WO2021129345A1 - 场景地图建立方法、设备及存储介质 - Google Patents
场景地图建立方法、设备及存储介质 Download PDFInfo
- Publication number
- WO2021129345A1 WO2021129345A1 PCT/CN2020/133735 CN2020133735W WO2021129345A1 WO 2021129345 A1 WO2021129345 A1 WO 2021129345A1 CN 2020133735 W CN2020133735 W CN 2020133735W WO 2021129345 A1 WO2021129345 A1 WO 2021129345A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- robot
- preset picture
- scene
- coordinates
- scene map
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/28—Navigation; 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
- G01C21/30—Map- or contour-matching
- G01C21/32—Structuring or formatting of map data
Definitions
- the present invention relates to the technical field of robots, in particular to a method, equipment and storage medium for establishing a scene map.
- the existing processing method is easy to accumulate errors, resulting in low mapping accuracy or even mapping. error.
- the existing processing method is to add a camera to the robot, through the VO (Visual Odometry, visual odometry) method, and combined with laser data assistance The method of mapping.
- the above method still has the problem of cumulative error, because the closed-loop method of pure laser has closed-loop errors, while the closed-loop method using visual local features such as SIFT (Scale-invariant feature transform) is difficult to deal with some repetitions. High scene. Therefore, what kind of closed-loop method is used to eliminate the accumulated error and improve the accuracy of mapping has become one of the problems to be solved urgently.
- the present invention provides a method, equipment and storage medium for establishing a scene map, and aims to provide a technology for assisting the establishment of a scene map by using pictures with non-repetitive content, and improve the accuracy of establishing a scene map.
- the present invention provides a method for establishing a scene map, and the method for establishing a scene map includes:
- the robot is used to collect the current image corresponding to the scene of the map to be established, the coordinates corresponding to the four corners of the preset picture existing in the collected current image are recognized, and the current pose of the robot is recorded at the same time.
- the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system are acquired.
- the current coordinates of the robot are calculated according to the acquired three-dimensional coordinates of the preset picture, and the current position information of the robot is obtained according to the current coordinates of the robot.
- the acquired current position information of the robot is used as closed-loop information for establishing a scene map, and the scene map is established by referring to the closed-loop information and the three-dimensional coordinates corresponding to the four corners of the acquired preset picture.
- the present invention also provides a scene map establishment device, the scene map establishment device includes:
- the image acquisition module is used to collect the current image corresponding to the scene of the map to be built by the robot, identify the coordinates corresponding to the four corners of the preset picture existing in the collected current image, and record the current pose of the robot at the same time .
- the coordinate conversion module is configured to obtain the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system according to the identified coordinates corresponding to the four corners of the preset picture.
- the position acquisition module is used to calculate the current coordinates of the robot according to the acquired three-dimensional coordinates of the preset picture when the robot acquires the preset picture again, and obtain the current position information of the robot according to the current coordinates of the robot.
- the map creation module is configured to use the acquired current position information of the robot as closed-loop information for creating a scene map, and to create a scene by referring to the closed-loop information and the three-dimensional coordinates corresponding to the four corners of the acquired preset picture map.
- the present invention also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a scene map creation program that can be run on the processor, and the scene map creation program When the program is run by the processor, the method for establishing a scene map is executed.
- the present invention also provides a computer storage medium with a scene map creation program stored on the storage medium, and the scene map creation program can be executed by one or more processors to realize the Steps of the scene map creation method.
- the method, equipment and storage medium for establishing a scene map of the present invention can achieve the following beneficial effects:
- the robot uses the robot to collect the current image corresponding to the scene of the map to be built, identify the coordinates corresponding to the four corners of the preset picture existing in the collected current image, and record the current pose of the robot at the same time; According to the coordinates corresponding to the four corners of the preset picture, the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system are obtained; when the robot collects the preset picture again, it will be based on the obtained preset picture.
- the three-dimensional coordinates of the picture, the current coordinates of the robot are calculated, and the current position information of the robot is obtained according to the current coordinates of the robot; the current position information of the robot obtained is used as the closed-loop information for establishing the scene map, referring to the closed-loop information and
- the obtained three-dimensional coordinates corresponding to the four corners of the preset picture are used to establish a scene map; the purpose of assisting the establishment of the scene map by identifying pictures whose contents in the scene to be established is not repeated is achieved, and the accuracy of the closed loop is improved. This improves the accuracy of the scene map creation; because each preset image has unique content, it avoids the problem of closed loop errors caused by the high scene repetition rate, and improves the accuracy of the scene map creation while also improving the creation of the map. effectiveness.
- FIG. 1 is a schematic flowchart of an implementation manner of a method for establishing a scene map of the present invention.
- Fig. 2 is a schematic diagram of functional modules of an embodiment of the device for establishing a scene map of the present invention.
- FIG. 3 is a schematic diagram of the internal structure of an embodiment of the electronic device of the present invention.
- the present invention provides a method, equipment and storage medium for establishing a scene map, which can improve the accuracy of the closed loop by identifying pictures whose contents are not repeated in the scene of the map to be established, thereby improving the accuracy of the scene map creation; Unique, thus avoiding the problem of closed loop errors caused by high scene repetition rate, improving the accuracy of scene map creation, and also improving the efficiency of map creation.
- Fig. 1 is a schematic flow chart of an embodiment of a method for establishing a scene map of the present invention
- a method for establishing a scene map of the present invention can be implemented as steps S10-S40 as described below:
- Step S10 Use the robot to collect the current image corresponding to the scene of the map to be established, identify the coordinates corresponding to the four corners of the preset picture existing in the collected current image, and record the current pose of the robot at the same time.
- a robot is used to collect the current image corresponding to the scene of the map to be established, and to identify whether there is a preset picture in the collected current image.
- the preset picture is pre-configured before performing step S10 of the embodiment described in FIG. 1 for the first time; that is, before running the scene map establishment method in FIG. 1 described in the present invention for the first time, the following steps need to be performed:
- the preset picture includes: a two-dimensional code and/or a fixed image.
- the above-mentioned preset pictures only need to be configured when the scene map creation method is run for the first time, and the configured preset pictures can be used directly when the method steps S10-S40 described in the embodiment of FIG. 1 are run subsequently.
- the coordinates corresponding to the four corners of the aforementioned preset picture existing in the current image are collected, and the current pose of the robot at this time is recorded at the same time.
- the coordinates corresponding to the four corners of the preset picture collected at this time are the coordinates in the coordinate system corresponding to the current image.
- the image acquisition sensor configured by the robot itself when the robot collects images in the scene where the map is to be established, the image acquisition sensor configured by the robot itself can be used; the image acquisition sensor includes, but is not limited to: various types of cameras and cameras, such as ordinary cameras , Depth camera, etc.
- the embodiment of the present invention does not limit the specific number and specific types of image collection sensors used by the robot to collect images in the scene; in actual use, it can be configured according to specific application scenarios.
- the preset pictures that are configured in the scene where the map is to be established are recognized with non-repetitive content, and more accurate and non-repetitive position information of the robot can be obtained.
- Step S20 Obtain the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system according to the identified coordinates corresponding to the four corners of the preset picture.
- the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system can be obtained.
- a multi-view geometric algorithm can be used to calculate the three-dimensional coordinates of the four corners of the aforementioned preset picture in the scene map coordinate system.
- the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system can also be obtained in combination with the type of image acquisition sensor configured by the robot itself.
- the depth camera loaded by the robot can be used to collect the depth picture corresponding to the preset picture.
- the preset picture can be obtained. Set the three-dimensional coordinates of the four corners of the picture in the scene map coordinate system.
- Step S30 When the robot collects the preset picture again, calculate the current coordinates of the robot according to the acquired three-dimensional coordinates of the preset picture, and obtain the current position information of the robot according to the current coordinates of the robot.
- Step S40 using the acquired current position information of the robot as closed-loop information for establishing a scene map, and referring to the closed-loop information and the three-dimensional coordinates corresponding to the four corners of the acquired preset picture to establish the scene map.
- the robot detects and collects the preset picture again, according to the acquired three-dimensional coordinates of the preset picture in the scene map, the current coordinates of the robot in the three-dimensional coordinate system corresponding to the scene map can be calculated.
- the parameters corresponding to the image acquisition sensor configured by the robot may be used in combination with The acquired three-dimensional coordinates of the preset picture are calculated by calculating the position information of the image acquisition sensor in the scene map coordinate system, and then the current coordinates of the robot are calculated.
- the position of the image acquisition sensor in the scene map coordinate system is calculated by using the acquired three-dimensional coordinates of the preset picture Information, the current coordinates of the robot are calculated, which can be implemented as follows:
- T in the relational expression (2) is a transformation matrix from the image acquisition sensor to the center of the robot, and the transformation matrix T can be obtained through preset calibration of the image acquisition sensor.
- the current position information corresponding to the robot can be obtained.
- the current position information of the robot is used to provide closed-loop information for the establishment of the scene map.
- the robot By using the robot to collect the corresponding preset pictures, the preset pictures that are not repeated in the scene are recognized, and the accuracy of the closed-loop information is improved; The image content is unique, thus avoiding the problem of closed loop errors caused by high scene repetition rate. Therefore, using the acquired current position information of the robot as closed-loop information for establishing a scene map, and according to the acquired three-dimensional coordinates corresponding to the four corners of the preset picture, a scene map with higher accuracy can be created.
- the scene map establishment method of the present invention uses a robot to collect the current image corresponding to the scene of the map to be established, recognizes the coordinates corresponding to the four corners of the preset picture existing in the collected current image, and records the current position of the robot at the same time.
- Pose According to the identified coordinates corresponding to the four corners of the preset picture, obtain the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system; when the robot collects the preset picture again, According to the acquired three-dimensional coordinates of the preset picture, the current coordinates of the robot are calculated, and the current position information of the robot is obtained according to the current coordinates of the robot; the acquired current position information of the robot is used as the closed-loop information for establishing the scene map Establishing a scene map by referring to the closed-loop information and the three-dimensional coordinates corresponding to the four corners of the acquired preset picture; it is achieved that the scene map is assisted in the establishment of the scene map by recognizing pictures whose contents in the scene to be established are not duplicated.
- the purpose is to improve the accuracy of the closed loop, thereby improving the accuracy of the scene map; because each preset picture has the unique content, it avoids the problem of closed loop errors caused by the high scene repetition rate, and improves the accuracy of the scene map. At the same time, the efficiency of mapping is also improved.
- an embodiment of the present invention also provides an apparatus for establishing a scene map; as shown in FIG. 2, FIG. 2 is an implementation manner of the apparatus for establishing a scene map of the present invention.
- the schematic diagram of the functional modules of the present invention, Figure 2 only describes the scene map establishment device of the present invention from the function; the scene map establishment device of the present invention functionally includes:
- the image acquisition module 100 is used to collect the current image corresponding to the scene of the map to be built by the robot, identify the coordinates corresponding to the four corners of the preset picture existing in the collected current image, and record the current position of the robot at the same time. posture.
- the coordinate conversion module 200 is configured to obtain the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system according to the identified coordinates corresponding to the four corners of the preset picture.
- the position acquiring module 300 is used to calculate the current coordinates of the robot according to the acquired three-dimensional coordinates of the preset picture when the robot acquires the preset picture again, and obtain the current position information of the robot according to the current coordinates of the robot.
- the map creation module 400 is configured to use the acquired current position information of the robot as closed-loop information for establishing a scene map, referring to the closed-loop information and the three-dimensional coordinates corresponding to the four corners of the acquired preset picture to establish Scene map.
- the device for establishing a scene map further includes a configuration module; the configuration module is configured to:
- the preset picture includes: a two-dimensional code and/or a fixed image.
- the image acquisition module 100 is used to:
- the image acquisition sensor configured by the robot itself is used to collect the current image corresponding to the scene to be created on the map.
- the image acquisition sensor includes a camera and/or a depth camera.
- the coordinate conversion module 200 is used to:
- a multi-view geometric algorithm is used to calculate the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system.
- the coordinate conversion module 200 is used to:
- Use the depth camera configured by the robot to collect the depth picture corresponding to the preset picture, and use the collected preset picture and cooperate with the depth picture to obtain the four corners of the preset picture in the scene map coordinate system The three-dimensional coordinates.
- the location acquisition module 300 is used to:
- the robot When the robot detects and collects a preset picture again, it uses the acquired three-dimensional coordinates of the preset picture according to the parameters corresponding to the image acquisition sensor configured by the robot itself, and calculates that the image acquisition sensor is in the scene map coordinate system Under the position information, the current coordinates of the robot are calculated.
- the location acquisition module 300 is used to:
- T in the relational expression (2) is a transformation matrix from the image acquisition sensor to the center of the robot, and the transformation matrix T can be obtained through preset calibration of the image acquisition sensor.
- the scene map establishment device of the present invention uses a robot to collect the current image corresponding to the scene of the map to be established, recognizes the coordinates corresponding to the four corners of the preset picture existing in the collected current image, and records the current position of the robot at the same time.
- Pose According to the identified coordinates corresponding to the four corners of the preset picture, obtain the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system; when the robot collects the preset picture again, According to the acquired three-dimensional coordinates of the preset picture, the current coordinates of the robot are calculated, and the current position information of the robot is obtained according to the current coordinates of the robot; the acquired current position information of the robot is used as the closed-loop information for establishing the scene map Establishing a scene map by referring to the closed-loop information and the three-dimensional coordinates corresponding to the four corners of the acquired preset picture; it is achieved that the scene map is assisted in the establishment of the scene map by recognizing pictures whose contents in the scene to be established are not duplicated.
- the purpose is to improve the accuracy of the closed loop, thereby improving the accuracy of the scene map; because each preset picture has the unique content, it avoids the problem of closed loop errors caused by the high scene repetition rate, and improves the accuracy of the scene map. At the same time, the efficiency of mapping is also improved.
- FIG. 3 is a schematic diagram of the internal structure of an embodiment of the electronic device of the present invention.
- the electronic device 1 may be a PC (Personal Computer, personal computer), or a terminal device such as a smart phone, a tablet computer, or a portable computer.
- the electronic device 1 at least includes a memory 11, a processor 12, a communication bus 13, and a network interface 14.
- the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, and the like.
- the memory 11 may be an internal storage unit of the electronic device 1 in some embodiments, such as a hard disk of the electronic device 1.
- the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in hard disk equipped on the electronic device 1, a smart memory card (Smart Media Card, SMC), Secure Digital (Secure Digital, SD) card, flash card (Flash Card), etc.
- the memory 11 may also include both an internal storage unit of the electronic device 1 and an external storage device.
- the memory 11 can be used not only to store application software and various data installed in the electronic device 1, such as the code of the scene map creation program 01, etc., but also to temporarily store data that has been output or will be output.
- the processor 12 may be a central processing unit (Central Processing Unit) in some embodiments.
- Central Processing Unit CPU
- controller microcontroller
- microprocessor or other data processing chip, used to run the program code or processing data stored in the memory 11, for example, execute the scene map creation program 01 and so on.
- the communication bus 13 is used to realize the connection and communication between these components.
- the network interface 14 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface), and is usually used to establish a communication connection between the electronic device 1 and other electronic devices.
- a standard wired interface and a wireless interface such as a WI-FI interface
- the electronic device 1 may further include a user interface.
- the user interface may include a display (Display) and an input unit such as a keyboard (Keyboard).
- the optional user interface may also include a standard wired interface and a wireless interface.
- the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, etc.
- the display can also be appropriately called a display screen or a display unit, which is used to display the information processed in the electronic device 1 and to display a visualized user interface.
- FIG. 3 only shows the electronic device 1 with components 11-14 and the scene map creation program 01.
- FIG. 2 does not constitute a limitation on the electronic device 1, and may include ratios Fewer or more parts are shown, or some parts are combined, or different parts are arranged.
- a scene map creation program 01 is stored in the memory 11; the scene map creation program 01 stored in the memory 11 can be stored in all When running on the processor 12, the scene map creation program 01 is executed by the processor 12 to implement the following steps:
- the robot is used to collect the current image corresponding to the scene of the map to be established, the coordinates corresponding to the four corners of the preset picture existing in the collected current image are recognized, and the current pose of the robot is recorded at the same time.
- the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system are acquired.
- the current coordinates of the robot are calculated according to the acquired three-dimensional coordinates of the preset picture, and the current position information of the robot is obtained according to the current coordinates of the robot.
- the acquired current position information of the robot is used as closed-loop information for establishing a scene map, and the scene map is established by referring to the closed-loop information and the three-dimensional coordinates corresponding to the four corners of the acquired preset picture.
- the scene map creation program 01 may also be run by the processor 12 to, before the step of "using a robot to collect the current image corresponding to the scene of the map to be created", it further includes:
- the preset picture includes: a two-dimensional code and/or a fixed image.
- the scene map creation program 01 can also be run by the processor 12 to collect the current image corresponding to the scene of the map to be created by the robot, including:
- the image acquisition sensor configured by the robot itself is used to collect the current image corresponding to the scene to be created on the map.
- the image acquisition sensor includes a camera and/or a depth camera.
- the scene map creation program 01 can also be run by the processor 12 to obtain four corners of the preset picture according to the identified coordinates corresponding to the four corners of the preset picture.
- the three-dimensional coordinates of each corner in the scene map coordinate system including:
- a multi-view geometric algorithm is used to calculate the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system.
- the scene map creation program 01 can also be run by the processor 12 to obtain four corners of the preset picture according to the identified coordinates corresponding to the four corners of the preset picture.
- the three-dimensional coordinates of each corner in the scene map coordinate system including:
- Use the depth camera configured by the robot to collect the depth picture corresponding to the preset picture, and use the collected preset picture and cooperate with the depth picture to obtain the four corners of the preset picture in the scene map coordinate system The three-dimensional coordinates.
- the scene map creation program 01 can also be run by the processor 12, so that when the robot collects a preset picture again, calculate the robot’s three-dimensional coordinates according to the acquired three-dimensional coordinates of the preset picture.
- Current coordinates including:
- the robot When the robot detects and collects a preset picture again, it uses the acquired three-dimensional coordinates of the preset picture according to the parameters corresponding to the image acquisition sensor configured by the robot itself, and calculates that the image acquisition sensor is in the scene map coordinate system. Under the position information, the current coordinates of the robot are calculated.
- the scene map creation program 01 can also be run by the processor 12 to use the acquired three-dimensional coordinates of the preset picture according to the parameters corresponding to the image acquisition sensor configured by the robot itself, and through calculation The position information of the image acquisition sensor in the scene map coordinate system is calculated to obtain the current coordinates of the robot, including:
- T in the relational expression (2) is a transformation matrix from the image acquisition sensor to the center of the robot, and the transformation matrix T can be obtained through preset calibration of the image acquisition sensor.
- the electronic device of the present invention uses a robot to collect the current image corresponding to the scene of the map to be established, recognizes the coordinates corresponding to the four corners of the preset picture existing in the collected current image, and records the current pose of the robot at the same time; According to the identified coordinates corresponding to the four corners of the preset picture, the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system are obtained; when the robot collects the preset picture again, according to the obtained According to the three-dimensional coordinates of the preset picture, the current coordinates of the robot are calculated, and the current position information of the robot is obtained according to the current coordinates of the robot; the current position information of the robot obtained is used as the closed-loop information for establishing the scene map, refer to The closed-loop information and the acquired three-dimensional coordinates corresponding to the four corners of the preset picture establish a scene map; the purpose of assisting the establishment of the scene map is achieved by identifying pictures whose contents in the scene to be established are not duplicated.
- the accuracy of the closed loop is improved, thereby improving the accuracy of the scene map creation; since each preset picture has the unique content, it avoids the problem of closed loop errors caused by the high scene repetition rate. While improving the accuracy of the scene map creation, Also improves the efficiency of mapping.
- the embodiment of the present invention also provides a computer storage medium, the computer storage medium stores a scene map creation program, and the scene map creation program can be executed by one or more processors to realize the following operations:
- the robot is used to collect the current image corresponding to the scene of the map to be established, the coordinates corresponding to the four corners of the preset picture existing in the collected current image are recognized, and the current pose of the robot is recorded at the same time.
- the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system are acquired.
- the current coordinates of the robot are calculated according to the acquired three-dimensional coordinates of the preset picture, and the current position information of the robot is obtained according to the current coordinates of the robot.
- the acquired current position information of the robot is used as closed-loop information for establishing a scene map, and the scene map is established by referring to the closed-loop information and the three-dimensional coordinates corresponding to the four corners of the acquired preset picture.
- the specific implementation of the computer-readable storage medium of the present invention is basically the same as the implementation principles of the foregoing corresponding embodiments of the scene map establishment method, device, and electronic equipment, and will not be repeated here.
- the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware.
- a robot is used to collect the current image corresponding to the scene of the map to be created, to identify the coordinates corresponding to the four corners of the preset picture existing in the collected current image, and to record the current pose of the robot at the same time;
- the recognized coordinates of the four corners of the preset picture correspond to the three-dimensional coordinates of the four corners of the preset picture in the scene map coordinate system; when the robot collects the preset picture again,
- the three-dimensional coordinates of the preset picture are calculated to obtain the current coordinates of the robot, and the current position information of the robot is obtained according to the current coordinates of the robot; the obtained current position information of the robot is used as the closed-loop information for establishing the scene map, referring to the current position information of the robot.
- the closed-loop information and the three-dimensional coordinates corresponding to the four corners of the acquired preset picture are used to establish a scene map; the purpose of assisting the establishment of the scene map is achieved by identifying pictures whose contents in the scene to be established are not duplicated.
- the accuracy of the closed loop is improved, and the mapping accuracy of the scene map is improved; since each preset image has the unique content, the problem of closed loop errors caused by the high scene repetition rate is avoided, and the accuracy of the scene map mapping is improved at the same time. Improve the efficiency of mapping. Therefore, it has industrial applicability.
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Processing Or Creating Images (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Analysis (AREA)
Abstract
一种场景地图建立方法、电子设备(1)及存储介质,利用机器人采集待建立地图的场景中对应的当前图像,识别采集的当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿(S10);根据识别出的预设图片的四个角分别对应的坐标,获取预设图片的四个角在场景地图坐标系下的三维坐标(S20);当机器人再次采集到预设图片时,根据获取的预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息(S30);将获取的机器人的当前位置信息作为建立场景地图的闭环信息,参照闭环信息以及获取的预设图片的四个角对应的三维坐标,建立场景地图(S40);提高了闭环的准确性和场景地图的建图精度。
Description
本发明涉及机器人技术领域,特别涉及一种场景地图建立方法、设备及存储介质。
目前,现有的利用激光导航机器人建立场景地图的方法,基本上大多数是采用单独使用激光传感器的方式来建立场景地图,这种处理方式容易积累误差,从而导致建图精度不高甚至建图错误。为解决单独使用激光传感器导致的建图精度不高甚至建图错误的问题,现有的处理方式是:在机器人上增加相机,通过VO(Visual Odometry,视觉里程计)方法,并联合激光数据辅助建图的方法。但上述方法仍然存在累积误差的问题,因为纯激光的闭环方法存在闭环错误的情况,而利用视觉局部特征比如SIFT(Scale-invariant feature transform,尺度不变特征变换)的闭环方法难以处理一些重复度高的场景。因此,采用何种闭环方法消除累计误差从而提高建图精度,成为目前亟待解决的问题之一。
本发明提供一种场景地图建立方法、设备及存储介质,旨在提供一种使用内容不重复的图片辅助建立场景地图的技术,提高场景地图建立的精度。
为实现上述目的,本发明提供了一种场景地图建立方法,所述场景地图建立方法包括:
利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿。
根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息。
将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图。
为实现上述目的,本发明还提供了一种场景地图建立装置,所述场景地图建立装置包括:
图像采集模块,用于利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿。
坐标转换模块,用于根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
位置获取模块,用于当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息。
地图建立模块,用于将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图。
为实现上述目的,本发明还提供了一种电子设备,所述电子设备包括存储器和处理器,所述存储器上存储有可在所述处理器上运行的场景地图建立程序,所述场景地图建立程序被所述处理器运行时,执行所述的场景地图建立方法。
为实现上述目的,本发明还提供了一种计算机存储介质,所述存储介质上存储有场景地图建立程序,所述场景地图建立程序可以被一个或者多个处理器执行,以实现如所述的场景地图建立方法的步骤。
本发明一种场景地图建立方法、设备及存储介质可以达到如下有益效果:
利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿;根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标;当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息;将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图;达到了通过识别待建立地图的场景中内容互不重复的图片辅助建立场景地图的目的,提高了闭环的准确性,进而提高了场景地图的建图精度;由于每张预设图片内容唯一,因此避免了场景重复率高所造成的闭环错误的问题,提高场景地图建图精度的同时,也提高了建图效率。
附图用来提供对本发明的进一步理解,并且构成说明书的一部分,与本发明的实施例一起用于解释本发明,并不构成对本发明的限制。在附图中:
图1是本发明场景地图建立方法的一种实施方式的流程示意图。
图2是本发明场景地图建立装置的一种实施方式的功能模块示意图。
图3是本发明电子设备的一种实施方式的内部结构示意图。
以下结合附图对本发明的优选实施例进行说明,应当理解,此处所描述的优选实施例仅用于说明和解释本发明,并不用于限定本发明。
本发明提供了一种场景地图建立方法、设备及存储介质,通过识别待建立地图的场景中内容互不重复的图片,提高闭环的准确性,进而提高场景地图的建图精度;由于每张图片唯一,从而避免了场景重复率高所造成的闭环错误的问题,提高场景地图建图精度的同时,也提高了建图效率。
如图1所示,图1是本发明场景地图建立方法的一种实施方式的流程示意图;本发明一种场景地图建立方法可以实施为如下描述的步骤S10-S40:
步骤S10、利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿。
本发明实施例中,利用机器人采集待建立地图的场景中对应的当前图像,并识别采集的当前图像中是否存在预设图片。
其中,所述预设图片为在首次执行图1所述实施例的步骤S10之前,预先配置的;即在首次运行本发明描述的图1中场景地图建立方法之前,还需执行如下步骤:
针对待建立地图的场景,在所述场景中的不同位置处,预先配置内容互不重复的预设图片。
其中,所述预设图片包括:二维码和/或固定图像。
上述预设图片通常情况下,仅需首次运行场景地图建立方法时配置,后续运行图1实施例描述的方法步骤S10-S40时,直接使用已配置的预设图片即可。
若识别出采集的当前图像中存在内容互不重复的预设图片时,采集当前图像中存在的上述预设图片的四个角所分别对应的坐标,同时记录此时机器人的当前位姿。此时所采集的预设图片的四个角所分别对应的坐标是在当前图像对应的坐标系下的坐标。
本发明实施例中,机器人采集待建立地图的场景中的图像时,可以利用机器人自身配置的图像采集传感器;所述图像采集传感器包括但不限于:各种类型的相机和摄像机等,比如普通相机、深度摄像机等。本发明实施例对机器人采集场景中的图像所采用的图像采集传感器的具体数量和具体类型,不进行限定;在实际使用时,可以根据具体的应用场景进行配置。通过机器人身上配置的图像采集传感器,识别待建立地图的场景中配置的内容互不重复的预设图片,即可获取较为精准的、不重复的机器人的自身位置信息。
步骤S20、根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
根据识别出的预设图片的四个角分别对应的在机器人采集的当前图像中的位置坐标,通过转换计算,即可获取上述预设图片的四个角在场景地图坐标系下的三维坐标。
本发明实施例中,在机器人多个位置识别上述预设图片的情况下,可以利用多视角几何算法,计算得到上述预设图片的四个角在场景地图坐标系下的三维坐标。
在另一个实施例中,也可以结合机器人自身配置的图像采集传感器的类型,来获取该预设图片的四个角在场景地图坐标系下的三维坐标。比如,若该机器人额外装载了深度摄像机,则可以利用机器人装载的深度摄像机采集预设图片对应的深度图片,通过利用采集的所述预设图片并配合所述深度图片,即可获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
步骤S30、当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息。
步骤S40、将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图。
当机器人再次检测并采集到预设图片时,根据获取的该预设图片在场景地图中的三维坐标,即可计算得到机器人在上述场景地图对应的三维坐标系下的当前坐标。
在一个实施例中,计算机器人的所述当前坐标时,由于机器人自身配置的图像采集传感器的参数和预设标定是已知的,因此,可以利用机器人自身配置的图像采集传感器对应的参数,结合获取的所述预设图片的三维坐标,通过计算所述图像采集传感器在所述场景地图坐标系下的位置信息,进而计算得到机器人的当前坐标。
进一步地,在一个实施例中,根据机器人自身配置的图像采集传感器对应的参数,利用获取的所述预设图片的三维坐标,通过计算所述图像采集传感器在所述场景地图坐标系下的位置信息,计算得到机器人的当前坐标,可以按照如下方式实施:
当机器人再次检测并采集到预设图片时,设p为所述预设图片的其中一个角在所述预设图片上的坐标;P为p在场景地图对应的三维坐标系下的坐标,K为所述机器人采集所述预设图片时使用的所述图像采集传感器对应的内参矩阵,且所述内参矩阵K可以通过所述图像采集传感器的参数或者所述图像采集传感器的预设标定获取;C为所述图像采集传感器的外参矩阵,即图像采集器在场景地图对应的三维坐标系下的位置信息,则满足关系式(1):
p = K C P; (1)
利用所述关系式(1),即可求解未知量C;
机器人的当前位置坐标R和C满足关系式(2):
R = T C; (2)
其中,所述关系式(2)中T为所述图像采集传感器到机器人中心的变换矩阵,所述变换矩阵T可以通过所述图像采集传感器的预设标定获取。
利用关系式(2)即可计算得到机器人的当前位置坐标R。
得到机器人的当前位置坐标R,即可得到机器人对应的当前位置信息。
通过机器人的当前位置信息为场景地图的建立提供闭环信息,通过利用机器人采集对应的预设图片,识别场景中不重复的预设图片,提高闭环信息的准确性;由于每个预设图片对应的图像内容唯一,因此避免了场景重复率高所造成的闭环错误的问题。因此,将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,根据获取的所述预设图片的四个角对应的三维坐标,即可创建得到精度较高的场景地图。
本发明场景地图建立方法,利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿;根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标;当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息;将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图;达到了通过识别待建立地图的场景中内容互不重复的图片辅助建立场景地图的目的,提高了闭环的准确性,进而提高了场景地图的建图精度;由于每张预设图片内容唯一,因此避免了场景重复率高所造成的闭环错误的问题,提高场景地图建图精度的同时,也提高了建图效率。
对应于图1实施例所描述的一种场景地图建立方法,本发明实施例还提供了一种场景地图建立装置;如图2所示,图2是本发明场景地图建立装置的一种实施方式的功能模块示意图,图2仅仅从功能上来描述本发明场景地图建立装置;本发明场景地图建立装置在功能上包括:
图像采集模块100,用于利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿。
坐标转换模块200,用于根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
位置获取模块300,用于当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息。
地图建立模块400,用于将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图。
在一个实施例中,所述场景地图建立装置还包括配置模块;所述配置模块用于:
针对待建立地图的场景,在所述场景中的不同位置处,预先配置内容互不重复的预设图片。
其中,所述预设图片包括:二维码和/或固定图像。
在一个实施例中,所述图像采集模块100用于:
利用机器人自身配置的图像采集传感器,采集待建立地图的场景中对应的当前图像。
其中,所述图像采集传感器包括相机和/或深度摄像机。
在一个实施例中,所述坐标转换模块200用于:
根据识别出的所述预设图片的四个角分别对应的坐标,利用多视角几何算法,计算出所述预设图片的四个角在场景地图坐标系下的三维坐标。
在一个实施例中,所述坐标转换模块200用于:
利用机器人自身配置的深度摄像机,采集所述预设图片对应的深度图片,利用采集的所述预设图片并配合所述深度图片,获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
在一个实施例中,所述位置获取模块300用于:
当机器人再次检测并采集到预设图片时,根据机器人自身配置的图像采集传感器对应的参数,利用获取的所述预设图片的三维坐标,通过计算所述图像采集传感器在所述场景地图坐标系下的位置信息,计算得到机器人的当前坐标。
在一个实施例中,所述位置获取模块300用于:
当机器人再次检测并采集到预设图片时,设p为所述预设图片的其中一个角在所述预设图片上的坐标;P为p在场景地图对应的三维坐标系下的坐标,K为所述机器人采集所述预设图片时使用的所述图像采集传感器对应的内参矩阵,且所述内参矩阵K可以通过所述图像采集传感器的参数或者所述图像采集传感器的预设标定获取;C为所述图像采集传感器的外参矩阵,则满足关系式(1):
p = K C P; (1)
利用所述关系式(1),即可求解未知量C;
机器人的当前位置坐标R和C满足关系式(2):
R = T C; (2)
其中,所述关系式(2)中T为所述图像采集传感器到机器人中心的变换矩阵,所述变换矩阵T可以通过所述图像采集传感器的预设标定获取。
利用关系式(2)即可计算得到机器人的当前位置坐标R。
本发明场景地图建立装置,利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿;根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标;当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息;将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图;达到了通过识别待建立地图的场景中内容互不重复的图片辅助建立场景地图的目的,提高了闭环的准确性,进而提高了场景地图的建图精度;由于每张预设图片内容唯一,因此避免了场景重复率高所造成的闭环错误的问题,提高场景地图建图精度的同时,也提高了建图效率。
对应于图1和图2所述实施例的描述,本发明还提供了一种电子设备,所述电子设备可以按照图1所述的场景地图建立方法来创建对应的场景地图。如图3所示,图3是本发明电子设备的一种实施方式的内部结构示意图。
在本实施例中,电子设备1可以是PC(Personal Computer,个人电脑),也可以是智能手机、平板电脑、便携计算机等终端设备。该电子设备1至少包括存储器11、处理器12,通信总线13,以及网络接口14。
其中,存储器11至少包括一种类型的可读存储介质,所述可读存储介质包括闪存、硬盘、多媒体卡、卡型存储器(例如,SD或DX存储器等)、磁性存储器、磁盘、光盘等。存储器11在一些实施例中可以是电子设备1的内部存储单元,例如该电子设备1的硬盘。存储器11在另一些实施例中也可以是电子设备1的外部存储设备,例如电子设备1上配备的插接式硬盘,智能存储卡(Smart
Media Card, SMC),安全数字(Secure Digital,
SD)卡,闪存卡(Flash Card)等。进一步地,存储器11还可以既包括电子设备1的内部存储单元也包括外部存储设备。存储器11不仅可以用于存储安装于电子设备1的应用软件及各类数据,例如场景地图建立程序01的代码等,还可以用于暂时地存储已经输出或者将要输出的数据。
处理器12在一些实施例中可以是一中央处理器(Central
Processing Unit, CPU)、控制器、微控制器、微处理器或其他数据处理芯片,用于运行存储器11中存储的程序代码或处理数据,例如执行场景地图建立程序01等。
通信总线13用于实现这些组件之间的连接通信。
网络接口14可选的可以包括标准的有线接口、无线接口(如WI-FI接口),通常用于在该电子设备1与其他电子设备之间建立通信连接。
可选地,该电子设备1还可以包括用户接口,用户接口可以包括显示器(Display)、输入单元比如键盘(Keyboard),可选的用户接口还可以包括标准的有线接口、无线接口。可选地,在一些实施例中,显示器可以是LED显示器、液晶显示器、触控式液晶显示器以及OLED(Organic
Light-Emitting Diode,有机发光二极管)触摸器等。其中,显示器也可以适当的称为显示屏或显示单元,用于显示在电子设备1中处理的信息以及用于显示可视化的用户界面。
图3仅示出了具有组件11-14以及场景地图建立程序01的电子设备1,本领域技术人员可以理解的是,图2示出的结构并不构成对电子设备1的限定,可以包括比图示更少或者更多的部件,或者组合某些部件,或者不同的部件布置。
基于图1、图2实施例的描述,在图3所示的电子设备1实施例中,存储器11中存储有场景地图建立程序01;所述存储器11上存储的场景地图建立程序01可在所述处理器12上运行,所述场景地图建立程序01被所述处理器12运行时实现如下步骤:
利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿。
根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息。
将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图。
在一个实施例中,所述场景地图建立程序01还可以被所述处理器12运行,以在步骤“利用机器人采集待建立地图的场景中对应的当前图像”之前,还包括:
针对待建立地图的场景,在所述场景中的不同位置处,预先配置内容互不重复的预设图片。
其中,所述预设图片包括:二维码和/或固定图像。
在一个实施例中,所述场景地图建立程序01还可以被所述处理器12运行,以利用机器人采集待建立地图的场景中对应的当前图像,包括:
利用机器人自身配置的图像采集传感器,采集待建立地图的场景中对应的当前图像。
其中,所述图像采集传感器包括相机和/或深度摄像机。
在一个实施例中,所述场景地图建立程序01还可以被所述处理器12运行,以根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标,包括:
根据识别出的所述预设图片的四个角分别对应的坐标,利用多视角几何算法,计算出所述预设图片的四个角在场景地图坐标系下的三维坐标。
在一个实施例中,所述场景地图建立程序01还可以被所述处理器12运行,以根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标,包括:
利用机器人自身配置的深度摄像机,采集所述预设图片对应的深度图片,利用采集的所述预设图片并配合所述深度图片,获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
在一个实施例中,所述场景地图建立程序01还可以被所述处理器12运行,以当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,包括:
当机器人再次检测并采集到预设图片时,根据机器人自身配置的图像采集传感器对应的参数,利用获取的所述预设图片的三维坐标,通过计算所述图像采集传感器在所述场景地图坐标系下的位置信息,计算得到机器人的当前坐标。
在一个实施例中,所述场景地图建立程序01还可以被所述处理器12运行,以根据机器人自身配置的图像采集传感器对应的参数,利用获取的所述预设图片的三维坐标,通过计算所述图像采集传感器在所述场景地图坐标系下的位置信息,计算得到机器人的当前坐标,包括:
当机器人再次检测并采集到预设图片时,设p为所述预设图片的其中一个角在所述预设图片上的坐标;P为p在场景地图对应的三维坐标系下的坐标,K为所述机器人采集所述预设图片时使用的所述图像采集传感器对应的内参矩阵,且所述内参矩阵K可以通过所述图像采集传感器的参数或者所述图像采集传感器的预设标定获取;C为所述图像采集传感器的外参矩阵,则满足关系式(1):
p = K C P; (1)
利用所述关系式(1),即可求解未知量C;
机器人的当前位置坐标R和C满足关系式(2):
R = T C; (2)
其中,所述关系式(2)中T为所述图像采集传感器到机器人中心的变换矩阵,所述变换矩阵T可以通过所述图像采集传感器的预设标定获取。
利用关系式(2)即可计算得到机器人的当前位置坐标R。
本发明电子设备,利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿;根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标;当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息;将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图;达到了通过识别待建立地图的场景中内容互不重复的图片辅助建立场景地图的目的,提高了闭环的准确性,进而提高了场景地图的建图精度;由于每张预设图片内容唯一,因此避免了场景重复率高所造成的闭环错误的问题,提高场景地图建图精度的同时,也提高了建图效率。
此外,本发明实施例还提供了一种计算机存储介质,所述计算机存储介质上存储有场景地图建立程序,所述场景地图建立程序可以被一个或者多个处理器执行,以实现下操作:
利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿。
根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息。
将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图。
本发明计算机可读存储介质具体实施方式与上述场景地图建立方法、装置和电子设备对应的各实施例的实施原理基本相同,在此不作赘述。
本领域内的技术人员应明白,本发明的实施例可提供为方法、系统、或计算机程序产品。因此,本发明可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。
显然,本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样,倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内,则本发明也意图包含这些改动和变形在内。
本发明实施例利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿;根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标;当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息;将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图;达到了通过识别待建立地图的场景中内容互不重复的图片辅助建立场景地图的目的,提高了闭环的准确性,进而提高了场景地图的建图精度;由于每张预设图片内容唯一,因此避免了场景重复率高所造成的闭环错误的问题,提高场景地图建图精度的同时,也提高了建图效率。因此,具有工业实用性。
Claims (10)
- 一种场景地图建立方法,所述场景地图建立方法包括:利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿;根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标;当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息;将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图。
- 如权利要求1所述的场景地图建立方法,其中,所述方法在步骤“利用机器人采集待建立地图的场景中对应的当前图像”之前,还包括:针对待建立地图的场景,在所述场景中的不同位置处,预先配置内容互不重复的预设图片;其中,所述预设图片包括:二维码和/或固定图像。
- 如权利要求1或2所述的场景地图建立方法,其中,所述利用机器人采集待建立地图的场景中对应的当前图像,包括:利用机器人自身配置的图像采集传感器,采集待建立地图的场景中对应的当前图像;其中,所述图像采集传感器包括相机和/或深度摄像机。
- 如权利要求1或2所述的场景地图建立方法,其中,所述根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标,包括:根据识别出的所述预设图片的四个角分别对应的坐标,利用多视角几何算法,计算出所述预设图片的四个角在场景地图坐标系下的三维坐标。
- 如权利要求1或2所述的场景地图建立方法,其中,所述根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标,包括:利用机器人自身配置的深度摄像机,采集所述预设图片对应的深度图片,利用采集的所述预设图片并配合所述深度图片,获取所述预设图片的四个角在场景地图坐标系下的三维坐标。
- 如权利要求1或2所述的场景地图建立方法,其中,所述当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,包括:当机器人再次检测并采集到预设图片时,根据机器人自身配置的图像采集传感器对应的参数,利用获取的所述预设图片的三维坐标,通过计算所述图像采集传感器在所述场景地图坐标系下的位置信息,计算得到机器人的当前坐标。
- 如权利要求6所述的场景地图建立方法,其中,所述根据机器人自身配置的图像采集传感器对应的参数,利用获取的所述预设图片的三维坐标,通过计算所述图像采集传感器在所述场景地图坐标系下的位置信息,计算得到机器人的当前坐标,包括:当机器人再次检测并采集到预设图片时,设p为所述预设图片的其中一个角在所述预设图片上的坐标;P为p在场景地图对应的三维坐标系下的坐标,K为所述机器人采集所述预设图片时使用的所述图像采集传感器对应的内参矩阵,且所述内参矩阵K可以通过所述图像采集传感器的参数或者所述图像采集传感器的预设标定获取;C为所述图像采集传感器的外参矩阵,则满足关系式(1):p = K C P; (1)利用所述关系式(1),即可求解未知量C;机器人的当前位置坐标R和C满足关系式(2):R = T C; (2)其中,所述关系式(2)中T为所述图像采集传感器到机器人中心的变换矩阵,所述变换矩阵T可以通过所述图像采集传感器的预设标定获取;利用关系式(2)即可计算得到机器人的当前位置坐标R。
- 一种场景地图建立装置,所述场景地图建立装置包括:图像采集模块,用于利用机器人采集待建立地图的场景中对应的当前图像,识别采集的所述当前图像中存在的预设图片的四个角所分别对应的坐标,同时记录机器人的当前位姿;坐标转换模块,用于根据识别出的所述预设图片的四个角分别对应的坐标,获取所述预设图片的四个角在场景地图坐标系下的三维坐标;位置获取模块,用于当机器人再次采集到预设图片时,根据获取的所述预设图片的三维坐标,计算得到机器人的当前坐标,并根据机器人的当前坐标,得到机器人的当前位置信息;地图建立模块,用于将获取的所述机器人的当前位置信息作为建立场景地图的闭环信息,参照所述闭环信息以及获取的所述预设图片的四个角对应的所述三维坐标,建立场景地图。
- 一种电子设备,所述电子设备包括存储器和处理器,所述存储器上存储有可在所述处理器上运行的场景地图建立程序,所述场景地图建立程序被所述处理器运行时,执行如权利要求1至7中任一项所述的场景地图建立方法。
- 一种计算机存储介质,所述存储介质上存储有场景地图建立程序,所述场景地图建立程序可以被一个或者多个处理器执行,以实现如权利要求1至7中任一项所述的场景地图建立方法的步骤。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911379873.7 | 2019-12-27 | ||
CN201911379873.7A CN111060118B (zh) | 2019-12-27 | 2019-12-27 | 场景地图建立方法、设备及存储介质 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021129345A1 true WO2021129345A1 (zh) | 2021-07-01 |
Family
ID=70304341
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2020/133735 WO2021129345A1 (zh) | 2019-12-27 | 2020-12-04 | 场景地图建立方法、设备及存储介质 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN111060118B (zh) |
WO (1) | WO2021129345A1 (zh) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111060118B (zh) * | 2019-12-27 | 2022-01-07 | 炬星科技(深圳)有限公司 | 场景地图建立方法、设备及存储介质 |
CN113183153A (zh) * | 2021-04-27 | 2021-07-30 | 北京猎户星空科技有限公司 | 一种地图创建方法、装置、设备及介质 |
CN113761255B (zh) * | 2021-08-19 | 2024-02-09 | 劢微机器人科技(深圳)有限公司 | 机器人室内定位方法、装置、设备及存储介质 |
CN114638894B (zh) * | 2022-03-18 | 2024-07-23 | 纯米科技(上海)股份有限公司 | 机器人行走的定位方法、系统、电子装置及存储介质 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050228555A1 (en) * | 2003-08-20 | 2005-10-13 | Samsung Electronics Co., Ltd. | Method of constructing artificial mark for autonomous driving, apparatus and method of determining position of intelligent system using artificial mark and intelligent system employing the same |
CN106846497A (zh) * | 2017-03-07 | 2017-06-13 | 百度在线网络技术(北京)有限公司 | 应用于终端的呈现三维地图的方法和装置 |
CN109556617A (zh) * | 2018-11-09 | 2019-04-02 | 同济大学 | 一种自动建图机器人的地图要素提取方法 |
CN109949422A (zh) * | 2018-10-15 | 2019-06-28 | 华为技术有限公司 | 用于虚拟场景的数据处理方法以及设备 |
CN110446164A (zh) * | 2019-07-23 | 2019-11-12 | 深圳前海达闼云端智能科技有限公司 | 移动终端定位方法、装置、移动终端和服务器 |
CN111060118A (zh) * | 2019-12-27 | 2020-04-24 | 炬星科技(深圳)有限公司 | 场景地图建立方法、设备及存储介质 |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101751697B (zh) * | 2010-01-21 | 2011-09-14 | 西北工业大学 | 一种基于统计模型的三维场景重建方法 |
CN107180215B (zh) * | 2017-05-31 | 2020-01-31 | 同济大学 | 基于库位和二维码的停车场自动建图与高精度定位方法 |
CN107505644B (zh) * | 2017-07-28 | 2020-05-05 | 武汉理工大学 | 基于车载多传感器融合的三维高精度地图生成系统及方法 |
CN109816769A (zh) * | 2017-11-21 | 2019-05-28 | 深圳市优必选科技有限公司 | 基于深度相机的场景地图生成方法、装置及设备 |
CN110570465B (zh) * | 2018-06-05 | 2022-05-20 | 杭州海康机器人技术有限公司 | 实时定位与地图构建方法、装置及计算机可读存储介质 |
CN109556616A (zh) * | 2018-11-09 | 2019-04-02 | 同济大学 | 一种基于视觉标记的自动建图机器人建图修整方法 |
-
2019
- 2019-12-27 CN CN201911379873.7A patent/CN111060118B/zh active Active
-
2020
- 2020-12-04 WO PCT/CN2020/133735 patent/WO2021129345A1/zh active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050228555A1 (en) * | 2003-08-20 | 2005-10-13 | Samsung Electronics Co., Ltd. | Method of constructing artificial mark for autonomous driving, apparatus and method of determining position of intelligent system using artificial mark and intelligent system employing the same |
CN106846497A (zh) * | 2017-03-07 | 2017-06-13 | 百度在线网络技术(北京)有限公司 | 应用于终端的呈现三维地图的方法和装置 |
CN109949422A (zh) * | 2018-10-15 | 2019-06-28 | 华为技术有限公司 | 用于虚拟场景的数据处理方法以及设备 |
CN109556617A (zh) * | 2018-11-09 | 2019-04-02 | 同济大学 | 一种自动建图机器人的地图要素提取方法 |
CN110446164A (zh) * | 2019-07-23 | 2019-11-12 | 深圳前海达闼云端智能科技有限公司 | 移动终端定位方法、装置、移动终端和服务器 |
CN111060118A (zh) * | 2019-12-27 | 2020-04-24 | 炬星科技(深圳)有限公司 | 场景地图建立方法、设备及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
CN111060118B (zh) | 2022-01-07 |
CN111060118A (zh) | 2020-04-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021129345A1 (zh) | 场景地图建立方法、设备及存储介质 | |
WO2022073427A1 (zh) | 物体抓取点的视觉定位方法、装置、存储介质及电子设备 | |
CN112767489B (zh) | 一种三维位姿确定方法、装置、电子设备及存储介质 | |
US10915998B2 (en) | Image processing method and device | |
WO2018098811A1 (zh) | 定位方法和装置 | |
US20190096089A1 (en) | Enabling use of three-dimensonal locations of features with two-dimensional images | |
EP3477543A1 (en) | System and method using augmented reality for efficient collection of training data for machine learning | |
WO2022000755A1 (zh) | 机器人及其行动控制方法、装置和计算机可读存储介质 | |
TW201346640A (zh) | 影像處理裝置及電腦程式產品 | |
US9069415B2 (en) | Systems and methods for finger pose estimation on touchscreen devices | |
CN110956131B (zh) | 单目标追踪方法、装置及系统 | |
CN110648363A (zh) | 相机姿态确定方法、装置、存储介质及电子设备 | |
JP6293386B2 (ja) | データ処理装置、データ処理方法及びデータ処理プログラム | |
CN109934873B (zh) | 标注图像获取方法、装置及设备 | |
CN110245643A (zh) | 目标跟踪拍摄方法、装置、电子设备 | |
JP2019106008A (ja) | 推定装置、推定方法、及び推定プログラム | |
WO2023179342A1 (zh) | 重定位方法及相关设备 | |
WO2024021340A1 (zh) | 机器人跟随方法、装置、机器人及计算机可读存储介质 | |
CN113129366A (zh) | 单目slam初始化方法、装置及电子设备 | |
CN110853098A (zh) | 机器人定位方法、装置、设备及存储介质 | |
TWM637241U (zh) | 運算裝置及模型產生系統 | |
CN113628284B (zh) | 位姿标定数据集生成方法、装置、系统、电子设备及介质 | |
CN105631938A (zh) | 一种图像处理方法及电子设备 | |
CN114329675A (zh) | 模型生成方法、模型生成装置、电子设备和可读存储介质 | |
TWI571753B (zh) | 用於產生一影像之一互動索引碼圖之電子計算裝置、其方法及其電腦程式產品 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20905936 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20905936 Country of ref document: EP Kind code of ref document: A1 |