WO2019104951A1

WO2019104951A1 - Artificial intelligence-based shopping device and system applied to intelligent trading areas

Info

Publication number: WO2019104951A1
Application number: PCT/CN2018/087232
Authority: WO
Inventors: 王坚
Original assignee: 特斯联（北京）科技有限公司
Priority date: 2017-11-28
Filing date: 2018-05-17
Publication date: 2019-06-06
Also published as: CN108053582A; CN108053582B

Abstract

Disclosed is an artificial intelligence-based shopping device, comprising a main body, a travel driving system, a body control interface, an external sensing system, a central intelligent system, a remote communication system, a positioning system, and a display touch system, wherein the external sensing system comprises twin cameras and an anti-collision radar. The artificial intelligence-based shopping device and system of the present invention can be self-driven for traveling at low speed, have the functions of automatic path navigation driving, obstacle avoidance, user following, and goods preservation, and can help users moving in an intelligent trading area implement rapid settlement, goods carrying and storage. The artificial intelligence-based shopping device fully integrates the technical means such as artificial intelligence and mobile payment, and can improve the experience and efficiency of users in the whole process of shopping and leisure activities in a large-scale trading area.

Description

Artificial intelligence shopping device and system thereof applied to smart business circle

Technical field

The invention relates to the field of artificial intelligence devices, in particular to an artificial intelligence shopping device applied to a smart business circle.

Background technique

The large commercial districts in the city generally focus on ultra-large shopping malls and supermarkets. There are a large number of different types of stores within a certain geographical space, covering life supermarkets, shopping, cinema lines, video games, early childhood education, A wealth of services such as sports and fitness.

Large-scale business districts are an important place for the lives of urban residents. Especially for holidays, many people tend to spend most of the day or even a whole day in the business district. In addition to shopping, they also engage in eating, entertainment, fitness, beauty, etc. Kind of activity. The number of lockers installed in large commercial districts is often very limited and cannot meet the needs of most customers. Therefore, customers often have to carry the purchased items or other activities while carrying the purchased items, when they purchase more items. Heavier customers can cause inconvenience, fatigue, and loss. This situation will also reduce the fun of customers and affect customers to enjoy other services in the business circle.

With the comprehensive coverage of mobile Internet and the increasing popularity of intelligent terminals, as well as the transformation of traditional business models by emerging technologies such as Internet of Things, big data, and artificial intelligence, large-scale business districts have informatized and intelligentized through software and hardware. Upgrade and transform into a new smart business district. New business forms such as product inquiries and shopping navigation, advertising pushes, online appointments, customization of personalized goods and services, and mobile payments are typical examples of smart business circles. The core value of the smart business circle is to provide users with more convenient and more personalized services to improve the user experience.

However, the smart business district that is actually put into use is mainly implemented by providing customer-oriented smart phones with information, navigation, payment, customer service, coupons and other online services. Although these online services can provide certain convenience, The actual experience defects in the real shopping process mentioned above cannot be fundamentally improved. At present, the hardware facilities for providing smart business-type services to customers in the real shopping scenes under the Internet are mainly various types of unmanned sales and unmanned stores. These facilities only focus on the purchase link itself, although it can be To the extent that the purchase process itself is saved, but it is not possible to provide customers with help in the whole process of shopping and leisure in a large business district, so as to help users improve their experience and increase their fun.

Summary of the invention

(1) Technical problems solved

In view of the deficiencies of the prior art, the present invention provides an artificial intelligence shopping device applied to a smart business circle. The artificial intelligence shopping device can independently drive low-speed travel, and has the functions of path navigation automatic driving, obstacle avoidance, user following, and item preservation, and can help the users who are active in the smart business circle to carry and store the articles.

(2) Technical plan

To achieve the above object, the present invention is achieved by the following technical solutions:

An artificial intelligence shopping device, comprising: a main body body, a traveling drive system, a body control interface, an external sensing system, a central intelligent system, a remote communication system, a positioning system, a display touch system; wherein the external The sensing system includes: dual camera, anti-collision radar;

The main body is for providing a space for accommodating the carried articles, and can control locking and unlocking the cover located on the top surface of the space, and for mounting the traveling drive system;

The traveling drive system is used to drive the device to autonomously travel, and adjust the traveling speed and direction;

The airframe control interface is configured to send an unlocking instruction to the main body to unlock the cover, and send an instruction to the traveling drive system to control its travel, stop, and speed;

The remote communication system is configured to connect to a background server and perform two-way communication with the background server;

The positioning system is configured to implement real-time positioning of the device;

The dual camera of the external sensing system is used for capturing a left-view video picture and a right-view video picture in front of the device; the anti-collision radar of the external sensing system is configured to detect whether there is an obstacle within a certain distance in the direction around the device;

The display touch system is a liquid crystal display and a capacitive touch panel;

The central intelligent system sets an autonomous driving path based on the user position captured by the dual camera, and controls the traveling driving system to travel according to the autonomous driving path through the body control interface, and the traveling real-time picture and the obstacle detection of the anti-collision radar according to the dual camera shooting Adjusting the process of traveling; the central intelligent system also controls the electronic locker to unlock the cover body through the human body interaction of the user through the body control interface; and is used to obtain a list of unpaid items carried by the device, and the unpaid amount is The item list is uploaded to the background server through the remote communication system; the payment QR code generated according to the unpaid item list is received from the background server, and the payment QR code is displayed on the display touch system for the user to scan the payment.

Preferably, the main body includes: an article carrying case, a chassis, a cover, and an electronic lock; the anti-collision radar is mounted on an outer wall of the side of the box of the article carrying case; and the outer wall of the side of the case is mounted with a dual camera in front a touch display system is further disposed on the outer wall of the side of the box; an electronic label reader is mounted on the inner wall of the side of the box of the article carrying box for reading information contained in the electronic label attached to the product placed in the box; The cover body is rotatably mounted on the article carrying case by a hinge pivot; the cover body has a latch at an end opposite to the hinge pivot, and the electronic lock is mounted on the inner wall of the side of the box of the article carrying case After the cover is fastened, the electronic lock locks the cover by a lock; and the electronic lock unlocks the cover according to an unlocking command sent by the airframe control interface; the chassis is located on the article At the bottom of the cabinet, a travel drive system is mounted on the chassis.

Preferably, the traveling drive system comprises an X-shaped bracket, a first rear wheel, a second rear wheel, a drive motor, a drive rod, a first front wheel, a second front wheel, a steering drive motor, and a steering control shaft; The front wheel and the two rear wheels are respectively mounted on the respective branches of the X-shaped bracket, wherein the two rear wheels are connected to the driving rod, and the driving motor drives the driving rod to rotate, thereby driving the two rear wheels to rotate, which is the autonomous travel of the device. Power is provided; the drive motor is fixedly mounted on the X-shaped bracket; the motor shaft of the steering drive motor is coupled to the steering control shaft, the steering control shaft is coupled to the two front wheels, and the two branches of the two front wheels and the X-shaped bracket are passed The universal joint is connected, and with the motor shaft output torque of the steering drive motor, the steering control shaft drives the two front wheels to adjust a certain angle, thereby adjusting the traveling direction of the device.

Preferably, the central intelligent system specifically includes: a target stereoscopic sensing subsystem, a user identification subsystem, an autonomous driving control subsystem, and a human-computer interaction subsystem; and the target stereoscopic sensing subsystem obtains the left-view video image from the dual camera. And a right-view video image, using a target recognition algorithm based on edge detection and vector information to extract respective targets in the video image, and determining three-dimensional position information of each target; the user recognition subsystem is configured to use the target stereo perception subsystem The extracted objects are compared with the user features collected and stored in advance, the user target is identified from all the extracted targets, and the three-dimensional position information of the user target is determined; the autonomous driving control subsystem sets the three-dimensional position information according to the user target. Defining the path that the device intends to travel autonomously, and controlling the traveling drive system to travel according to the autonomous driving path through the airframe control interface; the human-computer interaction subsystem obtains a user's touch input command from the display touch system, or through the remote communication system Get the user's far from the backend server input the command.

Preferably, the autonomous driving control subsystem obtains three-dimensional position information of the user target from the user identification subsystem in real time in the user following mode, and sets an autonomous driving path from the current position of the device to the current position of the user target; And the traveling drive system is controlled to travel according to the autonomous driving path through the airframe control interface.

Preferably, the target stereoscopic perception subsystem obtains the left-view video image and the right-view video image from a dual camera, and extracts an edge of each target region from the background by using a gradient-based edge detection algorithm to obtain an edge of the target region. The circumscribed rectangle obtains a center point of the circumscribed rectangle; establishes a feature vector starting from the center point and ending each edge pixel of the target area, and collecting all the feature vectors into a feature vector group as a target feature.

Preferably, the target stereo perception subsystem obtains plane coordinates (x, y) of the center point of each target area in the video picture, and can estimate a certain target distance in the video picture according to the disparity of the left and right view video pictures. The depth coordinate z of this device. The calculation method of the depth coordinate z is:

Where z is the depth coordinate of the pixel point, b is the distance between the center of the imaging plane of the left-view camera and the right-view camera, and f is the focal length of the left-view camera and the right-view camera, the focal lengths of the two are equal, and dis corresponds to The offset of the pixel at the same position of the target in the horizontal direction.

Preferably, the autonomous driving control subsystem controls the system to stop traveling and issue an alarm prompt according to the current position where the positioning system is located for the device and determines that the device has reached the boundary of the maximum area range that is allowed to travel.

Preferably, the human-machine interaction system obtains an unlock password from the background server through the remote communication system in the fixed-point waiting mode, and verifies that the password input by the user through the display touch system matches the unlock password algorithm; The interface controls the electronic lock to unlock the cover.

The present invention further provides an artificial intelligence shopping system applied to a smart business circle, which comprises: an artificial intelligence shopping device, a background server, and a user intelligent terminal; the artificial smart shopping device accesses the background server; The intelligent terminal is connected to the background server;

The user sends a lease request carrying the serial number of an artificial intelligence purchase device to the background server through the user smart terminal;

The background server binds the user to the artificial intelligence shopping device according to the lease request, and then issues an activation command to the corresponding artificial intelligence shopping device;

After the artificial intelligence shopping device is activated according to the command of the background server, the user's screen is first captured by the dual camera and the user feature of the user is extracted therefrom;

In the user following mode, the artificial intelligence shopping device follows the user's travel, receives the goods input by the user, and registers the goods to the unpaid item list; uploads the unpaid item list to the background server according to the checkout instruction input by the user, and from the background server Receiving the payment two-dimensional code, and then displaying the payment two-dimensional code; the user intelligent terminal scanning the payment two-dimensional code to complete the payment; in the fixed-point waiting mode, the artificial intelligence shopping device locks the cover to register the carried item, and The unlock password is received from the background server and compared with the password entered by the user; if it matches, the cover is unlocked.

(3) Beneficial effects

The artificial intelligence shopping device and the system thereof of the invention can independently drive low-speed travel, and have the functions of path navigation automatic driving, obstacle avoidance, user following, and item preservation, which can help users who are active in the smart business circle to perform quick checkout, The function of carrying and storing items. The artificial intelligence shopping device fully combines artificial intelligence, mobile payment and other technical means, and can improve the experience and improve efficiency for the whole process of shopping and leisure activities of the large business circle.

DRAWINGS

1 is a schematic overall structural diagram of an artificial intelligence shopping device according to the present invention;

2 is a schematic structural diagram of a main body of an artificial intelligence shopping device according to the present invention;

3 is a bottom plan view showing the structure of a traveling drive system of the artificial intelligence shopping device of the present invention;

4 is a schematic diagram of a shaft connection of a traveling drive system of an artificial intelligence shopping device according to the present invention;

5 is a schematic diagram of a central intelligent system of an artificial intelligence shopping device according to the present invention;

FIG. 6 is a schematic structural diagram of an artificial intelligence shopping system according to the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, but not all embodiments.

The examples of the embodiments are illustrated in the drawings, wherein the same or similar reference numerals indicate the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the drawings are intended to be illustrative of the invention and are not to be construed as limiting.

The invention provides an artificial intelligence shopping device applied to a smart business circle. The artificial intelligence shopping device can independently drive low-speed travel, and has the functions of path navigation automatic driving, obstacle avoidance, user following, and item preservation, and can help the users who are active in the smart business circle to carry and store the articles.

FIG. 1 is a schematic diagram of the overall structure of an artificial intelligence shopping device according to the present invention. The artificial intelligence shopping device comprises: a main body body, a traveling drive system, a body control interface, an external sensing system, a central intelligent system, a remote communication system, a positioning system, and a display touch system. Wherein, the external sensing system comprises a dual camera and an anti-collision radar.

The structure of the main body is as shown in FIG. 2 , and includes an article carrying case 201 , a chassis 202 , a cover 203 , and an electronic lock 204 . The side wall of the side wall of the article carrying case 201 may be mounted with at least one of the front, rear, left and right directions, and the anti-collision radar 205 may be installed in more directions. The display unit 207 is mounted on the outer wall of the side of the box. The display touch system 207 is a liquid crystal display and a capacitive touch panel. The inside of the article carrying case 201 forms a closed space, which is only open on the top surface of the box for accommodating the items placed by the user in the box; the article carrying case 201 is mounted on the inner wall of the side of the box whose height is close to the top surface There is an electronic tag reader 208 that can read information contained in an electronic tag attached to an optional item placed in the case. The cover body 203 is rotatably mounted on the article carrying case 201 by the hinge pivot 209. By rotating, the user can fasten the open top surface of the box body by the cover body 203 or open the cover body 203. . The cover body 203 has a latch 210 at an end opposite to the hinge pivot 209. The electronic lock 204 is mounted on the inner wall of the side of the box body adjacent to the top surface of the article carrying case 201. The electronic lock 204 is provided with a lock hole 211. The inside of the keyhole has an induction circuit and a latch. When the cover 203 is fastened, the lock 210 is inserted into the lock hole 211 of the electronic lock 204. The sensing circuit inside the lock hole senses that the lock 210 is inserted to cause the lock to be ejected. The lock 210 is inserted to realize the locking of the cover 203. After the lock, the electronic lock 204 receives the unlocking command from the body control interface to retract the lock to unlock the cover 203. The chassis 202 is located at the bottom of the article carrying case 201. The chassis 202 is mounted with a traveling drive system and an electronic motherboard of the device. The body control interface, the central intelligent system, the remote communication system, and the positioning system are all integrated in the electronic motherboard. on.

The travel drive system is used to drive the device to travel autonomously and to adjust the travel speed and direction. 3 shows a bottom view of the structure of the travel drive system, including an X-shaped bracket 301, a first rear wheel 302, a second rear wheel 303, a drive motor 304, a drive rod 305, a first front wheel 306, and a second front wheel. 307, a steering drive motor 308, a steering control shaft 309; two front wheels and two rear wheels are respectively mounted on the respective branches of the X-shaped bracket 301, wherein the two rear wheels are connected to the driving rod 305, and are driven by the driving motor 304. The rod 305 rotates to drive the two rear wheels to rotate to power the autonomous travel of the device; the drive motor 304 is fixedly mounted on the X-shaped bracket 301; the body of the steering drive motor 308 is fixedly mounted in the chassis 202, and The motor shaft of the steering drive motor 308 is coupled to the steering control shaft 309. As shown in FIG. 4, the motor shaft of the steering drive motor 308 and the steering control shaft 309 are connected in a T-shape, and the steering control shaft 309 connects the two front wheels. 306 and 307, and the two branches of the

front wheels

306, 307 and the X-shaped bracket 301 are connected by the universal joint 310, so that the steering control shaft 309 can be driven as the motor shaft output torque of the steering drive motor 308 is outputted.

Front wheel

306, 307 adjusts one The angle is set to adjust the direction of travel of the device. In addition to the above structure, the X-shaped bracket 301 has the function of supporting the main body of the apparatus and being fixed to the chassis. Wherein, the driving motor 304 controls the rotation speed of the rotation, the stationary and the rotation according to the speed control signal of the body control interface, thereby driving the running of the device and the traveling speed; the rotation angle of the steering drive motor 308 is also received. Control of the direction control signal of the body control interface.

The remote communication system can access the Internet based on the WIFI, 3G, and 4G protocols and connect to the background server of the system, and the remote communication system can also access the wireless network covered by the system dedicated base station deployed in the smart business circle, and pass the A dedicated wireless network is connected to the back-end server; the remote communication system enables the device to communicate with the back-end server in two-way communication.

The positioning system utilizes GPS and positioning technology applicable to the interior to realize real-time positioning of the device.

The central intelligent system is configured to set an autonomous driving path based on a user position captured by the dual camera, and control the traveling driving system to travel according to the autonomous driving path through the body control interface, the traveling real-time picture according to the dual camera and the anti-collision radar. Obstacle detection, adjusting the course of travel. The central intelligent system also controls the electronic identity of the user to control the identity of the user and then controls the electronic locker to unlock the cover body through the body control interface. The central intelligent system also obtains a list of unpaid items carried by the device, thereby facilitating the user's quick checkout. The central intelligent system communicates with the smart business circle back-end server or the user mobile device through the remote communication system, and acquires the real-time location of the device through the positioning system.

As shown in FIG. 5, the central intelligent system specifically includes the following subsystems: a target stereo perception subsystem, a user identification subsystem, an autonomous driving control subsystem, and a human-computer interaction subsystem.

The human-computer interaction subsystem is configured to obtain a user's touch input command from the display touch system, or obtain a user's remote input command from a background server through a remote communication system, where the remote input command is used by the device to register the user on the smart terminal. The APP enters and uploads to the background server, and the background server transfers the remote input command to the device.

The dual camera of the external sensing system includes a left view camera and a right view camera for respectively capturing a left view video picture and a right view video picture in front of the device. The target stereo perception subsystem obtains the left view video picture and the right view video picture from the dual camera, and extracts each target in the video picture by using the object recognition algorithm based on edge detection and vector information. Firstly, using the gradient-based edge detection algorithm, extract the edge of each target area from the background, obtain the circumscribed rectangle of the edge of the target area, and obtain the center point of the circumscribed rectangle; establish the center point as the starting point and the target area Each edge pixel is a feature vector of the end point, and all feature vectors are grouped into a feature vector group as a target feature. The plane coordinates (x, y) of the center point of each target area in the video picture are obtained, and the depth coordinate z of a certain target distance of the device in the video picture can be estimated according to the disparity of the left and right view video pictures. The calculation method of the depth coordinate z is:

Where z is the depth coordinate of the pixel point, b is the distance between the center of the imaging plane of the left-view camera and the right-view camera, and f is the focal length of the left-view camera and the right-view camera, the focal lengths of the two are equal, and dis corresponds to The offset of the pixel at the same position of the target in the horizontal direction. Thus, the target stereo perception subsystem takes the coordinates (x, y, z) as the target three-dimensional position information.

The user identification subsystem is configured to compare each target extracted by the target stereo perception subsystem with a pre-captured and stored user feature in a user following mode, that is, a feature vector group and pre-storage of each target area. The user feature vector groups are compared, and the feature vector consistency is determined, thereby identifying the user target from all the extracted targets, and determining the three-dimensional position information of the user target. When the user wants to rent the device, the user can use the APP on the smart terminal to scan the two-dimensional code printed on the device or displayed by the touch control system to obtain the serial number of the device, so that the rental request carrying the serial number will be carried. Uploaded to the background server; the background server issues an activation command to the device according to the lease request; when the device is activated from the idle state according to the command of the background server, the user recognition subsystem uses the dual camera to capture the user image and extracts therefrom A feature vector group starting from the center point of the user screen area and ending with each edge pixel of the user screen area is used as a user feature of the user for subsequent comparison recognition.

In the user following mode, the autonomous driving control subsystem obtains the current position information of the user target from the user identification subsystem in real time, that is, (x, y, z), and then sets the current position of the device from the current position of the device to the current position of the user target. The autonomous travel path is controlled; and the travel drive system is controlled to travel according to the autonomous travel path through the airframe control interface.

During the travel process, the device can utilize the autopilot navigation technology that has been matured in the prior art to achieve speed adjustment, obstacle avoidance, and path correction. For example, if it is detected that there is an obstacle on the travel path according to the travel real-time picture captured by the dual camera and the obstacle detection of the anti-collision radar, the device is controlled to stop traveling, wait for the obstacle to disappear, continue to travel, or be corrected by the autonomous travel control subsystem. path. The autonomous driving control subsystem also has the function of limiting the driving area of the device, that is, setting the maximum area range that the device is allowed to travel around the smart business circle, and determining that the device has reached the allowed driving according to the current position of the positioning system of the device. The boundary of the largest area range controls the system to stop traveling and issue an alarm prompt.

In the user follow mode, the user himself can also put the selected item into the item carrying case of the device, and the electronic tag reader in the box reads and registers the item into the unpaid item list; then, the user closes The cover body inputs a quick checkout instruction to the device through the display touch system, and the human-machine interaction subsystem responds to the instruction, and uploads the unpaid item list to the background server through the remote communication system; the background server aggregates the fee payable according to the unpaid item list. And generating a payment two-dimensional code; the background server sends the payment two-dimensional code to the device; after the human-computer interaction subsystem of the device receives the payment two-dimensional code through the remote communication system, displaying the same on the display touch system The two-dimensional code is paid, and the user uses the smart terminal to perform the scan code payment.

In the user follow mode, the user can also close and lock the cover, and instruct the device to enter the fixed-point waiting mode by displaying the touch system, and then engage in his own activities. The device will wait for the user in place. The user can use this time to engage in entertainment activities such as watching movies, fitness, eating, drinking coffee, etc. After the event ends, the waiting place is controlled to control the artificial intelligence shopping device to return to the user following mode, so that the user can purchase the item or The item that comes with the user is placed in the device. When the user wants to take out the items stored in the artificial intelligence shopping device, he may request an unlock password from the background server of the system; the background server simultaneously sends the unlock password to the artificial smart shopping device bound by the user; The password can be input by displaying the touch system, and the human-computer interaction subsystem verifies the password to unlock, so that the user can take out the registered items.

As shown in FIG. 6, the operator can place a certain number of artificial intelligence shopping devices of the present invention in the space of the smart business circle, and deploy a background server, and the artificial smart shopping device connects to the background server by accessing the Internet, or is based on The wireless network dedicated to this system is connected to the backend server. The user can install the APP of the system on a smart terminal such as a mobile phone, and connect to the background server of the system by enabling the APP.

When the user wants to rent, the user can use the APP on the smart terminal to scan the serial number of the artificial smart shopping device printed on the artificial intelligence shopping device of the present invention or displayed by the display touch system. Therefore, the lease request carrying the serial number is uploaded to the background server; the background server binds the user to the artificial smart shopping device in the background management system according to the lease request, and then issues an activation command to the corresponding artificial smart shopping device; After the artificial intelligence shopping device is activated according to the command of the background server from the idle state, the user's screen is first captured by the dual camera and the user characteristics of the user are extracted therefrom. In turn, the artificial intelligence shopping device enters the user follow mode, with the user acting in the smart business circle. When the user selects an item in the store, the item can be put into the item carrying case of the artificial intelligence shopping device, and the item is registered by the item management subsystem of the artificial intelligence shopping device to the unpaid item list; the artificial smart shopping device The display touch system can display the details, unit price and total price of all unpaid items listed in the device. After the user completes the purchase in the store, the display touch system can input the checkout instruction to the artificial intelligence shopping device, the device uploads the unpaid item list through the remote communication system, and receives the payment QR code from the background server, and then passes The human-computer interaction subsystem displays the payment two-dimensional code, and the user can complete the payment by scanning the payment two-dimensional code. The device can be switched to a fixed-point waiting mode to lock the cover to register the carried item and to remain in the waiting position until the user comes to pick up the purchased item.

It can be seen that the artificial intelligence shopping device and the system thereof of the invention can independently drive low-speed travel, and have the functions of path navigation automatic driving, obstacle avoidance, user following, and item preservation, which can help users who are active in the smart business circle to perform fast. Checkout, item bearing, and depositing functions. The artificial intelligence shopping device fully combines artificial intelligence, mobile payment and other technical means, and can improve the experience and improve efficiency for the whole process of shopping and leisure activities of the large business circle.

In the description of the present specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like means a specific feature described in connection with the embodiment or example. A structure, material or feature is included in at least one embodiment or example of the invention. In the present specification, the schematic representation of the above terms does not necessarily mean the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples.

The above is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any technical person skilled in the art within the technical scope disclosed by the present invention, the technical solution according to the present invention Equivalent substitutions or modifications of the inventive concept are intended to be included within the scope of the invention.

Claims

An artificial intelligence shopping device, comprising: a main body body, a traveling drive system, a body control interface, an external sensing system, a central intelligent system, a remote communication system, a positioning system, a display touch system; wherein the external The sensing system includes: dual camera, anti-collision radar;

The main body is for providing a space for accommodating the carried articles, and can control locking and unlocking the cover located on the top surface of the space, and for mounting the traveling drive system;

The traveling drive system is used to drive the device to autonomously travel, and adjust the traveling speed and direction;

The airframe control interface is configured to send an unlocking instruction to the main body to unlock the cover, and send an instruction to the traveling drive system to control its travel, stop, and speed;

The remote communication system is configured to connect to a background server and perform two-way communication with the background server;

The positioning system is configured to implement real-time positioning of the device;

The dual camera of the external sensing system is used for capturing a left-view video picture and a right-view video picture in front of the device; the anti-collision radar of the external sensing system is configured to detect whether there is an obstacle within a certain distance in the direction around the device;

The display touch system is a liquid crystal display and a capacitive touch panel;

The central intelligent system sets an autonomous driving path based on the user position captured by the dual camera, and controls the traveling driving system to travel according to the autonomous driving path through the body control interface, and the traveling real-time picture and the obstacle detection of the anti-collision radar according to the dual camera shooting Adjusting the process of traveling; the central intelligent system also controls the electronic locker to unlock the cover body through the human body interaction of the user through the body control interface; and is used to obtain a list of unpaid items carried by the device, and the unpaid amount is The item list is uploaded to the background server through the remote communication system; the payment QR code generated according to the unpaid item list is received from the background server, and the payment QR code is displayed on the display touch system for the user to scan the payment.
The artificial intelligence shopping device according to claim 1, wherein the main body comprises: an article carrying case, a chassis, a cover, and an electronic lock; and the outer side wall of the case of the article carrying case is mounted with the anti-theft Collision radar; a double camera is installed on the outer side wall of the box body, and a touch control system is further mounted on the outer wall of the side of the box; an electronic label reader is installed on the inner wall of the side of the box of the article carrying box for reading into the box The electronic tag attached to the body product contains information; the cover body is rotatably mounted on the article carrying case by a hinge pivot; the cover body has a latch at an end opposite to the hinge pivot, and the electronic locker Mounted on the inner wall of the side of the box of the article carrying case; when the cover is fastened, the electronic lock locks the cover by the lock; and the electronic lock is sent according to the body control interface The unlocking command unlocks the cover; the chassis is located at the bottom of the article carrying case, and the traveling drive system is mounted on the chassis.
The artificial intelligence shopping device according to claim 1, wherein the travel drive system comprises an X-shaped bracket, a first rear wheel, a second rear wheel, a drive motor, a drive rod, a first front wheel, and a second front Wheel, steering drive motor, steering control shaft; two front wheels and two rear wheels are respectively mounted on each branch of the X-shaped bracket, wherein two rear wheels are connected to the driving rod, and the driving motor drives the driving rod to rotate, thereby driving Two rear wheels rotate to power the autonomous travel of the device; the drive motor is fixedly mounted on the X-shaped bracket; the motor shaft of the steering drive motor is connected to the steering control shaft, and the steering control shaft connects the two front wheels, and two The two branches of the front wheel and the X-shaped bracket are connected by a universal joint. With the motor shaft output torque of the steering drive motor, the steering control shaft drives the two front wheels to adjust a certain angle, thereby adjusting the travel of the device. direction.
The artificial intelligence shopping device according to claim 1, wherein the central intelligent system comprises: a target stereoscopic perception subsystem, a user identification subsystem, an autonomous driving control subsystem, a human-computer interaction subsystem; and a target stereoscopic perception. Obtaining the left view video picture and the right view video picture from the dual camera, extracting each target in the video picture by using a target recognition algorithm based on edge detection and vector information, and determining three-dimensional position information of each target; the user identification The subsystem is configured to compare each target extracted by the target stereo perception subsystem with a user feature collected and stored in advance, identify a user target from all the extracted targets, and determine three-dimensional position information of the user target; The driving control subsystem sets the path that the device intends to autonomously travel according to the three-dimensional position information of the user target, and controls the traveling driving system to travel according to the autonomous driving path through the body control interface; the human-computer interaction subsystem is in the slave display touch system. Get the user's touch input command, or pass The remote communication system obtains the user's remote input command from the background server.
The artificial intelligence shopping device according to claim 4, wherein the autonomous driving control subsystem obtains three-dimensional position information of the user target from the user identification subsystem in real time in the user following mode, and sets the device from the device. The current position points to the autonomous travel path of the current position of the user target; and the travel drive system is controlled to travel according to the autonomous travel path through the airframe control interface.
The artificial intelligence shopping device according to claim 4, wherein the target stereoscopic perception subsystem obtains the left-view video image and the right-view video image from a dual camera, using a gradient-based edge detection algorithm from the background Extracting the edge of each target area, obtaining the circumscribed rectangle of the edge of the target area, obtaining the center point of the circumscribed rectangle; establishing a feature vector starting from the center point and ending with each edge pixel of the target area, all the features Vector collections are feature vector groups that serve as target features.
The artificial intelligence shopping device according to claim 4, wherein the target stereoscopic perception subsystem obtains plane coordinates (x, y) of a center point of each target area in a video picture, and can be based on left and right viewing angles. The parallax of the video picture estimates the depth coordinate z of a certain target in the video picture from the device. The calculation method of the depth coordinate z is:

Where z is the depth coordinate of the pixel point, b is the distance between the center of the imaging plane of the left-view camera and the right-view camera, and f is the focal length of the left-view camera and the right-view camera, the focal lengths of the two are equal, and dis corresponds to The offset of the pixel at the same position of the target in the horizontal direction.
The artificial intelligence shopping device according to claim 4, wherein the autonomous driving control subsystem determines that the device has reached the boundary of the maximum range of the allowed driving according to the current position of the positioning system of the device, and then controls The system stops traveling and issues an alarm.
The artificial intelligence shopping device according to claim 4, wherein the human-machine interaction system obtains an unlock password from the background server through the remote communication system in the fixed-point waiting mode, and verifies the password entered by the user through the display touch system. The unlocking password algorithm is matched; if it matches, the electronic locker is controlled by the body control interface to unlock the cover.
An artificial intelligence shopping system applied to a smart business circle, comprising: an artificial intelligence shopping device, a background server, and a user intelligent terminal; the artificial intelligence shopping device accesses the background server; the user intelligent terminal is connected to the background server;

Sending, by the user intelligent terminal, a rent request carrying a serial number of an artificial intelligence shopping device to the background server;

The background server binds the user to the artificial intelligence shopping device according to the lease request, and then issues an activation command to the corresponding artificial intelligence shopping device;

After the artificial intelligence shopping device is activated according to the command of the background server, the user's screen is first captured by the dual camera and the user feature of the user is extracted therefrom;

In the user following mode, the artificial intelligence shopping device follows the user's travel, receives the goods input by the user, and registers the goods to the unpaid item list; uploads the unpaid item list to the background server according to the checkout instruction input by the user, and from the background server Receiving the payment two-dimensional code, and then displaying the payment two-dimensional code; the user intelligent terminal scanning the payment two-dimensional code to complete the payment; in the fixed-point waiting mode, the artificial intelligence shopping device locks the cover to register the carried item, and The unlock password is received from the background server and compared with the password entered by the user; if it matches, the cover is unlocked.