1. An AI-assisted teaching robot, comprising: the upper surface center of the base is movably provided with an upper mounting table-board through a lifting mechanism, the modes of the upper mounting table-board for placing the display equipment at least comprise flat placement, fixed embedding and detachable embedding, the rear side of the upper surface of the base is provided with a laser radar, and the front side in the base is provided with a drop-proof sensor, the base is also internally provided with a control processing module, a communication module, a power supply module and an ultrasonic sensor which are respectively connected with the control processing module, the laser radar and the anti-falling sensor are respectively connected with the control processing module, the control processing module is also electrically connected with two AI cameras, the AI camera is arranged at the rear side of the upper mounting table-board, the AI camera adopts a three-mesh camera, the laser radar, the anti-falling sensor, the communication module, the ultrasonic sensor and the AI camera are all connected with the power module.

2. The AI-assisted teaching robot of claim 1 wherein the AI camera employs a multi-view camera.

3. The AI auxiliary teaching robot according to claim 1, wherein driving wheels are rotatably mounted at the centers of the left and right sides of the bottom surface of the base, a driving motor is mounted on each of the driving wheels in a transmission manner, the driving motor is fixedly mounted in the base, the driving motor is connected with the control processing module and the power module, auxiliary wheels are disposed at four corners of the bottom surface of the base, and universal wheels are disposed at the centers of the front and rear sides of the bottom surface of the base.

4. The AI auxiliary teaching robot of claim 3, wherein an emergency stop switch is connected between the drive motor and the power module, the emergency stop switch being mounted on a rear side of an upper surface of the upper mounting deck.

5. The AI teaching assistance robot of claim 1, wherein the lifting mechanism comprises a support rod fixedly mounted at the center of the upper surface of the base, a placing table fixedly mounted at the top end of the support rod, and an electric lifting rod fixedly mounted at the center of the top surface of the placing table and connected with the upper mounting table, a slide rail is mounted on the side wall of the electric lifting rod, a stabilizing slide block is slidably mounted on the slide rail up and down, an inner cylinder is fixedly mounted outside the stabilizing slide block, the inner cylinder is fixedly connected with the upper mounting table, an outer cylinder is slidably mounted outside the inner cylinder, and the outer cylinder is fixedly mounted on the placing table.

6. The AI auxiliary teaching robot of claim 1, wherein the power module is further connected with a USB socket and a power socket, both of which are mounted on a rear side of the upper mounting table.

7. The AI auxiliary teaching robot of claim 1, wherein a charging port is further connected to the power module, the charging port is installed at a rear side of the base, and a power detection sensor is further connected to the control processing module.

8. The AI auxiliary teaching robot of claim 1, wherein an access door is provided to open and close at a position on the upper surface of the base directly above the power module.

9. The AI auxiliary teaching robot of claim 1, wherein broadcasters are provided on both the left and right sides of the upper surface of the upper mounting table, and the broadcasters are respectively connected to the power module and the control processing module.

10. The AI-assisted teaching robot of claim 1, wherein a visual recognition algorithm, an indoor navigation algorithm, a path planning algorithm, a motion control algorithm, an obstacle-avoiding and obstacle-avoiding algorithms, and a seq2seq source-opening framework are implanted in the control processing module.

11. The AI-assisted teaching robot of claim 10, wherein the visual recognition algorithm consists of yolov5 algorithm, IOU algorithm, and a number of residual networks.

12. The AI-assisted teaching robot of claim 10, wherein the psycho-directed dialog function of the AI-assisted teaching robot is implemented by a seq2seq open source framework.

Background

In the prior campus environment, a teacher often needs to hold a lot of complex articles such as teaching aids, operation and computers in the teaching process; in the teaching and answering process, the fixed platform is often required to be returned for answering and playing; in the teaching process, when an emergency occurs, the students need to leave the classroom temporarily, and the students cannot monitor and communicate remotely for self study, so that potential risks exist; at present, the classroom environment consists of a traditional podium which cannot be moved and a portable podium which can be pushed by hand, even if the intelligent podium is declared on the market, a plurality of display and voice interfaces are integrated to realize the functions of an electronic screen and sound amplification, but the problem of moving from an office to a classroom is still not solved; the indoor single camera cannot perform voice interaction, and all devices cannot be controlled in an interconnected mode and need to be controlled independently; in the teaching process of teachers, the degree of knowledge mastered by all students cannot be accurately judged, and the judgment on the teaching effect is insufficient; the number of students is more, uncertain potential risk is more, falls down occasionally, conflict etc. takes place, and the mr is because of not being on-the-spot, and the reaction is not in time, and student's personal safety is guaranteed quality and is slightly less than apparent.

Disclosure of Invention

The invention aims to provide an AI auxiliary teaching robot which helps teachers to finish teaching notes and books to be carried automatically according to daily teaching application scenes of the teachers, and meanwhile assists the teachers to better perform classroom teaching, remote teaching and remote monitoring.

In order to solve the above technical problem, the present invention provides an AI-assisted teaching robot, comprising: the base, there is the facial make-up mesa at base upper surface center through elevating system movable mounting, just the mode that the display device was laid to the facial make-up mesa is at least including keeping flat, fixed embedded and can dismantle embedded, base upper surface rear side is provided with laser radar, just the inside front side of base is provided with the dropproof sensor, still be provided with control processing module in the base and communication module, power module and the ultrasonic sensor be connected with control processing module respectively, laser radar and dropproof sensor are connected with control processing module respectively, control processing module still electricity is connected with two AI cameras, the AI camera is installed at facial make-up rear side, laser radar, dropproof sensor, communication module, ultrasonic sensor and AI camera all are connected with power module.

Furthermore, the AI camera adopts the multi-view camera.

Further, the center of the left side and the center of the bottom surface of the base are both rotated to be provided with driving wheels, the driving wheels are both installed with driving motors in a transmission mode, the driving motors are fixedly installed in the base and are connected with the control processing module and the power supply module, auxiliary wheels are arranged at four corners of the bottom surface of the base, and universal wheels are arranged at the centers of the front side and the rear side of the bottom surface of the base.

Furthermore, an emergency stop switch is connected between the driving motor and the power supply module, and the emergency stop switch is installed on the rear side of the upper surface of the upper installation table board.

Further, elevating system includes that fixed mounting is at the bracing piece of base upper surface center, fixed mounting is at the platform of placing on bracing piece top and fixed mounting place the platform top surface center and with the electric lift pole of being connected of facial features, the slide rail is installed to electric lift pole lateral wall, slidable mounting has firm slider about the slide rail, the outer fixed mounting of firm slider has the inner tube, inner tube and facial features of features.

Furthermore, the power module is further connected with a USB socket and a power socket, and the USB socket and the power socket are both arranged on the rear side of the upper mounting table board.

Furthermore, power module still is connected with the mouth and the electric quantity detection sensor that charge, the mouth that charges is installed in the base rear side, the electric quantity detection sensor still is connected with control processing module.

Furthermore, the upper surface of the base is provided with an access door which is opened and closed at a position right above the power module.

Furthermore, the left side and the right side of the upper surface of the upper mounting table board are respectively provided with a broadcaster, and the broadcasters are respectively connected with the power module and the control processing module.

Furthermore, a visual identification algorithm, an indoor navigation algorithm, a path planning algorithm, a motion control algorithm, an obstacle avoidance algorithm and a seq2seq source-opening framework are implanted in the control processing module.

Further, the visual recognition algorithm consists of yolov5 algorithm, IOU algorithm and several residual error networks.

Furthermore, the mind grooming conversation function of the AI auxiliary teaching robot is realized through a seq2seq open source framework.

Compared with the prior art, the invention has the following advantages:

(1) according to the invention, based on data acquisition equipment such as a laser radar, an ultrasonic sensor and an AI high-definition camera, through technologies such as a visual identification algorithm, an indoor navigation algorithm, a path planning algorithm, a motion control algorithm, a barrier-avoiding algorithm and the like, the product can complete self-walking and autonomous following functions in an indoor teaching scene, a school or other stepless scenes, and the moving problem of a platform is solved; through the mobility of the classroom desk, the space utilization rate of each classroom is greatly improved, the classroom is more functional, and teachers do not need to carry heavy articles such as textbooks, homework and teaching aids;

(2) the invention judges the classroom listening effect of students through behavior recognition and algorithm learning, autonomously reminds teachers whether to carry out additional supplementary teaching, autonomously recommends books or extraclass reading corresponding to teaching, contains an intelligent voice interaction system, can autonomously answer questions and interact, supports communication interconnection, is accessed to a school or other education self-owned servers, helps the teachers to improve the teaching effect, assists the teachers to teach the questions, and enables the teachers to go to the classes more easily and more effectively;

(3) according to the invention, whether students are in a potential risk area or action is automatically identified through a computer vision technology, teachers are fed back in real time through remote communication to check or remotely talk, and the students in the risk area or action are reminded by voice, so that the safety monitoring and reminding early warning functions of the students are realized.

(4) According to the invention, the seq2seq open source framework is implanted into the system, so that the mind dispersion dialogue function of the AI-assisted teaching robot is effectively realized, and the teaching effect is further improved.

Drawings

FIG. 1 is a schematic view of the overall structure of an AI-assisted teaching robot according to the present invention;

FIG. 2 is a schematic view of an electric lifting rod mounting structure of the AI assisted teaching robot according to the present invention;

FIG. 3 is a top view of the overall structure of the AI-assisted teaching robot of the present invention;

FIG. 4 is an overall circuit block diagram of the AI-assisted teaching robot of the present invention;

FIG. 5 is a schematic structural view of the AI auxiliary teaching robot according to the present invention with a simple top;

FIG. 6 is a schematic structural diagram of the AI auxiliary teaching robot according to the present invention, in which the touch display screen is fixedly mounted in an embedded manner;

FIG. 7 is a schematic structural diagram of the AI-assisted teaching robot according to the present invention with an embedded tablet computer on the top-loading platform;

fig. 8 is a schematic diagram of a seq2seq open source framework of the AI-assisted teaching robot according to the present invention.

Detailed Description

The AI-assisted teaching robot of the present invention will now be described in greater detail with reference to the accompanying schematic drawings, in which preferred embodiments of the invention are shown, it being understood that one skilled in the art may modify the invention herein described while still achieving the advantageous effects of the invention. Accordingly, the following description should be construed as broadly as possible to those skilled in the art and not as limiting the invention.

The invention is described in more detail in the following paragraphs by way of example with reference to the accompanying drawings. Advantages and features of the present invention will become apparent from the following description and from the claims. It is to be noted that the drawings are in a very simplified form and are not to precise scale, which is merely for the purpose of facilitating and distinctly claiming the embodiments of the present invention.

As shown in fig. 1 to 4, an embodiment of the present invention provides an AI-assisted teaching robot, including: a base 1, an upper mounting table-board 3 is movably arranged at the center of the upper surface of the base 1 through a lifting mechanism 2, and the modes of the upper mounting table-board 3 for placing the display equipment at least comprise flat placement, fixed embedding and detachable embedding, the rear side of the upper surface of the base 1 is provided with a laser radar 4, and the front side in the base 1 is provided with a falling-prevention sensor 5, the base 1 is also internally provided with a control processing module 6, a communication module 7, a power module 8 and an ultrasonic sensor 9 which are respectively connected with the control processing module 6, the laser radar 4 and the anti-falling sensor 5 are respectively connected with a control processing module 6, the control processing module 6 is also electrically connected with two AI cameras 10, AI camera 10 is installed at the facial features 3 rear side of facial features, lidar 4, dropproof sensor 5, communication module 7, ultrasonic sensor 9 and AI camera 10 all are connected with power module 8.

In one embodiment, the AI camera 10 is a multi-view camera, such as a binocular camera, a trinocular camera, and the like, and those skilled in the art can select the cameras according to actual needs.

Specifically, the upper table-board 3 can include three forms, the first is a simple table-board, as shown in fig. 5; the second is that a touch display screen 29 is embedded and fixedly installed on the upper mounting table board 3, and teaching demonstration is performed by controlling the touch display screen 29, so that the teaching effect is improved, as shown in fig. 6; the third is to embed the tablet pc 30 on the upper table top 3, so that the tablet pc 30 can be taken out, and the tablet pc 30 can also be placed on the upper table top 3 for charging and operation, so as to cooperate with teaching demonstration, as shown in fig. 7.

All rotate in base 1 bottom surface left and right sides center and install drive wheel 11, two drive wheel 11 equal transmission installs driving motor 12, driving motor 12 fixed mounting is in base 1, driving motor 12 is connected with control processing module 6 and power module 8, 1 bottom surface four corners of base all is provided with auxiliary wheel 13, just both sides center all is provided with universal wheel 14 around the 1 bottom surface of base. In the embodiment, the linear motion of the base 1 is realized by controlling the two driving motors 12 to move at the same speed, and the curvilinear motion of the base 1 is realized by driving the two driving motors 12 to move at different speeds in cooperation with the auxiliary wheels 13 and the universal wheels 14, so that the self-walking capability of the base 1 is realized.

Elevating system 2 includes that fixed mounting is at the bracing piece 15 of base 1 upper surface center, fixed mounting is at the platform 16 and the electric lift pole 17 of fixed mounting on the bracing piece 15 top of placing platform 16 top surface center and being connected with facial make-up table 3, slide rail 18 is installed to electric lift pole 17 lateral wall, slide rail 18 is last under sliding mounting has firm slider 19, 19 outer fixed mounting of firm slider has inner tube 20, inner tube 20 and facial make-up table 3 fixed connection, just 20 outer sliding mounting of inner tube has urceolus 21, urceolus 21 fixed mounting is on placing platform 16. In this embodiment, the electric lifting rod 15 is controlled to lift, so as to lift the upper table top 3, and the stability of the upper table top 3 in the lifting process is improved by the arrangement of the slide rail 18 and the stable slide block 19.

An emergency stop switch 22 is connected between the driving motor 12 and the power module 8, and the emergency stop switch 22 is installed on the rear side of the upper surface of the upper installation table board 3. In the present embodiment, the emergency stop switch 22 is provided to realize an emergency stop during the movement of the base 1.

The power module 8 is also connected with a USB socket 23 and a power socket 24, and the USB socket 23 and the power socket 24 are both arranged on the rear side of the upper mounting table board 3. In this embodiment, the USB socket 23 and the power outlet 24 are provided to facilitate charging of the mobile device or other devices.

The power module 8 is further connected with a charging port 25 and an electric quantity detection sensor 26, the charging port 25 is installed on the rear side of the base 1, and the electric quantity detection sensor 26 is further connected with the control processing module 6. In the present embodiment, the electric quantity sensor 26 is provided to monitor the battery capacity in real time.

An access door 27 is arranged on the upper surface of the base 1 and is opened and closed at a position right above the power module 8. In this embodiment, the access door 27 is provided to facilitate battery replacement.

The upper surface left and right sides of facial make-up mesa 3 all is provided with broadcaster 28, broadcaster 28 is connected with power module 8 and control processing module 6 respectively. In the present embodiment, the provision of the broadcaster 28 realizes the broadcasting function of the apparatus.

And a visual identification algorithm, an indoor navigation algorithm, a path planning algorithm, a motion control algorithm, an obstacle-avoiding algorithm and a seq2seq source-opening frame are implanted in the control processing module 6. In the embodiment, the arrangement of the visual recognition algorithm, the indoor navigation algorithm, the path planning algorithm, the motion control algorithm and the obstacle-crossing and obstacle-avoiding algorithm provides the functions of indoor self-walking, remote monitoring and control and the like for the device; seq2seq is an upgrade of a recurrent neural network RNN and lstm, and is commonly used in service scenes with strong directionality such as natural language processing, machine translation, voice recognition and character recognition, in order to improve model training and reasoning speed, a bucket mechanism is added in seq2seq, and input sentences are calculated after being partitioned into buckets, as shown in FIG. 8, seq2seq consists of a plurality of lstms and can output sentences with different input lengths as dialogue answering sentences, firstly, the lstm structure reduces or enlarges the information amount transmitted by the lstm at the previous time through an information gate, a forgetting gate and an output gate, thereby eliminating the gradient dispersion problem when the neural network is reversely updated; secondly, the seq2seq inverts the input statement and the output statement in terms of architecture, so that the length of the output statement can be different from that of the input statement, the output statement is more consistent with an actual conversation scene, and the psychological leading conversation function of the AI-assisted teaching robot is realized through the setting of the seq2seq open source framework.

The visual recognition algorithm consists of yolov5 algorithm, IOU algorithm and a plurality of residual error networks. In the embodiment, yolov5 is developed on the basis of original yolov3 and yolov4, and compared with the former two versions, the image is subjected to multi-channel segmentation in the input stage of the image, so that various combinations of image feature extraction are increased, and the long tail effect of pixel extraction is eliminated; meanwhile, the residual errors are updated more accurately by matching with a plurality of residual error networks, and the detection precision is improved while the detection speeds of yolov3 and yolov4 are kept by matching with an IOU algorithm.

The following is a list of preferred embodiments of the AI-assisted teaching robot to clearly illustrate the contents of the present invention, and it should be understood that the contents of the present invention are not limited to the following embodiments, and other modifications by conventional technical means of those skilled in the art are within the scope of the idea of the present invention.

The embodiment of the invention provides a using method of an AI auxiliary teaching robot, which comprises the following steps:

(1) self-walking to classroom: in the debugging stage, scanning and mapping by the laser radar 4 are carried out to complete mapping; in the practical application, according to the APP instruction, plan the route to the target classroom, then begin to remove, at the in-process that removes, AI camera 10, ultrasonic sensor 9 and laser radar 4 can constantly 360 discernments around, discernment in advance and judge the obstacle, after discerning the obstacle, slow down according to moving speed and target distance. If the environment allows, how to avoid and block the obstacle is calculated at the same time, when the device is run-in due to sudden obstacle, the scram switch 22 is started to carry out the emergency stop of the device and report prompt voice through the broadcaster 28, and when the device is close to a target classroom, steering control is carried out, wherein the steering method comprises but is not limited to differential steering, and meanwhile, the universal wheels 14 and the auxiliary wheels 13 are matched in a cooperative mode, so that the device finishes steering action within the allowable range of steering radius, and then the device enters the classroom and carries out obstacle avoiding and block avoiding operation when the falling-prevention sensor 5 monitors that the suspension distance of the ground in front is larger than a set value;

(2) the upper mounting table top 3 is lifted: the electric lifting rod 17 is controlled to drive through an APP issuing instruction, so that the upper mounting table top 3 can be lifted, and the upper mounting table top 3 can be lifted through the device body;

AI auxiliary teaching: the AI camera 10 collects emotion and body language of students in class, calculates and scores through a visual recognition algorithm, analyzes class listening effect of the students, autonomously reminds teachers whether to perform additional supplementary teaching, autonomously recommends corresponding tutoring books or extraclass reading for teaching, and transmits teaching scenes to schools or other education self-owned servers through the communication module 7;

(3) AI safety monitoring reminding: the AI camera 10 collects the actions of the students and the areas where the students are located, judges whether the students are in potential dangerous areas or whether the students have dangerous actions such as falling or collision and the like through a visual recognition algorithm, and feeds the actions back to the teacher for on-site voice reminding;

(4) self-circuit charging: when the electric quantity detection sensor 26 detects that the electric quantity of the battery is too low, an instruction is sent to the control processing module 6, and the control device returns to a charging point for automatic charging.

In summary, compared with the prior art, the invention has the following advantages:

(5) according to the invention, based on data acquisition equipment such as a laser radar, an ultrasonic sensor and an AI high-definition camera, through technologies such as a visual identification algorithm, an indoor navigation algorithm, a path planning algorithm, a motion control algorithm, a barrier-avoiding algorithm and the like, the product can complete self-walking and autonomous following functions in an indoor teaching scene, a school or other stepless scenes, and the moving problem of a platform is solved; through the mobility of the classroom desk, the space utilization rate of each classroom is greatly improved, the classroom is more functional, and teachers do not need to carry heavy articles such as textbooks, homework and teaching aids;

(6) the invention judges the classroom listening effect of students through behavior recognition and algorithm learning, autonomously reminds teachers whether to carry out additional supplementary teaching, autonomously recommends books or extraclass reading corresponding to teaching, contains an intelligent voice interaction system, can autonomously answer questions and interact, supports communication interconnection, is accessed to a school or other education self-owned servers, helps the teachers to improve the teaching effect, assists the teachers to teach the questions, and enables the teachers to go to the classes more easily and more effectively;

(7) according to the invention, whether students are in a potential risk area or action is automatically identified through a computer vision technology, teachers are fed back in real time through remote communication to check or remotely talk, and the students in the risk area or action are reminded by voice, so that the safety monitoring and reminding early warning functions of the students are realized.

(8) According to the invention, the seq2seq open source framework is implanted into the system, so that the mind dispersion dialogue function of the AI-assisted teaching robot is effectively realized, and the teaching effect is further improved.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.