单的方法如以下公式(9)所示:
w=v·C+λ 1·(-Δd)+λ 2Δθ        (9)
其中,λ 1和λ 2是两个参数,表示控制角度偏差和位置偏差所带来的控制权重。
通过该方法,可使机器人沿色带行走,保证行走到色带终点处时,可检测到终点处标识。当经过标识时,就可以基于上面描述的方式来校正机器人的位姿。这里,图10是示出根据本申请实施例的计算机器人与色带之间的偏差的示意图。因此,在上述结合标识的同时定位和地图创建导航方法中,在多个第二标识中的两个相邻第二标识之间包括连接该多个第二标识的辅助定位带,以在当该定位对象经过该辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得该定位对象沿着该辅助定位带行进,确保所述定位对象可以通过并识别到多个第二标识。
在另一实施例中,多个第二标识不仅可以包括两种标识,还可以包括至少两种标识。至少两种标识的设定位置可以为矩阵、矩形等不同形状,且在任意两种标识之间设定辅助定位带。通过根据本申请实施例的结合标识的SLAM导航方法,可以提高SLAM导航系统的定位精度和鲁棒性,且使其适应于需要精准定位的工业环境。
根据本申请实施例的另一方面,提供了一种结合标识的同时定位和地图创建导航装置,包括:初始化区域设置单元,设置为设置定位对象初始化区域;第一标识添加单元,设置为在该定位对象初始化区域添加第一标识,该第一标识包括位姿信息或者标识信息;定位对象启动单元,设置为使得定位对象从该定位对象初始化区域启动,并基于该第一标识确定定位对象的初始位姿;第二标识添加单元,设置为在定位对象的行进路径上添加第二标识;和位姿更新单元,设置为在定位对象经过第二标识的情况下,基于该第二标识更新定位对象的当前位姿。
图11是根据本申请实施例的结合标识的同时定位和地图创建导航装置的示意性框图。如图11所示,根据本申请实施例的结合标识的同时定位和地图创建导航装置200包括:初始化区域设置单元210,设置为设置定位对象初始化区域;第一标识添加单元220,设置为在该初始化区域设置单元210设置的定位对象初始化区域添加第一标识,该第一标识包括位姿信息或者标识信息;定位对象启动单元230,设置为使得定位对象从该初始化区域设置单元210设置的定位对象初始化区域启动,并基于该第一标识添加单元220添加的第一标识确定定位对象的初始位姿;第二标识添加单元240,设置为在定位对象的行进路径上添加第二标识;和位姿更新单元250,设置为在定位对象经过第二标识添加单元240添加的第二标识的情况下,基于该第二标识更新定位对象的当前位姿。
在上述结合标识的同时定位和地图创建导航装置中,该定位对象是同时定位和地图创建导航中行进的机器人和携带传送带或者机械臂中的移动装置。
在上述结合标识的同时定位和地图创建导航装置中,该第一标识包含位姿信息,且该定位对象启动单元设置为:基于该第一标识相对于定位对象的位姿,经由坐标转换从该第一标识的位姿信息直接提取该定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,该第一标识包含标识信息,且该定位对象启动单元设置为:利用该第一标识的标识信息匹配,记录定位对象的多个位姿信息与误差,并转换为第一标识的多个位姿信息与误差; 通过高斯模型融合该第一标识的多个位姿信息以估计该第一标识的最优位姿信息和误差;基于该第一标识相对于定位对象的位姿,经由坐标转换从该第一标识的最优位姿信息直接提取该定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,该第二标识包含位姿信息、标识信息,或者非标识类图形信息。
在上述结合标识的同时定位和地图创建导航装置中,该第二标识包含位姿信息,且该位姿更新单元设置为:基于该第二标识相对于定位对象的位姿,经由坐标转换从该第二标识的位姿信息直接提取该定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,该第二标识包含标识信息,且该位姿更新单元设置为:利用该第二标识的标识信息匹配,记录定位对象的多个位姿信息与误差,并转换为第二标识的多个位姿信息与误差;通过高斯模型融合该多个位姿信息以估计该第二标识的最优位姿信息和误差;基于该第二标识相对于定位对象的位姿,经由坐标转换从该第二标识的最优位姿信息直接提取该定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,该第二标识包含非标识类图形信息,且该位姿更新单元设置为:利用定位对象经过同一标识时位姿的相似性,确定定位对象所经过的标识是否为第二标识;记录经过第二标识的定位对象的多个位姿信息与误差,并转换为第二标识的多个位姿信息和误差;通过高斯模型融合第二标识的多个位姿信息以估计该第二标识的最优位姿信息和误差;基于该第二标识相对于定位对象的位姿,经由坐标转换从该第二标识的最优位姿信息直接提取该定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航装置中,其中,当所述第二标识包括多个第二标识时,所述第一标识添加单元,设置为在定位对象的行进路径上添加多个第二标识;且该位姿更新单元还设置为在定位对象经过该多个第二标识的情况下,基于该多个第二标识更新定位对象的当前位姿。
在上述结合标识的同时定位和地图创建导航装置中,还包括:辅助定位单元,设置为添加连接该多个第二标识中的两个相邻第二标识的辅助定位带,并在该定位对象经过该辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得该定位对象沿着该辅助定位带行进,确保所述定位对象可以通过并识别到第二标识。
在上述结合标识的同时定位和地图创建导航装置中,还包括:第三标识添 加单元,设置为在所述定位对象的行进路径上的对接装置上添加第三标识;对接单元,设置为在所述定位对象经过所述第三标识的情况下,基于所述第三标识与所述对接装置完成对接操作。
这里,本领域技术人员可以理解,在根据本申请实施例的结合标识的同时定位和地图创建导航装置中,第一标识添加单元和第二标识添加单元可以集成为用于添加标识的同一标识添加单元,并且定位对象启动单元中的部分功能也可以包括在位姿更新单元中,以用于基于标识更新定位对象的位置信息。
另外,本领域技术人员可以理解,根据本申请实施例的结合标识的同时定位和地图创建导航装置的其它细节与之前关于根据本申请实施例的结合标识的同时定位和地图创建导航方法描述的相应细节相同,为了避免冗余便不再岁数。
根据本申请的又一方面,提供了一种结合标识的同时定位和地图创建导航系统,包括:定位对象,设置为在场景中实时定位并构建地图;第一标识,包括位姿信息或者标识信息;第二标识;同时定位和地图创建装置,包括:初始化区域设置单元,设置为设置定位对象初始化区域;第一标识添加单元,设置为将该第一标识添加到该定位对象初始化区域;定位对象启动单元,设置为使得定位对象从该定位对象初始化区域启动,并基于该第一标识确定定位对象的初始位姿;第二标识添加单元,设置为将第二标识添加到该定位对象的行进路径上;和位姿更新单元,设置为在定位对象经过第二标识的情况下,基于该第二标识更新定位对象的当前位姿。
图12是根据本申请实施例的结合标识的同时定位和地图创建导航系统的示意性框图。如图12所示,根据本申请实施例的结合标识的同时定位和地图创建导航系统300包括:定位对象301,设置为在场景中实时定位并构建地图;第一标识302,包括位姿信息或者标识信息;第二标识303;同时定位和地图创建装置310,包括:初始化区域设置单元311,设置为设置定位对象初始化区域;第一标识添加单元312,设置为将该第一标识302添加到该初始化区域设置单元311设置的定位对象初始化区域;定位对象启动单元313,设置为使得定位对象301从该初始化区域设置单元311设置的定位对象初始化区域启动,并基于该第一标识添加单元312添加的第一标识302确定定位对象301的初始位姿;第二标识添加单元314,设置为将第二标识303添加到该定位对象的行进路径上;和位姿更新单元315,设置为在定位对象301经过该第二标识添加单元314添加的第二标识303的情况下,基于该第二标识303更新定位对象301的当前位姿。
在上述结合标识的同时定位和地图创建导航系统中,该定位对象是同时定位和地图创建导航中行进的机器人和携带传送带或者机械臂的移动装置。
在上述结合标识的同时定位和地图创建导航系统中,该第一标识包含位姿信息,且该定位对象启动单元还设置为:基于该第一标识相对于定位对象的位姿,经由坐标转换从该第一标识的位姿信息直接提取该定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,该第一标识包含标识信息,且该定位对象启动单元还设置为:利用该第一标识的标识信息匹配,记录定位对象的多个位姿信息与误差,并转换为第一标识的多个位姿信息与误差;通过高斯模型融合该第一标识的多个位姿信息以估计该第一标识的最优位姿信息和误差;基于该第一标识相对于定位对象的位姿,经由坐标转换从该第一标识的最优位姿信息直接提取该定位对象的初始位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,该第二标识包含位姿信息、标识信息,或者非标识类图形信息。
在上述结合标识的同时定位和地图创建导航系统中,该第二标识包含位姿信息,且该位姿更新单元还设置为:基于该第二标识相对于定位对象的位姿,经由坐标转换从该第二标识的位姿信息直接提取该定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,该第二标识包含标识信息,且该位姿更新单元还设置为:利用该第二标识的标识信息匹配,记录定位对象的多个位姿信息与误差,并转换为第二标识的多个位姿信息与误差;通过高斯模型融合该多个位姿信息以估计该第二标识的最优位姿信息和误差;基于该第二标识相对于定位对象的位姿,经由坐标转换从该第二标识的最优位姿信息直接提取该定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,该第二标识包含非标识类图形信息,且该位姿更新单元还设置为:利用定位对象经过同一标识时位姿的相似性,确定定位对象所经过的标识是否为第二标识;记录经过第二标识的定位对象的多个位姿信息与误差,并转换为第二标识的多个位姿信息和误差;通过高斯模型融合第二标识的多个位姿信息以估计该第二标识的最优位姿信息和误差;基于该第二标识相对于定位对象的位姿,经由坐标转换从该第二标识的最优位姿信息直接提取该定位对象的当前位姿信息。
在上述结合标识的同时定位和地图创建导航系统中,其中,当所述第二标 识包括多个第二标识时,所述第一标识添加单元,设置为将多个第二标识添加到定位对象的行进路径上;且该位姿更新单元还设置为在定位对象经过该多个第二标识的情况下,基于该多个第二标识更新定位对象的当前位姿。
在上述结合标识的同时定位和地图创建导航系统中,该在多个第二标识中的两个相邻第二标识之间包括辅助定位带,该同时定位和地图创建装置包括:辅助定位单元,设置为在该定位对象经过该辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得该定位对象沿着该辅助定位带行进。
在上述结合标识的同时定位和地图创建导航系统中,所述同时定位和地图创建装置,还包括:第三标识添加单元,设置为在所述定位对象的行进路径上的对接装置上添加第三标识;对接单元,设置为在所述定位对象经过所述第三标识的情况下,基于所述第三标识与所述对接装置完成对接操作。
另外,本领域技术人员可以理解,根据本申请实施例的结合标识的同时定位和地图创建导航装置的其它细节与之前关于根据本申请实施例的结合标识的同时定位和地图创建导航方法描述的相应细节相同,为了避免冗余便不再岁数。
本申请提供的结合标识的同时定位和地图创建导航方法、装置和系统可以提高同时定位和地图创建导航系统的定位精度和鲁棒性,且使其适应于需要精准定位的工业环境。
本领域的技术人员应理解,上述描述及附图中所示的本申请的实施例只作为举例而并不限制本申请。本申请的目的已经完整并有效地实现。本申请的功能及结构原理已在实施例中展示和说明,在没有背离该原理下,本申请的实施方式可以有任何变形或修改。

Claims (30)

  1. 一种结合标识的同时定位和地图创建导航方法,包括:
    设置定位对象初始化区域;其中,所述定位对象初始化区域、所述定位对象的行进路径上和所述定位对象的对接装置处中的至少之一设置有标识,所述标识包括位姿信息、标识信息和非标识类图形信息中的至少一种;
    控制定位对象执行以下操作中的至少一种:
    从所述定位对象初始化区域启动,并基于所述标识确定定位对象的初始位姿;
    在行进路径上经过所述标识的情况下,基于所述标识更新所述定位对象的当前位姿;
    在与对接装置进行对接时,基于所述标识调整所述定位对象与所述对接装置的相对位姿。
  2. 如权利要求1所述的方法,其中,所述定位对象包括同时定位和地图创建导航中行进的机器人和需要与对接装置进行对接的移动装置。
  3. 如权利要求1或2所述的方法,其中,所述对接装置包括传送带或者机械臂。
  4. 如权利要求1所述的方法,其中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含位姿信息,且,
    基于所述第一标识确定定位对象的初始位姿包括:
    基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的位姿信息直接提取所述定位对象的初始位姿信息。
  5. 如权利要求1所述的方法,其中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含标识信息,且,
    基于所述第一标识确定定位对象的初始位姿包括:
    在所述第一标识上启动时,记录所述第一标识包含的标识信息,以所述标识信息为坐标原点进行定位并制作地图;
    在所述定位对象重复多次经过所述第一标识时,利用所述第一标识的标识信息进行匹配获得所述定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第一标识的多个位姿信息和误差;
    通过高斯模型融合所述第一标识的多个位姿信息以估计所述第一标识的最优位姿信息和误差;和
    基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识 的最优位姿信息直接确定所述定位对象的初始位姿信息。
  6. 如权利要求1所述的方法,其中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含位姿信息,且,
    基于所述第二标识更新定位对象的当前位姿包括:
    基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的位姿信息直接提取所述定位对象的当前位姿信息。
  7. 如权利要求1所述的方法,其中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含标识信息,且,
    基于所述第二标识更新定位对象的当前位姿包括:
    在所述定位对象重复多次经过所述第二标识时,利用所述第二标识的标识信息进行匹配获得所述定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;
    通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和
    基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
  8. 如权利要求1所述的方法,其中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含非标识类图形信息,且,
    基于所述第二标识更新定位对象的当前位姿包括:
    利用所述定位对象经过同一标识时位姿的相似性,确定所述定位对象重复多次所经过的标识是否为第二标识;
    如果是所述第二标识,记录经过所述第二标识的所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;
    通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和
    基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
  9. 如权利要求1所述的方法,其中,所述定位对象的行进路径上设置的标识为第二标识,且设置有多个第二标识。
  10. 如权利要求9所述的方法,其中,在所述多个第二标识的两个相邻第二标识之间包括连接所述两个相邻第二标识的辅助定位带,以在所述定位对象 经过所述辅助定位带时校正所述定位对象相对于所述辅助定位带的角度偏差,以使得所述定位对象沿着所述辅助定位带行进。
  11. 一种结合标识的同时定位和地图创建导航装置,包括:
    初始化区域设置单元,设置为设置定位对象初始化区域;其中,所述定位对象初始化区域、所述定位对象的行进路径上和所述定位对象的对接装置处中的至少之一设置有标识,所述标识包括位姿信息、标识信息和非标识类图形信息中的至少一种;
    定位对象启动单元,设置为控制定位对象执行以下操作中的至少一种:从所述定位对象初始化区域启动,并基于所述标识确定定位对象的初始位姿;
    位姿更新单元,设置为在行进路径上经过标识的情况下,基于所述标识更新定位对象的当前位姿;
    对接单元,设置为在与对接装置进行对接时,基于所述标识调整所述定位对象与所述对接装置的相对位姿。
  12. 如权利要求11所述的装置,其中,所述定位对象包括同时定位和地图创建的导航中行进的机器人和需要与对接装置进行对接的移动装置。
  13. 如权利要求11或12所述的装置,其中,所述对接装置包括携带传送带或者机械臂。
  14. 如权利要求11所述的装置,其中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含位姿信息,且,
    所述定位对象启动单元还设置为:
    基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的位姿信息直接提取所述定位对象的初始位姿信息。
  15. 如权利要求11所述的装置,其中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含标识信息,且,
    所述定位对象启动单元还设置为:
    在所述第一标识上启动时,记录所述第一标识包含的标识信息,以所述标识信息为坐标原点进行定位并制作地图;
    在所述定位对象重复多次经过所述第一标识时,利用所述第一标识的标识信息进行匹配获得所述定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第一标识的多个位姿信息与误差;
    通过高斯模型融合所述第一标识的多个位姿信息以估计所述第一标识的最 优位姿信息和误差;和
    基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的最优位姿信息直接确定所述定位对象的初始位姿信息。
  16. 如权利要求11所述的装置,其中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含位姿信息,且
    所述位姿更新单元还设置为:
    基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的位姿信息直接提取所述定位对象的当前位姿信息。
  17. 如权利要求11所述的装置,其中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含标识信息,且,
    所述位姿更新单元还设置为:
    在所述定位对象重复多次经过所述第二标识时,利用所述第二标识的标识信息进行匹配获得所述定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;
    通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和
    基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
  18. 如权利要求11所述的装置,其中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含非标识类图形信息,且,
    所述位姿更新单元设置为:
    利用所述定位对象经过同一标识时位姿的相似性,确定所述定位对象重复多次所经过的标识是否为第二标识;
    如果是所述第二标识,记录经过第二标识的所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;
    通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和
    基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
  19. 如权利要求11所述的装置,其中,所述定位对象的行进路径上设置的标识为第二标识,且设置有多个第二标识。
  20. 如权利要求19所述的装置,还包括:
    辅助定位单元,设置为在多个第二标识的两个相邻第二标识之间添加连接所述两个相邻第二标识的辅助定位带,并在所述定位对象经过所述辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得所述定位对象沿着所述辅助定位带行进。
  21. 一种结合标识的同时定位和地图创建导航系统,包括:
    定位对象,设置为在场景中实时定位并构建地图;
    标识,包括位姿信息、标识信息和非标识类图形信息中的至少一种;
    同时定位和地图创建装置,包括:
    初始化区域设置单元,设置为设置定位对象初始化区域;其中,所述定位对象初始化区域、所述定位对象的行进路径上和所述定位对象的对接装置处中的至少之一设置有标识;
    定位对象启动单元,设置为控制定位对象执行以下操作中的至少一种:从所述定位对象初始化区域启动,并基于所述标识确定定位对象的初始位姿;
    位姿更新单元,设置为在行进路径上经过标识的情况下,基于所述标识更新定位对象的当前位姿;
    对接单元,设置为在与对接装置进行对接时,基于所述标识调整所述定位对象与所述对接装置的相对位姿。
  22. 如权利要求21所述的系统,其中,所述定位对象包括同时定位和地图创建的导航中行进的机器人和需要与对接装置进行对接的移动装置。
  23. 如权利要求21所述的系统,其中,所述对接装置包括携带传送带或者机械臂。
  24. 如权利要求21所述的系统,其中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含位姿信息,且,
    所述定位对象启动单元设置为:
    基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的位姿信息直接提取所述定位对象的初始位姿信息。
  25. 如权利要求21所述的系统,其中,所述定位对象初始化区域设置的标识为第一标识,所述第一标识包含标识信息,且,
    所述定位对象启动单元设置为:
    在所述第一标识上启动时,记录所述第一标识包含的标识信息,以所述标 识信息为坐标原点进行定位并制作地图;
    在所述定位对象重复多次经过所述第一标识时,利用所述第一标识的标识信息进行匹配获得所述定位对象的位姿信息,记录定位对象的多个位姿信息和误差,并转换为所述第一标识的多个位姿信息与误差;
    通过高斯模型融合所述第一标识的多个位姿信息以估计所述第一标识的最优位姿信息和误差;和
    基于所述第一标识相对于定位对象的位姿,经由坐标转换从所述第一标识的最优位姿信息直接确定所述定位对象的初始位姿信息。
  26. 如权利要求21所述的系统,其中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含位姿信息,且
    所述位姿更新单元还设置为:
    基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的位姿信息直接提取所述定位对象的当前位姿信息。
  27. 如权利要求21所述的系统,其中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含标识信息,且,
    所述位姿更新单元还设置为:
    在所述定位对象重复多次经过所述第二标识时,利用所述第二标识的标识信息进行匹配获得所述定位对象的位姿信息,记录所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息与误差;
    通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最优位姿信息和误差;和
    基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
  28. 如权利要求21所述的系统,其中,所述定位对象的行进路径上设置的标识为第二标识,所述第二标识包含非标识类图形信息,且
    所述位姿更新单元还设置为:
    利用所述定位对象经过同一标识时位姿的相似性,确定所述定位对象重复多次所经过的标识是否为第二标识;
    如果是所述第二标识,记录经过第二标识的所述定位对象的多个位姿信息和误差,并转换为所述第二标识的多个位姿信息和误差;
    通过高斯模型融合所述第二标识的多个位姿信息以估计所述第二标识的最 优位姿信息和误差;
    基于所述第二标识相对于定位对象的位姿,经由坐标转换从所述第二标识的最优位姿信息直接确定所述定位对象的当前位姿信息。
  29. 如权利要求21所述的系统,其中,所述定位对象的行进路径上设置的标识为第二标识,且设置有多个第二标识。
  30. 如权利要求29所述的系统,其中,在多个第二标识中的两个相邻第二标识之间包括辅助定位带,
    所述同时定位和地图创建装置还包括:
    辅助定位单元,设置为在所述定位对象经过所述辅助定位带时校正定位对象相对于辅助定位带的角度偏差,以使得所述定位对象沿着所述辅助定位带行进。
PCT/CN2018/100869 2017-08-16 2018-08-16 结合标识的同时定位和地图创建导航方法、装置及系统 WO2019034115A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
JP2018558212A JP6781771B2 (ja) 2017-08-16 2018-08-16 指標を使用した同時位置決め地図作成ナビゲーション方法、装置及びシステム
EP18832973.4A EP3598072A4 (en) 2017-08-16 2018-08-16 LABEL WITH METHOD, DEVICE AND SYSTEM FOR SIMULTANEOUS LOCATION AND CARD NAVIGATION
US16/315,574 US11340628B2 (en) 2017-08-16 2018-08-16 Marker-combined simultaneous localization and mapping navigation method, device and system
SG11202001328PA SG11202001328PA (en) 2017-08-16 2018-08-16 Marker-combined simultaneous localization and mapping navigation method, device and system
KR1020207007437A KR102367438B1 (ko) 2017-08-16 2018-08-16 마커를 결합한 동시 위치결정과 지도작성 내비게이션 방법, 장치 및 시스템

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710703818.3 2017-08-16
CN201710703818.3A CN107727104B (zh) 2017-08-16 2017-08-16 结合标识的同时定位和地图创建导航方法、装置及系统

Publications (1)

Publication Number Publication Date
WO2019034115A1 true WO2019034115A1 (zh) 2019-02-21

Family

ID=61205194

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/100869 WO2019034115A1 (zh) 2017-08-16 2018-08-16 结合标识的同时定位和地图创建导航方法、装置及系统

Country Status (7)

Country Link
US (1) US11340628B2 (zh)
EP (1) EP3598072A4 (zh)
JP (1) JP6781771B2 (zh)
KR (1) KR102367438B1 (zh)
CN (1) CN107727104B (zh)
SG (1) SG11202001328PA (zh)
WO (1) WO2019034115A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111123926A (zh) * 2019-12-20 2020-05-08 上海点甜农业专业合作社 一种基于二维码路标自动导航场景的搭建方法
CN112327818A (zh) * 2019-07-19 2021-02-05 杭州海康机器人技术有限公司 校正导引图形的位姿偏差的方法及行走设备
CN112596070A (zh) * 2020-12-29 2021-04-02 四叶草(苏州)智能科技有限公司 一种基于激光及视觉融合的机器人定位方法
CN114537555A (zh) * 2022-03-14 2022-05-27 恒达富士电梯有限公司 基于多机器人和多agv协同的电梯协同柔性生产线

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107727104B (zh) 2017-08-16 2019-04-30 北京极智嘉科技有限公司 结合标识的同时定位和地图创建导航方法、装置及系统
KR102396731B1 (ko) * 2018-02-27 2022-05-11 삼성전자주식회사 정밀 지도 데이터 제공 방법 및 이를 위한 시스템
CN108955668A (zh) * 2018-08-02 2018-12-07 苏州中德睿博智能科技有限公司 一种融合二维码与色带的复合导航方法、装置及系统
CN108931982A (zh) * 2018-09-04 2018-12-04 赵凯雅 用于移动机器人设备的视觉导航系统和方法
CN114999308A (zh) * 2018-11-19 2022-09-02 深圳市海柔创新科技有限公司 地图生成方法、装置、计算机可读存储介质和计算机设备
CN113409401A (zh) * 2018-11-20 2021-09-17 深圳市海柔创新科技有限公司 定位码贴设方法、装置、计算机设备和存储介质
CN109571408B (zh) * 2018-12-26 2020-03-10 北京极智嘉科技有限公司 一种机器人、库存容器的角度校准方法及存储介质
CN109813319B (zh) * 2019-03-07 2021-09-28 山东大学 一种基于slam建图的开环优化方法及系统
CN110197509B (zh) * 2019-04-30 2023-07-11 上海理工大学 一种基于彩色人工标识的相机位姿求解法
CN110108283B (zh) * 2019-05-10 2020-11-17 成都四相致新科技有限公司 一种基于多二维码视觉的高精度定位方法
KR102208708B1 (ko) * 2019-08-14 2021-01-28 한국과학기술연구원 공통 좌표계 기반의 가상공간에서 가상 컨텐츠 제공 방법 및 장치
DE102019128927A1 (de) * 2019-10-25 2021-04-29 Vorwerk & Co. Interholding Gmbh Verfahren zur Erstellung einer Umgebungskarte für ein sich selbsttätig fortbewegendes Bodenbearbeitungsgerät, sowie System aus zwei sich selbsttätig fortbewegenden Bodenbearbeitungsgeräten
CN111830969B (zh) * 2020-06-12 2024-03-26 北京布科思科技有限公司 一种基于反光板及二维码的融合对接方法
CN112153568B (zh) * 2020-08-28 2022-08-30 汉海信息技术(上海)有限公司 基于业务场景的Wi-Fi识别、绑定方法、装置及设备
CN112338910A (zh) * 2020-09-22 2021-02-09 北京无线体育俱乐部有限公司 空间地图确定方法、机器人、存储介质及系统
CN112130567A (zh) * 2020-09-22 2020-12-25 广州小鹏自动驾驶科技有限公司 一种数据处理方法和装置
CN113310484B (zh) * 2021-05-28 2022-06-24 杭州艾米机器人有限公司 一种移动机器人定位方法和系统
CN113515126B (zh) * 2021-07-12 2024-05-10 北京经纬恒润科技股份有限公司 一种车辆定位方法及装置
CN113733043B (zh) * 2021-09-24 2024-06-18 深圳先进技术研究院 一种自动化递送机器人及其自动驱动方法
CN114779755A (zh) * 2022-03-01 2022-07-22 国以贤智能科技(上海)股份有限公司 基于混合导航方式的单舵轮叉车运动控制方法及系统
DE102022206041A1 (de) 2022-06-15 2023-12-21 Robert Bosch Gesellschaft mit beschränkter Haftung Verfahren zum Bestimmen von Objekten in einer Umgebung für SLAM
KR102585353B1 (ko) * 2022-11-10 2023-10-06 주식회사 클로봇 무인 이동체, 라이다 및 브이 형상의 마커를 활용한 무인 이동체의 도킹 제어 방법 및 프로그램

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102402225A (zh) * 2011-11-23 2012-04-04 中国科学院自动化研究所 一种实现移动机器人同时定位与地图构建的方法
CN103294059A (zh) * 2013-05-21 2013-09-11 无锡普智联科高新技术有限公司 基于混合导航带的移动机器人定位系统及其方法
US20140022281A1 (en) * 2012-07-18 2014-01-23 The Boeing Company Projecting airplane location specific maintenance history using optical reference points
CN103869814A (zh) * 2012-12-17 2014-06-18 联想(北京)有限公司 一种终端定位和导航方法以及可移动的终端
CN104062973A (zh) * 2014-06-23 2014-09-24 西北工业大学 一种基于图像标志物识别的移动机器人slam方法
CN104142683A (zh) * 2013-11-15 2014-11-12 上海快仓智能科技有限公司 基于二维码定位的自动引导车导航方法
CN104848858A (zh) * 2015-06-01 2015-08-19 北京极智嘉科技有限公司 二维码以及用于机器人的视觉-惯性组合导航系统及方法
CN106527432A (zh) * 2016-11-04 2017-03-22 浙江大学 基于模糊算法和二维码自矫正的室内移动机器人协作系统
CN106708051A (zh) * 2017-01-10 2017-05-24 上海极络智能科技有限公司 基于二维码的导航系统和方法、导航标记物及导航控制器
CN106803261A (zh) * 2015-11-20 2017-06-06 沈阳新松机器人自动化股份有限公司 机器人相对位姿估计方法
CN106989746A (zh) * 2017-03-27 2017-07-28 远形时空科技(北京)有限公司 导航方法及导航装置
CN107727104A (zh) * 2017-08-16 2018-02-23 北京极智嘉科技有限公司 结合标识的同时定位和地图创建导航方法、装置及系统

Family Cites Families (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006065563A2 (en) * 2004-12-14 2006-06-22 Sky-Trax Incorporated Method and apparatus for determining position and rotational orientation of an object
US7693654B1 (en) * 2005-11-23 2010-04-06 ActivMedia Robotics/MobileRobots Method for mapping spaces with respect to a universal uniform spatial reference
US7693757B2 (en) * 2006-09-21 2010-04-06 International Business Machines Corporation System and method for performing inventory using a mobile inventory robot
US8874261B2 (en) * 2007-07-25 2014-10-28 Deere & Company Method and system for controlling a mobile robot
JP5168134B2 (ja) * 2008-12-26 2013-03-21 富士通株式会社 環境地図生成プログラム、環境地図生成方法及び移動ロボット
DE102010012187B4 (de) * 2010-03-19 2020-12-31 Sew-Eurodrive Gmbh & Co Kg Verfahren zur Bestimmung der Position zumindest eines ersten und eines zweiten Fahrzeuges innerhalb einer Anlage
WO2011160677A1 (en) * 2010-06-22 2011-12-29 Tomtom International B.V. Navigation device & method
DE112011102132T5 (de) * 2010-06-25 2013-05-23 Trimble Navigation Ltd. Verfahren und Vorrichtung für eine imagebasierte Positionierung
JP2012064131A (ja) * 2010-09-17 2012-03-29 Tokyo Institute Of Technology 地図生成装置、地図生成方法、移動体の移動方法、及びロボット装置
US9286810B2 (en) * 2010-09-24 2016-03-15 Irobot Corporation Systems and methods for VSLAM optimization
US9538892B2 (en) * 2012-10-05 2017-01-10 Irobot Corporation Robot management systems for determining docking station pose including mobile robots and methods using same
CN103631264A (zh) * 2013-12-04 2014-03-12 苏州大学张家港工业技术研究院 一种同时定位与地图创建方法及装置
EP2916189B1 (en) * 2014-03-06 2019-05-08 Hexagon Technology Center GmbH Quality assured manufacturing
CN103968846B (zh) * 2014-03-31 2017-02-08 小米科技有限责任公司 定位导航方法和装置
CN106662631A (zh) * 2015-01-22 2017-05-10 广州艾若博机器人科技有限公司 基于rfid的定位与地图构建方法及其装置
WO2016115714A1 (zh) * 2015-01-22 2016-07-28 江玉结 基于色块标签的定位与地图构建方法及其装置
US9758305B2 (en) * 2015-07-31 2017-09-12 Locus Robotics Corp. Robotic navigation utilizing semantic mapping
JP6849330B2 (ja) * 2015-08-28 2021-03-24 パナソニック インテレクチュアル プロパティ コーポレーション オブ アメリカPanasonic Intellectual Property Corporation of America 地図生成方法、自己位置推定方法、ロボットシステム、およびロボット
US9908702B2 (en) * 2016-02-05 2018-03-06 Invia Robotics, Inc. Robotic navigation and mapping
CN106643759A (zh) * 2016-09-29 2017-05-10 王德龙 一种基于电子导航地图的实时定位导航信息共享方法
CN106323294B (zh) 2016-11-04 2023-06-09 新疆大学 变电站巡检机器人定位方法及定位装置

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102402225A (zh) * 2011-11-23 2012-04-04 中国科学院自动化研究所 一种实现移动机器人同时定位与地图构建的方法
US20140022281A1 (en) * 2012-07-18 2014-01-23 The Boeing Company Projecting airplane location specific maintenance history using optical reference points
CN103869814A (zh) * 2012-12-17 2014-06-18 联想(北京)有限公司 一种终端定位和导航方法以及可移动的终端
CN103294059A (zh) * 2013-05-21 2013-09-11 无锡普智联科高新技术有限公司 基于混合导航带的移动机器人定位系统及其方法
CN104142683A (zh) * 2013-11-15 2014-11-12 上海快仓智能科技有限公司 基于二维码定位的自动引导车导航方法
CN104062973A (zh) * 2014-06-23 2014-09-24 西北工业大学 一种基于图像标志物识别的移动机器人slam方法
CN104848858A (zh) * 2015-06-01 2015-08-19 北京极智嘉科技有限公司 二维码以及用于机器人的视觉-惯性组合导航系统及方法
CN106803261A (zh) * 2015-11-20 2017-06-06 沈阳新松机器人自动化股份有限公司 机器人相对位姿估计方法
CN106527432A (zh) * 2016-11-04 2017-03-22 浙江大学 基于模糊算法和二维码自矫正的室内移动机器人协作系统
CN106708051A (zh) * 2017-01-10 2017-05-24 上海极络智能科技有限公司 基于二维码的导航系统和方法、导航标记物及导航控制器
CN106989746A (zh) * 2017-03-27 2017-07-28 远形时空科技(北京)有限公司 导航方法及导航装置
CN107727104A (zh) * 2017-08-16 2018-02-23 北京极智嘉科技有限公司 结合标识的同时定位和地图创建导航方法、装置及系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3598072A4

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112327818A (zh) * 2019-07-19 2021-02-05 杭州海康机器人技术有限公司 校正导引图形的位姿偏差的方法及行走设备
CN112327818B (zh) * 2019-07-19 2023-06-23 杭州海康机器人股份有限公司 校正导引图形的位姿偏差的方法及行走设备
CN111123926A (zh) * 2019-12-20 2020-05-08 上海点甜农业专业合作社 一种基于二维码路标自动导航场景的搭建方法
CN112596070A (zh) * 2020-12-29 2021-04-02 四叶草(苏州)智能科技有限公司 一种基于激光及视觉融合的机器人定位方法
CN112596070B (zh) * 2020-12-29 2024-04-19 四叶草(苏州)智能科技有限公司 一种基于激光及视觉融合的机器人定位方法
CN114537555A (zh) * 2022-03-14 2022-05-27 恒达富士电梯有限公司 基于多机器人和多agv协同的电梯协同柔性生产线
CN114537555B (zh) * 2022-03-14 2023-09-01 恒达富士电梯有限公司 基于多机器人和多agv协同的电梯协同柔性生产线

Also Published As

Publication number Publication date
SG11202001328PA (en) 2020-03-30
US11340628B2 (en) 2022-05-24
JP2019537077A (ja) 2019-12-19
EP3598072A1 (en) 2020-01-22
US20210333797A1 (en) 2021-10-28
CN107727104A (zh) 2018-02-23
EP3598072A4 (en) 2020-05-06
JP6781771B2 (ja) 2020-11-04
KR102367438B1 (ko) 2022-02-24
KR20200041355A (ko) 2020-04-21
CN107727104B (zh) 2019-04-30

Similar Documents

Publication Publication Date Title
WO2019034115A1 (zh) 结合标识的同时定位和地图创建导航方法、装置及系统
Yang et al. Cubeslam: Monocular 3-d object slam
CN112985416B (zh) 激光与视觉信息融合的鲁棒定位和建图方法及系统
Wang et al. Visual servoing trajectory tracking of nonholonomic mobile robots without direct position measurement
Nieto et al. Recursive scan-matching SLAM
CN108180917B (zh) 一种基于位姿图优化的顶标地图构建方法
Ranganathan et al. Light-weight localization for vehicles using road markings
WO2023104207A1 (zh) 一种协同三维建图方法及系统
CN105912971A (zh) 一种用于agv导航的规则图形码码阵及其读码方法
Bischoff et al. Fusing vision and odometry for accurate indoor robot localization
CN108225327A (zh) 一种顶标地图的构建与定位方法
WO2022078513A1 (zh) 定位方法、装置、自移动设备和存储介质
WO2022078488A1 (zh) 定位方法、装置、自移动设备和存储介质
US20220390954A1 (en) Topology Processing for Waypoint-based Navigation Maps
CN112677148A (zh) 机器人系统
Ruchanurucks et al. Automatic landing assist system using IMU+ P n P for robust positioning of fixed-wing UAVs
CN112101160A (zh) 一种面向自动驾驶场景的双目语义slam方法
Lin et al. A sparse visual odometry technique based on pose adjustment with keyframe matching
JP6410231B2 (ja) 位置合わせ装置、位置合わせ方法及び位置合わせ用コンピュータプログラム
Huang et al. Joint ego-motion estimation using a laser scanner and a monocular camera through relative orientation estimation and 1-DoF ICP
Yan et al. Image-based visual servoing system for components alignment using point and line features
Fang et al. A motion tracking method by combining the IMU and camera in mobile devices
WO2019113859A1 (zh) 基于机器视觉的虚拟墙构建方法及装置、地图构建方法、可移动电子设备
US20230079899A1 (en) Determination of an absolute initial position of a vehicle
Strasdat et al. Multi-cue localization for soccer playing humanoid robots

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2018558212

Country of ref document: JP

Kind code of ref document: A

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18832973

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20207007437

Country of ref document: KR

Kind code of ref document: A