WO2019062814A1

WO2019062814A1 - Landmark-based positioning robot system, landmark and robot used thereby, and robot landmark-based positioning method

Info

Publication number: WO2019062814A1
Application number: PCT/CN2018/107964
Authority: WO
Inventors: 凌子龙
Original assignee: 凌子龙
Priority date: 2017-09-29
Filing date: 2018-09-27
Publication date: 2019-04-04
Also published as: CN109582011A

Abstract

A robot positioning landmark, comprising: an RFID card (1) and a graphic identifier (2) comprising an identification origin and coordinate pointing information. A landmark-based positioning robot, comprising: a robot body capable of moving, an RFID card reader (3) and a camera module (4). An antenna of the RFID card reader is provided at the bottom of the robot, a capturing window (5) is provided at the bottom of the robot, and the camera module (4) is provided within the robot body and the camera is disposed to face the capturing window (5) at the bottom of the robot. A landmark-based positioning robot system, comprising a plurality of positioning landmarks, at least one robot and an upper computer, the upper computer comprising a wireless communication module, and the robot also comprising a wireless communication module.

Description

Landmark positioning robot system and landmarks and robot and robot landmark positioning method used thereby

Technical field

The present invention relates to the field of robot technology, and in particular to a landmark positioning robot system and a landmark and robot and robot landmark positioning method therefor.

Background technique

At present, with the development of robot technology, the handling and sorting operations of storage robots and sorting robots in the canonical area have been well applied. Mobile robots, especially VGA robots, have strong dependence on landmarks, with Amazon's kiva For example, the robot system uses two-dimensional code landmarks to assist the robot to self-position and realize navigation. Landmarks are particularly important as robots that rely on landmarks for self-positioning. With the rapid development of information technology, QR codes are widely used in various enterprises and Products, QR code brings together information about the company and products, by scanning the QR code to obtain relevant information, and can also be used as a landmark to mark the current location information, the robot navigates according to the current location and the target position of the robot; Another method is to use an RFID card as a robot landmark, and the robot reads the position information corresponding to the currently read RFID card through its own RFID card reader.

technical problem

However, only the use of two-dimensional code landmarks requires a high working environment for the robot. The robot may be unrecognizable or misidentified due to ambient light interference or two-dimensional code landmark pollution, and the camera is required to continuously capture each frame of the video image. The recognition and judgment are made. When the robot runs at a high speed, there is a high requirement for the number of frames and the shutter speed of the camera; when the RFID card is used as the robot landmark, the robot reads the currently read RFID through its own RFID card reader. The location information corresponding to the card, but due to the non-contact characteristics of the RFID card, the robot does not easily determine the precise position when reading the card, and it is more difficult to determine the current orientation information of the robot by reading an RFI D card; in the warehouse logistics and points The professional robots in the field of sorting operate in a systematic group. When individual landmarks cannot be read or the robot positioning is inaccurate, it has a great impact on the stable operation of the robot system.

Problem solution

Technical solution

In order to solve the above problems, the present invention proposes a landmark positioning robot system and a landmark and robot and robot landmark positioning method therefor.

The technical solution adopted by the present invention is.

A robot positioning landmark includes an RFID card and a graphic identifier including an identification origin and coordinate pointing information. The graphic identifier is a graphic identifier that can be imaged by the camera. The RFID card is a card containing an RFID chip and a radio frequency antenna. To prevent mutual interference, a passive near field RFID card is selected, and the RFID card in the landmark of different positions records different. Location information, but all the robotic positioning landmarks may have the same graphic identifier. The above-mentioned robot positioning landmark proposed by the present invention requires a movable robot to be equipped with an RFID card reader and a camera, and the RFID card reader is used to read the RFID card storage in the landmark. The location information of the landmark is captured by the camera, and then the identification image is captured by the captured image, and the identification origin and coordinate pointing information are intercepted therefrom, and the robot accurately determines the current precise position of the robot through the identification origin and the coordinate pointing information on the image. Accurately pointing, the above method for recognizing the image of the captured logo can be realized by using the captured image and the information stored in the coordinate axis of the logo image pre-stored by the system for comparison and recognition; and can be read by the radio frequency card Trigger camera capture or trigger pair The current frame image captured by the camera is recognized; when the NFC card is used in place of the RFID card in the landmark of the present invention, its use characteristics are not significantly different from those of the RFID card, and it can be understood that the NFC card is used as the RFID card. An adhesive layer is pre-positioned on the other side of the RFID card of the card. In order to facilitate the laying of the landmark of the present invention, a predetermined adhesive layer on one side of the RFID card may be fixed to the ground by gluing. The graphic is made using a reflective material. In order to better increase the degree of recognition, the graphic may be printed using a reflective coating or may be cut into a graphic using a reflective material. The landmark of the present invention is formed by an inlay or paste method, and the method of printing the graphic includes reverse printing.

The graphic identifier is an asymmetric identifier or a non-central symmetric graphic identifier. The graphic identifier is a graphic identifier that can be imaged by the camera. The RFID card is a card containing an RFID chip and a radio frequency antenna. To prevent mutual interference, a passive near field RFID card is selected, and the RFID card in the landmark of different positions records different. Location information, but all non-central symmetric graphic identifiers of the robot positioning landmarks may be the same. The above-mentioned robot positioning landmarks proposed by the present invention require a movable robot to be equipped with an RFID reader and a camera, and the RFID reader is used to read the landmarks. The location information of the landmark is stored by the RFID card, and the camera captures the asymmetric identification image, and then recognizes the captured asymmetric identification image, and intercepts the identification origin and the coordinate pointing information, and the robot relatively judges the identification origin and the coordinate pointing information on the image. Taking out the current precise position and precise pointing of the robot, the above method for recognizing the captured asymmetrically-identified image may use the captured image to compare and recognize the information in the coordinate axis of the asymmetric identification image pre-stored by the system. Achieved; The reading of the RF card triggers the camera to capture or trigger the recognition of the current frame image captured by the camera; when the NFC card is used in place of the RFID card in the landmark of the present invention, the use characteristics are not significantly different from the RFID card, and can be understood as using NFC. The card acts as an RFID card. An adhesive layer is pre-positioned on the other side of the RFID card of the card. In order to facilitate the laying of the landmark of the present invention, a predetermined adhesive layer on one side of the RFID card may be fixed to the ground by gluing. The non-central symmetrical pattern is made using a reflective material. In order to better increase the degree of recognition, the non-central symmetrical pattern may be printed using a reflective paint or may be cut into a pattern using a reflective material. The landmark of the present invention is formed by an inlay or paste method with an RFID card, and the method for printing the graphic includes the reverse White printing.

The graphic identification including the identification origin and the coordinate pointing information is printed on the RFID card. The graphic is printed on the face (or front) of the RFID card facing up, which facilitates the laying.

A landmark positioning robot includes a robot body with a mobile function, an RFID card reader and a camera module, the RFID card reader antenna is disposed at a bottom of the robot, and a bottom of the robot is provided with a shooting window, and the camera module Set in the robot body and point the camera toward the bottom of the robot to shoot the window. The RFID reader at the bottom of the robot can read the RFID card in the landmark, the camera module can capture the graphic image on the landmark, and the robot can read the position coordinate information of the landmark from the RFID card in the landmark, the camera module The graphic image on the photographed landmark can help the robot to further locate the coordinate center point and the coordinate axis; it is used in conjunction with the landmark combination of the RFID card and the non-central symmetric graphic identifier, and the RFID reader of the robot is used to read the RFID card stored in the landmark. The location information of the landmark, and the camera captures the graphic identification image of the landmark, and then recognizes the captured asymmetric identification image, and intercepts the identification origin and the coordinate pointing information. The robot determines the current state of the robot through the identification origin and the coordinate pointing information on the image. Accurate position and precise pointing, the above method for recognizing the captured asymmetrically-identified image can be realized by using the captured image and the information stored in the coordinate axis of the pre-stored asymmetric identification image in the coordinate axis for recognition and recognition; The robot is a mobile robot, generally Driven robot, the robot body comprises a casing, a driving wheel, a driving wheel motor, a transmission device, a power module, a main control circuit module and an external communication module, and the external communication module generally refers to a robot and a host computer. Communication is usually implemented by using a WiFi communication module.

The camera module triggers the camera to capture or trigger the recognition of the current frame image captured by the camera by reading the RFID card. Continuous shooting and recognition not only increases the computing load, but also does not guarantee the position of each frame of the image to be photographed. It is more efficient and accurate to use the RFID card reader to read the landmark when triggering the shutter. Of course, the camera can also be used. Continuous operation, by triggering a high-brightness flash to obtain a high-quality image, and more specifically identifying the current frame image when the RFID reader is triggered to read the information.

The RFID reader antenna is annular, and the antenna surrounds a shooting window at the bottom of the robot facing the camera. In order to increase the accuracy of coordinate positioning, the graphic identification pattern of the landmark may be printed on the RFID card to form an overlap, so that the robot's camera is read by the reader at the bottom of the robot when the RFID card is read from any angle above the landmark. The module just points to the graphic logo of the landmark, which is good for shooting and better positioning.

The robot body comprises a casing, a driving wheel, a driving wheel motor, a transmission device, a power module, a main control circuit module and an external communication module. The above external communication module generally refers to communication between the robot and the host computer, and is usually implemented by using a WiFi communication module.

A landmark positioning robot system comprising a plurality of the above-mentioned landmarks and at least one of the above-mentioned robots and a host computer, the host computer comprising a wireless communication module, the robot also comprising a wireless communication module. A set of landmark system generally first sets the landmark on the potential inflection point of all the robots' travelable routes in the field, and then the robot needs to start on one of the landmarks. Under this premise, the landmark navigation robot system can be under the command of the upper computer, all the robots. They are moved between the punctuation points in the venue according to the instructions.

The landmark positioning robot system is a VGA robot system. AGV is the abbreviation of "Automated Guided Vehicle", which means "automatic guided transport vehicle". It is equipped with automatic guiding device such as electromagnetic or optical. It can travel along the prescribed guiding path, with safety protection and various shifts. The transport vehicle with the function, the AGV belongs to the category of Wheeled Mobile Robot (WMR). The landmark positioning robot in the present invention can be an RFID card and a graphic logo as a landmark, by reading the RFID card. And the identification of the landmark on the landmark to realize the automatic guided transport vehicle, the above-mentioned landmark of the present invention and the automatic guided transport vehicle robot conforming to the robot feature of the present invention constitute a VGA robot system.

The steps of the positioning method of the landmark positioning robot system are as follows:

In the first step, the robot reads the RFID card information of the first coordinate point above a certain landmark point, the information records the coordinate position of the landmark point, and the robot obtains the coordinate position information;

In the second part, the robot captures the logo pattern image of a certain landmark, and the robot applies the preset image recognition information to determine the coordinate pointing angle of the original origin and the logo pattern in the image; the robot reverses the position of the logo origin in the image. The positional relationship between the center point of the robot body and the origin on the landmark image is pushed to determine the current precise position of the robot; the robot determines the orientation and landmark of the current robot according to the relative angle of the pixel vertical coordinate of the image and the coordinate of the pattern on the image. The relationship between the coordinates of the orientation.

The premise of this method is that the robot internally stores a coordinate recognition algorithm for the pattern on the landmark. The basis of the algorithm is based on the coordinates of the preset pattern and the origin definition. The process of robot positioning can be understood as finding a preset pattern on the image. The process, correctly identifying the preset pattern, is to identify the position and angle of the landmark logo pattern on the image on the captured image.

The steps of the robot navigation method of the landmark positioning robot system are as follows:

In the first step, the robot determines the precise position and orientation of the current robot through the above position determination method;

The second part calculates a waypoint landmark to be passed according to the target position coordinate or path information received by the wireless signal;

In the third part, the robot aligns to point to the next waypoint until the RFID card reader recognizes the waypoint information;

The fourth part determines whether the new waypoint information is the next waypoint of the last route of the planned route; if yes, repeats the third step, and if not, returns to the second step to repeat;

In the fifth step, the RFID card reader completes the navigation after identifying the target waypoint information.

In the above navigation process, the information of all landmark locations in the map and the map may be pre-stored in the robot, or the target location and the route map may be sent to the robot by the upper computer through the wireless signal, but if the robot deviates from the route to the path At the time of the point, the path is calculated again by the host computer and the path map is acquired. This process can also add inertial navigation to navigate the driving state of the robot between the two waypoints.

Advantageous effects of the invention

Beneficial effect

The core point of the landmark positioning robot system proposed by the present invention is that the venue is arranged by setting an RFID card and a landmark that can indicate the coordinate orientation and the pattern of the origin. The robot reads the RFID card using the RFID card reader to obtain the coordinate information of the current position. Shooting and recognizing the graphical logo on the landmark to assist the robot in determining its current positional deviation and orientation deviation relative to the origin of the coordinate, so that the robot can comprehensively judge its current absolute position and absolute orientation by combining the RFID reading card with the camera to capture the image. Information, this system has strong environmental adaptability and accurate positioning.

Brief description of the drawing

DRAWINGS

FIG. 1 is a schematic diagram 1 of a robot positioning landmark of Embodiment 1.

FIG. 2 is a schematic diagram 2 of a robot positioning landmark of Embodiment 1.

FIG. 3 is a schematic diagram 1 of a landmark positioning robot of Embodiment 1.

FIG. 4 is a schematic diagram 2 of a landmark positioning robot of Embodiment 1.

The meanings of the marks in the figure are as follows: 1. RFID card; 2, non-central symmetric graphic logo; 3, RFID card reader; 4, camera module; 5, shooting window.

Invention embodiment

Embodiments of the invention

Examples of specific embodiments are set forth below in conjunction with the accompanying drawings.

As shown in FIGS. 1 to 4, a robot positioning landmark includes an RFID card and a non-central symmetric graphic logo, wherein the RFID card selects an operating frequency of 13.56 MHz, a communication protocol: ISO14443, and the outer casing is packaged in a plastic 100 mm*100 mm*1 mm. A non-central symmetrical graphic mark as shown in the figure is printed on the front surface of the package casing, and a double-sided tape is pre-set on the back surface of the package casing. A landmark positioning robot, comprising: a robot body with a mobile function, an RFID card reader and a camera module, the RFID reader antenna is arranged at the bottom of the robot, and the working frequency is 13.56 MHz, and the communication protocol is ISO14443. The bottom of the robot is provided with a shooting window of 160 mm*160 mm. The camera module is provided with an 8 megapixel camera in the robot body and the camera is directed toward the bottom of the robot. The robot body includes a cast aluminum alloy chassis and an ABS cover. The casing, the two driving wheels, the driving wheel motor selects 1kW servo motor, the transmission device selects planetary reducer, the power module selects 18650 lithium battery pack with power management circuit, the main control circuit takes ARM as the core module and external The communication module selects the WiFi communication module to realize communication between the robot and the host computer. When laying the system, the map information (including all landmark information on the map) and the identification pattern recognition method on the landmark are stored in the robot in advance. The positioning method of the landmark positioning robot system is as follows: First, the robot is above a certain landmark point, Reading the RFID card information to the first coordinate point, the information records the coordinate position of the landmark point, and the robot obtains the coordinate position information; the second part, the robot captures the logo pattern image of a certain landmark, and the robot applies the image to the pre-preparation The image recognition information is used to determine the coordinate pointing angle of the identification origin and the logo pattern in the image; the robot reverses the positional relationship between the center point of the robot body and the origin on the landmark image according to the position of the logo origin in the image, and determines the current accuracy of the robot. Position; the robot determines the relationship between the orientation of the current robot and the coordinate orientation of the landmark indication according to the relative angle of the pixel vertical coordinate of the image and the coordinate of the pattern on the image. The process of robot positioning can be understood as the process of finding a preset pattern on an image. Correctly identifying the preset pattern means recognizing the position and angle of the landmark marking pattern on the image on the captured image. When the robot is required to move from a position to a certain target position, the target position landmark coordinates are first sent to the robot, and then the robot determines the precise position and orientation of the current robot through the above position determination method; the target position received according to the wireless signal The coordinates or path information calculates the waypoint landmarks that need to pass; the robot aligns to point to the next waypoint until the RFID reader recognizes the waypoint information; repeats the above steps until the robot passes all terminal waypoints, the RFID card reads The navigation is completed after the target waypoint information is recognized.

Claims

A robot positioning landmark, comprising: an RFID card and a graphic identifier including an identification origin and coordinate pointing information.
A robot positioning landmark according to claim 1, wherein the graphic identifier is an asymmetric identifier or a non-central symmetric graphic identifier.
A robot positioning landmark according to claim 1, wherein the graphic identifier including the identification origin and the coordinate pointing information is printed on the RFID card.
A landmark positioning robot, comprising: a robot body having a mobile function, an RFID card reader and a camera module, wherein the RFID card reader antenna is disposed at a bottom of the robot, and the bottom of the robot is provided with a shooting window. The camera module is disposed in the robot body and faces the camera toward the bottom of the robot.
The landmark positioning robot according to claim 4, wherein the camera module triggers the camera to capture or trigger the recognition of the current frame image captured by the camera by reading the RFID card.
A landmark positioning robot according to claim 4, wherein the RFID reader antenna is in a ring shape, and the antenna surrounds a shooting window at the bottom of the robot facing the camera.
The landmark positioning robot according to claim 4, wherein the robot body comprises a casing, a driving wheel, a driving wheel motor, a transmission device, a power module, a main control circuit module and an external communication module. .
A landmark positioning robot system of a robot positioning landmark according to claim 1 and a landmark positioning robot according to claim 4, comprising a plurality of said landmarks and at least one of said robots and a host computer, said upper position The machine includes a wireless communication module, and the robot also includes a wireless communication module.
A landmark positioning robot system according to claim 8, wherein the landmark positioning robot system is a VGA robot system.
The landmark positioning robot system according to claim 8, wherein the positioning method of the landmark positioning robot system is as follows:

In the first step, the robot reads the RFID card information of the first coordinate point above a certain landmark point, the information records the coordinate position of the landmark point, and the robot obtains the coordinate position information;

In the second part, the robot captures the logo pattern image of a certain landmark, and the robot applies the preset image recognition information to determine the coordinate pointing angle of the original origin and the logo pattern in the image; the robot reverses the position of the logo origin in the image. The positional relationship between the center point of the robot body and the origin on the landmark image is pushed to determine the current precise position of the robot; the robot determines the orientation and landmark of the current robot according to the relative angle of the pixel vertical coordinate of the image and the coordinate of the pattern on the image. The relationship between the coordinates of the orientation.
A landmark positioning robot system according to claim 8, wherein the steps of the robot navigation method of the landmark positioning robot system are as follows:

In the first step, the robot determines the precise position and orientation of the current robot through the above position determination method;

The second part calculates a waypoint landmark to be passed according to the target position coordinate or path information received by the wireless signal;

In the third part, the robot aligns to point to the next waypoint until the RFID card reader recognizes the waypoint information;

The fourth part determines whether the new waypoint information is the next waypoint of the last route of the planned route; if yes, repeats the third step, and if not, returns to the second step to repeat;

In the fifth step, the RFID card reader completes the navigation after identifying the target waypoint information.