A method of guiding

Abstract

The present application relates to a guide method, based on the identity information, position information, route demand of tourists, combined with the position information of the associated walking device, the walking device can provide guide service at any time. And, the direct and indirect interaction between the tourists and the associated walking device can be carried out, the walking device can ensure the effective guidance to the tourists, and the situation that the tourists are lost due to the surrounding environment can be avoided. The present application plans all tourists, and the navigation route optimization of all tourists in the scenic area is carried out, which can improve the tour experience of tourists, relieve the congestion degree of scenic spots, and improve the operation efficiency of the scenic area. The present application carries out navigation route optimization based on the combination of multiple tourists, provides route change suggestions to the tourists combined with as much detailed tourist information as possible, improves the tour experience of tourists based on the navigation route optimization, relieves the congestion degree of scenic spots, and improves the operation efficiency of the scenic area.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of navigation technology, more particularly, to a tour guide method. BACKGROUND

[0002] For large scenic spots, tourists are difficult to grasp the overall situation, terrain, path of the scenic spot, and it is also difficult to grasp the tour route. In order to improve the tour experience of tourists, the scenic spot usually provides a portable explanation device for indicating the approximate position of the tourists, playing the explanation words of the corresponding position, etc. When using, the tourists still need to actively explore the tour route, and the problems of repeated tour, getting lost, missing scenic spots, etc. often occur, and the overall tour experience is poor.

[0003] In the prior art, Chinese patent application 201310715227.X discloses a tour guide method of an intelligent tour guide robot. In the technical solution, the intelligent tour guide robot mainly realizes navigation positioning according to WIFI, and explains by recognizing objects. However, since the WIFI layout involves a large number of network points, it needs to invest a large number of network point layouts for large scene modes. This is only suitable for small-scale scene applications, and cannot realize accurate recognition and positioning and adaptive scene matching for specific scene modes in large scene modes, such as a number of scene pavilions in a scenic spot or scenic spots in a large range.

[0004] In order to solve the above problems, Chinese patent application 201611004729.1 discloses a method for guiding a tour by an intelligent robot. In the technical solution, a route in a tour guide area is set based on a GPS positioning map; an intelligent robot is driven to provide tour guide services according to the set route; a target object picture in the tour guide area is obtained based on a structure similarity visual tracking method on the tour guide service route; it is judged whether the target object picture is constructed with a tour guide explanation scene database; when it is judged that the target object picture is constructed with a tour guide explanation scene database, tour guide explanation scene content is output based on a voice output unit.

[0005] The application patent application based on GPS positioning cannot be applied to any scenic spot with obstruction, and almost cannot be applied to any scenic spot from the real situation; the intelligent robot has no association with the tourists and cannot directly or indirectly interact, and has single function; the intelligent robot identifies target objects through visual technology, which is affected by the surrounding environment and has high requirement on identification technology but low identification precision; the intelligent robot cannot optimize the tour route and relieve the queuing pressure of the scenic spot, the explanation content is fixed condition activation fixed explanation word, has no interaction ability, no adaptability, no ability to understand the real intention of the tourists, cannot solve the scenes and problems without preset answers in advance, cannot consider the route conflict, play project conflict, space and resource conflict among multiple tourists at the same time, cannot form a guide function, and the realized function can only be close to the "read and write machine". SUMMARY

[0006] The purpose of the present application is to overcome the shortcomings of the prior art, provide a guide method, which can accurately associate the walking device with the tourists, ensure accurate guidance to the tourists, be applicable to various different terrain features, different range size scenic spots, and realize the overall optimization of the tour efficiency of all tourists.

[0007] The technical solution of the present application is as follows:

[0008] A guide method, which identifies the identity of the tourists and obtains the position information of the tourists; an interactive platform obtains the route demand of the tourists through the interaction between the artificial intelligence guide and the tourists; based on the identity recognition result of the tourists, the walking device is matched with the tourists, the walking device obtains the real-time position information and route demand of the matched tourists, and moves to meet the tourists, and based on the real-time position information and route demand of the current tourists and the real-time position information and route demand of other tourists, the optimal navigation route is calculated in real time to guide the current tourists in an interactive manner.

[0009] As a preferred, the position information of the walking device is obtained, and the real-time travel route of the walking device moving to meet the tourists is calculated in real time in combination with the real-time position information of the matched tourists.

[0010] As a preferred, the action data, expression data and expression data of the tourists are obtained, a pre-judgment model pre-trained through machine learning is used to pre-judge the next action of the tourists, including staying, playing, resting, accelerating walking, decelerating walking or running, so as to generate a walking device control instruction to control the walking device to adjust the motion posture, action and speed; if the next action of the tourists is not accurately pre-judged, the control instruction is corrected based on the real-time action data of the tourists, and the surrounding personnel density, environmental information, play content, interactive content, action data, expression data and expression data of the tourists are added to the sample library as training samples for further training and adjustment of the pre-judgment model.

[0011] As preferred, after the walking device meets the visitor, the distance between the walking device and the visitor is determined, and the environmental factors between the walking device and the visitor are combined. If the distance between the walking device and the visitor is too large, or the environmental factors form a visual obstruction or a walking obstacle between the walking device and the visitor, the walking device waits for the visitor to approach, or the visitor approaches to eliminate the visual obstruction; or the walking device stops moving until the visual obstruction or the walking obstacle is eliminated; or before the visual obstruction or the walking obstacle is formed, the walking device waits for the visitor to approach to a distance where the visual obstruction cannot be formed.

[0012] As preferred, the environmental factors include moving objects, fixed objects;

[0013] When the moving object forms a visual obstruction between the walking device and the visitor, the walking device waits for the visitor to approach to eliminate the visual obstruction, or the walking device stops moving until the visual obstruction is eliminated;

[0014] When the moving object forms a walking obstacle between the walking device and the visitor, the walking device stops moving until the visual obstruction is eliminated;

[0015] When the fixed object forms a visual obstruction between the walking device and the visitor, the walking device waits for the visitor to approach to eliminate the visual obstruction;

[0016] Before the moving object forms a visual obstruction or a walking obstacle between the walking device and the visitor, the walking device waits for the visitor to approach to a distance where the visual obstruction cannot be formed;

[0017] Before the fixed object forms a visual obstruction between the walking device and the visitor, the walking device waits for the visitor to approach to a distance where the visual obstruction cannot be formed.

[0018] As preferred, when the moving object moves between the walking device and the visitor, a visual obstruction or a walking obstacle is formed; when the walking device moves and the visitor is located on both sides of the fixed object, and the fixed object forms a visual blind area for the visual area of the visitor towards the walking device, a visual obstruction is formed.

[0019] As preferred, if the visitor changes the route requirement by interacting with the artificial intelligence guide, the walking device obtains the latest route requirement and calculates the optimal navigation route in real time based on the real-time location information of the visitor and the latest route requirement.

[0020] As preferred, alternatively, the interaction platform does not directly obtain the route requirement of the visitor; the interaction platform customizes the route requirement for the current visitor based on the interaction information, including the selection of scenic spots, the tour sequence of scenic spots, and the estimated play time of each scenic spot.

[0021] As preferred, according to the route demand of all tourists, combined with the real-time location information, route demand, and current queue number of all tourists, the queue number of the attractions contained in the current tourist's route demand at the time when the current tourist is expected to arrive is calculated; if the queue number at the time when the current tourist is expected to arrive corresponds to an expected waiting time that is greater than the waiting time threshold, a route demand change suggestion is proposed to the tourist, and the order of visiting the attractions is adjusted to guide the tourist to other attractions with an expected waiting time that is less than the original next attraction.

[0022] As preferred, or if the queue number at the time when the current tourist is expected to arrive corresponds to an expected waiting time that is greater than the waiting time threshold, the interactive platform actively modifies the route demand, the artificial intelligence guide adjusts the guide words, the transition words that connect the original next attraction are adjusted to the transition words that correspond to the next attraction of the latest route demand, and the tourist is guided to the next attraction of the latest route demand.

[0023] As preferred, the guide words and transition words are generated in real time according to the personality characteristics, interest data, knowledge background, and tourism experience of the tourist.

[0024] As preferred, the route demand change suggestion and the modification of the navigation route corresponding to the route demand are based on the combination of multiple tourists, and specifically, according to the authorization of the tourist, part or all of the following information of the tourist is obtained: tourist social platform public information, tourist age, residence region, work type, education background, and tourism history obtained from government platforms and enterprise platforms, combined with the language habits and interaction content of the tourist, a route adjustment model pre-trained through machine learning is used to determine the most likely accepted route adjustment scheme of multiple tourists, and a route demand change suggestion for multiple tourists is generated.

[0025] As preferred, after the first round of interaction between the tourist and the interactive platform through the artificial intelligence guide about the route demand change suggestion, for the tourist who accepts the route demand change suggestion and the tourist who expresses complete rejection of the route demand change suggestion, no route demand change suggestion is actively proposed within the preset time or the best time determined by the route adjustment model based on the aforementioned obtained tourist information.

[0026] In the second round of interaction, the feedback of the tourist who does not accept the route demand change suggestion, as well as the accepted route demand change suggestion and the route demand change suggestion that is expressed to be completely rejected, are used as parameters for navigation route optimization, combined with the queue situation, regional density, road density, and tourist information of each attraction, a batch of tourists who disagree with the previous route demand change suggestion but do not express complete rejection of the route demand change suggestion are selected as the target of this round of navigation route optimization, and specific best persuasion sentences are generated for each tourist to persuade the tourist to accept the route demand change suggestion.

[0027] As preferred, the personality characteristics, interest data, knowledge background, travel experience, or tourist information of the tourist is obtained through interaction with the artificial intelligence guide; or, the tourist is able to browse the publicly available information on the Internet; or, the tourist is authorized to obtain the data through barrier-free data interaction.

[0028] As preferred, the peer-to-peer network is used to identify the non-specific characteristics and location of the target, including the tourist, the walking device or the moving object.

[0029] The peer-to-peer network includes a plurality of node devices, and there is no primary and secondary relationship between all the node devices. The node device is provided with a data collection device and an operation module. The data collection device includes at least one type of sensor for collecting different corresponding types of sensing data. The node device arranged at different collection positions collects at least one point sample of the consumer, and the point sample is the sensing data corresponding to the sensor type.

[0030] For a certain node device, the collected sensing data is processed to obtain result data, and the result data is propagated to other node devices. The other node devices receiving the result data use the result data as one of the collected original data, and the result data of the other node devices is affected by the result data. Based on this, the plurality of node devices in the peer-to-peer network perform collaborative calculation without obtaining the identity information of the target, to determine that each unique target is itself, to realize non-specific characteristic recognition, and to identify the location of the tourist, the walking device or the moving object.

[0031] As preferred, the current node device receives the result data output by the other node devices. For the current node device, the collected sensing data is combined with the result data from the other node devices to calculate the result data of the current node device, and is sent to the other node devices. The node devices in the peer-to-peer network perform collaborative calculation with the collection of sensing data and the calculation of result data.

[0032] As preferred, in the peer-to-peer network, for a certain point sample of a certain target, the result data transmitted from the node device collecting the point sample to other node devices is adjusted by the subsequent node device according to the characteristics of the point sample, or the characteristics of the point sample are reported to the subsequent node device to adjust the sensing attention. If the subsequent other node device does not detect the characteristics of the point sample, but can determine from the characteristics of other point samples that the undetected characteristics of the point sample still belong to the target, the undetected characteristics of the point sample are still expressed in the result data of the current node device and transmitted to other node devices.

[0033] As preferred, the method for reporting the feature of the point sample for the subsequent node device to adjust the perceptual attention is: adjusting the parameters of the data processing model of the subsequent node device based on the result data provided by the previous node device expressing the feature of the point sample, or the feature of the point sample, so that the subsequent node device improves the computing power for identifying the feature of the point sample; or the subsequent node device uses the perceptual attention model to match the received feature of the point sample or the result data expressing the feature of the point sample for computing power adjustment.

[0034] As preferred, when the node device processes the result data output by several previous node devices, based on the data processing model, when the targets described by several previous node devices can be determined as the same target through certain common point sample features, the point sample features and other information described by each node device are combined into the same target.

[0035] As preferred, when the result data received by the node device shows that the mark used by the current node device to identify the target before the current time of receiving the result data is different from the mark used by other node devices to identify the target, and the mark allocated by other node devices for the target is updated, the mark used by the current node device to identify the target before the current time of receiving the result data is converted.

[0036] As preferred, the method for converting the mark used by the current node device to identify the target before the current time of receiving the result data is:

[0037] replacing the mark used by the current node device to identify the target before the current time of receiving the result data with the latest mark allocated by other node devices for the target;

[0038] or, recording the conversion relationship between the mark used by the current node device to identify the target before the current time of receiving the result data and the updated mark allocated by other node devices for the target, and converting when the result data currently received by the current node device is needed;

[0039] or, the node device deploys a conversion model to convert the marks of multiple targets according to the input original data or result data.

[0040] As preferred, for one or more point samples collected by node devices at different collection positions in sequence, if the feature values of a certain point sample or multiple point samples at different collection positions respectively meet the preset proximity condition or are determined by a certain model to have relevance reaching a threshold, and are unique at each collection position, it is determined that the point sample of the certain type at different collection positions has relevance.

[0041] As preferred, the one or more point samples collected by the node devices at different collection positions are associated if the node devices at different collection positions collect the same space field, and there is only one target in the space field, or the point samples collected can correctly point to one of the multiple targets to which they belong.

[0042] As preferred, the data collection device of the node device comprises one or a combination of image acquisition device, electromagnetic induction device, temperature measurement device, vibration frequency sensing device and laser radar. The data collected by the above devices and the three-dimensional point cloud collected by the laser radar or the point cloud generated from the images collected by the multiple image acquisition devices are jointly calculated to obtain three-dimensional points with data. The image color, contour, line, reflectivity, motion trend, electromagnetic characteristics, temperature, temperature change trend, vibration frequency and vibration frequency change trend based on two-dimensional perception are taken as additional attributes of the corresponding three-dimensional points to form a three-dimensional point cloud with attributes. The correspondence between each part or relevant part of the 3D appearance of the consumer and each region of the three-dimensional point cloud with attributes is determined in combination with the change characteristics of electromagnetic induction, temperature, vibration frequency, motion correlation and reflectivity.

[0043] As preferred, when the identity information of the target needs to be obtained, an identity information acquisition command is triggered, the identity information acquisition command is taken as one of the inputs for the calculation of the result data of the node device, and the identity information of the target is obtained by driving the node device connected in the peer-to-peer network to respond to the corresponding result data under the condition that the barrier-free data collection condition for obtaining the identity information of the target is met.

[0044] As preferred, the peer-to-peer network determines the authority of the target by verifying the authenticity of the identity information of the target; wherein the node device capable of obtaining the identity information in the peer-to-peer network does not provide the identity information, but only expresses the verification result of the authenticity of the identity information in the result data of the node device according to the verification requirement of the authenticity of the identity information in the received result data.

[0045] As preferred, the node device capable of obtaining the identity information in the peer-to-peer network does not provide the identity information, and the information source device providing the identity information is driven to establish an encrypted file transmission channel between the input terminal of the node device requiring the identity information and the information source device providing the identity information, or to establish an encrypted information transmission channel in other network communication mode; the identity information is taken as one of the inputs of the node device.

[0046] As preferred, the data collection device comprises one or more of image acquisition device, audio acquisition device, temperature measurement device, vibration frequency sensing device, laser radar, chemical sensor and electromagnetic induction device.

[0047] As preferred, the walking device is provided with at least one execution device; each execution device is connected to one or more node devices in the peer-to-peer network; the interaction platform submits a request command to the peer-to-peer network; if it is determined based on the collaborative calculation that the current node device needs to respond to the request command, the current node device will send an instruction to the execution device connected to the current node device according to the calculated result data, so as to control the execution device to complete the response action.

[0048] As preferred, the result data corresponding to the request command is a result representation; the execution device receives the result data output by the connected node device; if a specific element in the result data indicates that the execution device needs to respond, or the result data is used as one of the inputs of the data processing model of the node device, and it is determined by calculation that the corresponding execution device needs to respond, then the execution device performs the corresponding operation.

[0049] As preferred, when the execution device needs to respond, the execution device combines the received result data output by other node devices to calculate its own result data, and controls the execution components on the execution device to perform the corresponding operation through the obtained result data.

[0050] As preferred, the execution device receives the result data output by other node devices, and the principle is: when the request command needs the corresponding execution device to respond, if the result data obtained by the calculation of one or more node devices can determine the execution device that needs to respond, then the corresponding execution device is added to the node list that delivers the current result data; the one or more node devices directly deliver the result data to the execution device or the node device connected to the execution device; or the execution device receives the result data output by other node devices in a layer-by-layer delivery manner.

[0051] As preferred, the corresponding execution device is added to the node list that delivers the result data according to the preset condition or algorithm output, model output.

[0052] As preferred, the execution device is a node device connected to a specific function execution component, and the execution feedback information of the execution component of the execution device is fed back to the execution device to participate in the calculation of the subsequent result data of the execution device.

[0053] As preferred, the execution device of the walking device includes but is not limited to a movement component, a brake component, a steering component, an energy component, a voice playing device, a projection device, a light device, a video playing device, a wireless communication module, and a special-purpose host; the special-purpose host is used to support the augmented reality glasses and functional props worn by tourists.

[0054] The beneficial effects of the present application are as follows:

[0055] The guide method provided by the application realizes that the walking device provides guide services at any time based on the identity information, position information and route demand of the tourists and in combination with the position information of the associated walking device, and direct and indirect interaction can be carried out between the tourists and the associated walking device, the walking device can ensure effective guidance of the tourists and avoid the situation that the tourists are lost due to the surrounding environment. The application plans all the tourists, optimizes the navigation route of all the tourists in the scenic area, improves the touring experience of the tourists, relieves the congestion degree of the scenic spot and improves the operation efficiency of the scenic area.

[0056] The application optimizes the navigation route based on the combination of multiple tourists, provides the route change suggestion to the tourists in combination with as much detailed tourist information as possible, improves the effect of the navigation route optimization through the continuously iterated tourists, convinces the tourists to accept the route change suggestion through the targeted best persuasion sentences for each tourist, realizes the navigation route optimization, improves the touring experience of the tourists based on the navigation route optimization, relieves the congestion degree of the scenic spot and improves the operation efficiency of the scenic area.

[0057] The application uses the peer-to-peer network for collaborative computing, performs non-specific feature recognition and position recognition on the tourists and the walking device to complete identity recognition and positioning, all node devices in the peer-to-peer network have no primary and secondary relationship, there is no fixed connection path between the node devices, the node devices only receive the calculation results of other node devices, send the calculation results of themselves to the outside, and respond to the discovery of events and / or corresponding execution devices without relying on a single node device for identification and control, but through collaborative computing of multiple node devices in the peer-to-peer network for common confirmation, in the case that specific features are not required and specific identity information is obtained, each unique target can be determined as itself to realize non-specific feature recognition. The application realizes target recognition, identity confirmation or event monitoring through the non-specific feature recognition mode, not only the recognition result is accurate, but also the position recognition is accurate. The application can realize target recognition and identity confirmation without relying on specific features, protect privacy, and at the same time, solve the problems of traffic, education, medical convenience, epidemic prevention, livelihood services, emergency, public security, anti-terrorism, community management and services, market behavior, safety production, civilized behavior and the like.

[0058] The application adopts non-specific feature recognition, on the other hand, can effectively prevent the risk caused by theft or imitation of specific features, greatly improve the security. The application adopts a non-contact passive mode to perform non-inductive recognition on the identity of the target, greatly improve the convenience of execution. The application is based on the peer-to-peer network, and the coverage range is on the order of hundreds of meters or hundreds of kilometers for easy layout, and is suitable for various levels of geographical range.

[0059] In the present application, the execution device responds to the execution based on the calculation result obtained by cooperative calculation, and the response efficiency is high, avoiding false execution or non-execution during execution caused by network attacks and other illegal responses. In order to avoid hijacking, the present application can also use multiple node devices to cooperatively control the execution device, further improving the immunity to hijacking attacks. DETAILED DESCRIPTION

[0060] The present application will be further described in detail below in combination with embodiments.

[0061] The present application provides a guide method to solve the poor guide effect and the inability to optimize the navigation route to improve the tour experience and the overall efficiency of the scenic area in the prior art. The guide method can accurately associate the walking device with the tourists and ensure accurate guidance to the tourists. It is suitable for scenic areas with different terrain characteristics and different range sizes, and the tour efficiency of all tourists is optimized as a whole.

[0062] In the guide method of the present application, the identity of the tourist is identified and the position information of the tourist is obtained during the journey of the tourist to the scenic area or in the scenic area, so as to realize real-time tracking of the tourist. The interactive platform obtains the route requirements of the tourist through the interaction between the artificial intelligence guide and the tourist, including the selection of scenic spots, the tour order of scenic spots, and the play time of each scenic spot. The interactive platform can be implemented as an interactive software platform such as APP, WeChat mini program or public number, or a hardware facility deployed in the scenic area. In the present application, the walking device is matched with the tourist based on the identity recognition result of the tourist, and the walking device is used to lead the tourist, forming an association between the walking device and the tourist in a matching relationship, and providing one-to-one or one-to-many guide services. The walking device obtains the real-time position information and route requirements of the matched tourist, and moves to meet the tourist (based on the real-time position information of the walking device and the tourist, the walking device can catch up with the tourist through the planned path, during which the tourist can continue to walk or wait at any position for the walking device). In the present application, the optimal navigation route is calculated in real time based on the real-time position information and route requirements of the current tourist and the position information and route requirements of other tourists, and the walking device walks based on the navigation route to lead the tourist; and the interactive platform interacts with the current tourist through the artificial intelligence guide.

[0063] Since the tourist may change the route requirement at any time, in order to ensure the tourist to be guided in real time with the optimal navigation route, in the present application, if the tourist changes the route requirement by interacting with the artificial intelligence guide, the walking device acquires the latest route requirement, and calculates the optimal navigation route in real time based on the real-time position information of the tourist and the latest route requirement. Alternatively, the interaction platform does not directly acquire the route requirement of the tourist, but the interaction platform customizes the route requirement for the current tourist based on the interaction information according to the interaction between the artificial intelligence guide and the tourist, including the selection of the scenic spots, the visiting sequence of the scenic spots, and the expected playing time of each scenic spot.

[0064] After the walking device acquires the matched real-time position information of the tourist and the route requirement, further, the position information of the walking device is acquired, the walking route for the walking device to move to meet the tourist is calculated in real time in combination with the matched real-time position information of the tourist, and the walking device moves towards the matched tourist based on the walking route. Since the walking route can be calculated and updated in real time, the walking device can accurately move towards the tourist regardless of whether the tourist is in a moving state or a staying state, until the walking device meets the tourist, thereby avoiding the waste of time caused by the tourist stopping the tour to wait for the walking device, and avoiding the situation that the tourist cannot meet the walking device due to the movement of the tourist.

[0065] After the walking device meets the tourist and leads the tourist, in order to make the walking device consistent with the visiting pace of the tourist as much as possible, in the present application, further, the action data, the manner data, and the expression data of the tourist are acquired, the next action of the tourist is predicted by using the pre-judgment model pre-trained by machine learning, including staying, playing, resting, accelerating walking, decelerating walking, or running, so as to generate the walking device control instruction, control the walking device to adjust the motion posture, the action, and the speed, so as to realize the state adjustment of the walking device following the visiting pace of the tourist, and make the visiting pace of the walking device consistent with the visiting pace of the tourist as much as possible.

[0066] If the prediction of the next action of the tourist is not accurate, the control instruction is corrected based on the real-time action data of the tourist, and the surrounding personnel density, the environmental information, the playing content, the interaction content, the action data, the manner data, and the expression data of the tourist are added to the sample library as training samples, for further training and adjustment of the prediction model, so as to improve the accuracy of the prediction model.

[0067] In order to avoid the situation that the tourist loses the following of the walking device after the walking device meets the tourist and leads the tourist, the distance between the walking device and the tourist is judged in the application, and the environmental factors (including the moving object and the fixed object) between the walking device and the tourist are combined. If the distance between the walking device and the tourist is too large, or the environmental factors form the visual obstruction or the walking obstacle between the walking device and the tourist (i.e. there is an obstacle on the straight line connecting the tourist and the walking device, including the moving object and the fixed object, which causes the tourist to be unable to see the walking device or the tourist to be unable to advance), the walking device waits for the tourist to approach (the tourist is in the advancing state), or the tourist approaches to eliminate the visual obstruction (i.e. the tourist is in the advancing state, and the tourist moves between the obstacle and the walking device, so that the visual obstruction formed by the obstacle is eliminated); or the walking device stops moving until the visual obstruction or the walking obstacle is eliminated (the visual obstruction or the walking obstacle is actively or passively eliminated); or before the visual obstruction or the walking obstacle is formed, the walking device waits for the tourist to approach to the distance at which the visual obstruction cannot be formed (the formation of the visual obstruction or the walking obstacle is predicted, and the distance between the walking device and the tourist is shortened before the visual obstruction or the walking obstacle is formed, so that the formation of the visual obstruction or the walking obstacle is actively avoided).

[0068] In the embodiment, the environmental factors include the moving object (including people or objects) and the fixed object (usually an object, such as the structure of a building itself or a fixedly arranged article). When the moving object moves between the walking device and the tourist, the visual obstruction or the walking obstacle is formed. When the walking device moves and the tourist is located on the two sides of the fixed object respectively, and the fixed object forms the visual blind area for the visual area of the tourist towards the walking device, i.e. the fixed object is blocked on the straight line connecting the tourist and the walking device, the visual obstruction is formed.

[0069] Specifically, when the moving object forms the visual obstruction between the walking device and the tourist, if the tourist can still advance, the walking device (decreases the speed or stops moving) waits for the tourist to approach to eliminate the visual obstruction (i.e. the tourist is in the advancing state, and the tourist moves between the moving object and the walking device, so that the visual obstruction formed by the moving object is eliminated), or the walking device stops moving until the visual obstruction is eliminated (the moving object moves by itself until the visual obstruction is eliminated).

[0070] When the moving object forms the walking obstacle between the walking device and the tourist, the walking device stops moving until the visual obstruction is eliminated (the moving object moves by itself until the walking obstacle is eliminated).

[0071] When a fixed object forms a visual occlusion between the walking device and the visitor (for example, a corner of a building, and the walking device and the visitor are located on the two sides of the corner, respectively, and thus a visual occlusion is formed, that is, the corner of the building blocks the line of sight of the visitor towards the walking device), the walking device waits for the visitor to approach until the visual occlusion is eliminated (the walking device slows down or stops moving, and the visitor continues to move forward until the visitor can see the walking device).

[0072] Before a moving object forms a visual occlusion between the walking device and the visitor or a walking obstacle is formed, the walking device waits for the visitor to approach until the distance between the walking device and the visitor is shortened to a distance at which the moving object cannot enter, and thus a visual occlusion or a walking obstacle cannot be formed; that is, based on the real-time position information of the moving object, the motion trajectory of the moving object is predicted, and the motion state of the walking device is actively adjusted, so that the distance between the walking device and the visitor is shortened.

[0073] Before a fixed object forms a visual occlusion between the walking device and the visitor, the walking device waits for the visitor to approach until the distance between the walking device and the visitor is shortened to a distance at which a visual occlusion cannot be formed; that is, based on the position information of the identified or known fixed object, and based on the dynamic state of the walking device and the visitor, including the speed, direction, etc., it is determined whether the distance between the walking device and the visitor will be such that the walking device and the visitor are located on the two sides of the fixed object, and thus a visual occlusion is formed; if so, the motion state of the walking device is actively adjusted, so that the distance between the walking device and the visitor is shortened to a distance at which a visual occlusion cannot be formed.

[0074] Based on this, the effective following between the walking device and the visitor can be further ensured, and the visitor can be prevented from losing the following of the walking device.

[0075] In order to avoid the visitor waiting for a long time at a popular scenic spot and effectively improve the overall operation efficiency of the scenic area, in the present application, according to the route requirements of all visitors, in combination with the real-time position information of all visitors, the route requirements, and the current queue number of each scenic spot, the queue number of the scenic spot included in the route requirement of the current visitor at the expected arrival time of the current visitor is calculated; if the expected waiting time corresponding to the queue number at the expected arrival time of the current visitor is greater than the waiting time threshold, a route requirement change suggestion is made to the visitor, and the order of visiting the scenic spots is adjusted to guide the visitor to other scenic spots with an expected waiting time less than the original next scenic spot.

[0076] The route demand change suggestion is based on real-time route demand change suggestions of all tourists, and whether the tourists accept the route demand change suggestion, that is, the latest route demand and navigation route corresponding to all tourists, and the queuing number of each scenic spot at the current tourist expected arrival time is updated in real time; and within a certain time, the queuing number of each scenic spot will reach a dynamic balance, and the route demand change suggestion is optimal within a certain time, that is, there will be no situation that the route demand change is continuously carried out and the scenic spot cannot be reached.

[0077] Or, if the queuing number at the current tourist expected arrival time corresponds to an expected waiting time greater than the waiting time threshold, the interactive platform actively modifies the route demand, the artificial intelligence guide adjusts the guide words, the connecting words connecting the original next scenic spot are adjusted to the connecting words corresponding to the next scenic spot of the latest route demand, and the tourists are guided to the next scenic spot of the latest route demand. Since the interactive platform actively modifies the route demand, that is, it does not need to be agreed by the tourists, but in order to avoid the modification of the route demand causing the adjustment of the navigation route, causing the guide words not to correspond, or being inconsistent with the expected situation of the tourists, the guide words need to be actively adjusted according to the modification of the route demand, that is, the connecting words connecting the original next scenic spot are adjusted to the connecting words corresponding to the next scenic spot of the latest route demand, and the tourists are guided to the next scenic spot of the latest route demand, so as to avoid causing the tourists to feel abrupt. Specifically, the guide words and the connecting words are generated in real time according to the personality characteristics, interest data, knowledge background and tourism experience of the tourists, so as to maximize the matching of the preferences and acceptance degree of the tourists.

[0078] In the present application, the data of multiple tourists is used for navigation route optimization, and in specific implementation, the route demand change suggestion and the modification of the navigation route corresponding to the route demand are based on the combination of multiple tourists, and specifically, according to the authorization of the tourists, part or all of the following information of the tourists is obtained: tourist social platform public information, tourist age, residence region, work type, education background and tourism history obtained from government platforms and enterprise platforms, language habits and interactive content of the tourists, and a route adjustment model pre-trained through machine learning is used to determine the most likely accepted route adjustment scheme of multiple tourists, and a route demand change suggestion for multiple tourists is generated.

[0079] After generating the route demand change suggestion for multiple tourists, the first round of interaction is carried out between the tourists and the interactive platform through the artificial intelligence guide. After the first round of interaction between the tourists and the interactive platform through the artificial intelligence guide about the route demand change suggestion, there can be complete acceptance, partial acceptance, complete non-acceptance, etc. Among them, for the tourists who accept the route demand change suggestion and the tourists who express complete non-acceptance of the route demand change suggestion, no longer actively propose the route demand change suggestion within the preset time or the best time judged by the route adjustment model according to the tourist information obtained in the foregoing.

[0080] In the second round of interaction, the feedback of the tourists who do not accept the route demand change suggestion, the accepted route demand change suggestion, and the route demand change suggestion expressed as complete non-acceptance are taken as parameters for navigation route optimization, combined with the queuing situation of each scenic spot, regional density, road density, and tourist information, a batch of tourists who disagree with the previous route demand change suggestion but do not express complete non-acceptance of the route demand change suggestion are selected as the target of this round of navigation route optimization, and the best persuasion statement is generated for each tourist to persuade the tourist to accept the route demand change suggestion, so as to avoid excessive interaction of the route demand change suggestion, which not only wastes computing power but also affects the tourist's touring experience.

[0081] In order to follow the principle of protecting the privacy of tourists, in the present application, the personality characteristics, interest data, knowledge background, and tourism experience of tourists, or tourist information, are obtained through interaction with the artificial intelligence guide (i.e. voluntarily provided), or obtained by browsing online public information that can be browsed by tourists, or obtained through authorization of the tourists in an accessible data interaction manner.

[0082] In specific implementation, a traditional single-point recognition method can be used to identify the identity of the tourists at a set position to achieve the purpose of associating the position information with the identity confirmation, or a non-specific feature identity recognition based on peer-to-peer network collaborative computing provided by the present application can be used. The peer-to-peer network of the present application is based on collaborative computing and does not rely on single-point recognition, distributing the computing function throughout the network, reducing the software and hardware requirements of single-point computing, improving the execution efficiency, and greatly improving the attack resistance. The information between the node devices is in a relatively symmetrical state, which can prevent illegal tampering of data. Even if a single node device is physically cracked and its data is tampered with, because the network-wide operation is a super-high redundancy complex calculation and super-multiple dimension verification, the tampering of the data of a single node device does not affect the network-wide calculation result, and the faulty and tampered node device can be quickly located, ensuring the credibility of the network-wide calculation result, and further, the contradiction between data sharing and information security between departments can be solved.

[0083] The result data transmitted between the node devices can be the processing result of information, rather than the information itself, and thus the original data (i.e. the sensing data) collected can not be stored, the node devices only receive the calculation results output by other node devices and send their own calculation results, the information contained in a single calculation result is insufficient to restore any event and target information, and a determined result can be obtained only by joint cooperative calculation of the calculation results, multi-dimensional data matrix elements and the corresponding relationship between physical space and facilities on the entire peer-to-peer network, the cooperative calculation has less dependence on the information transmitted by a small number of node devices, and thus can fundamentally change the essence of traditional informationization single-point security sensitivity.

[0084] In the present application, the acquisition of the identity information of the tourists and the acquisition of the position information of the walking devices and the moving objects can be obtained by the cooperative calculation of the peer-to-peer network provided by the present application. Specifically, the present application uses the peer-to-peer network to perform non-specific feature recognition and position recognition on a target, and the target includes a tourist, a walking device or a moving object. The "non-specific feature recognition" is different from the general "recognition" in strict conceptual definition. The general "recognition" means to determine the image or the specific identity information of the target, such as who (including the name and specific information indicating the identity of the target) and what (such as a car, a person, etc.). The "recognition" in the "non-specific feature recognition" of the present application means to determine each unique target (i.e. a tourist, a walking device or a moving object) as itself; that is, for a certain to-be-recognized object, its existence is unique, and after the "non-specific feature recognition" is implemented by the present application, it is determined that the to-be-recognized object (i.e. the target that has not been recognized or identified) is itself, rather than other to-be-recognized objects, and the result of the "non-specific feature recognition" does not need to determine the specific features of the to-be-recognized object or the identity information or image of the to-be-recognized object. For example, for a person, it is regarded as to-be-identified object A, and for an object, it is regarded as to-be-identified object B, and after the "non-specific feature recognition" is implemented, it is not necessary to identify that to-be-identified object A is a person and who the specific identity is, and it is not necessary to identify that to-be-identified object B is an object and what the specific object is; instead, it is necessary to determine that to-be-identified object A is to-be-identified object A itself and to-be-identified object B is to-be-identified object B itself. Then, the corresponding service or control can be performed on to-be-identified object A or to-be-identified object B.

[0085] The peer-to-peer network comprises a plurality of node devices, and there is no primary and secondary relationship among all the node devices, forming a decentralized network and computing framework. Unlike the traditional information-based single-point convergence computing mode, there is no fixed and preset path relationship in the direction of data transmission between the node devices of the application. In the peer-to-peer network described in the application, for a certain node device, the collected raw data is processed to obtain result data, and the result data is propagated to other node devices; the other node devices that receive the result data take the result data as one of the collected raw data, and the result data influences the result data of other node devices. For convenience of description, the aforementioned "a certain node device" is referred to as "current node device", and "other node devices" are referred to as "subsequent node devices". One of the influences is that the result data obtained by the subsequent node device is not completely determined by the raw data collected by itself, but is jointly determined with the result data output by the current node device; wherein the result data output by the current node device may change the data processing model and parameters used by the subsequent node device to calculate the result data, thereby affecting the result data of the subsequent node device. For example, if the result data output by the current node device is associated with the raw data collected by the subsequent node device, the influence of the result data output by the current node device on the accuracy of the result data of the subsequent node device must be considered; specifically, for the perception of a certain target, if the result data obtained by calculating only based on the raw data collected by the subsequent node device can only reflect the real-time (including real-time location and time) single-point result judgment of the target within the perception range of the subsequent node device; and the result data output by the current node device reflects the direct perception data and result judgment of the target at other locations and other times, or indirectly related other perception data and result judgment, which helps to improve the accuracy and comprehensiveness of the result data of the subsequent node device, including superimposed calculation of the same dimension, correlation reference of different dimensions.

[0086] Due to the absence of master-slave relationship between the node devices in the peer-to-peer network, point-to-point transmission can be performed between the node devices, and thus, for a certain perception data of a certain target embodied in the result data output by a certain node device, the information is relatively symmetrical in other node devices receiving the result data; other node devices take the received result data as input, combine the perception data of their own sensors, calculate their own result data, and naturally cover the received result data and the information embodied by their own sensors, and transmit to other node devices of the next layer, and thus, for a certain perception data of a certain target, the information is in a relatively symmetrical state in all node devices, which can prevent the influence of the calculation process and the calculation result of a single node device from being tampered and counterfeited on the result data, and also becomes a means to find out the fault or tampered node device and the node device with non-compliant performance, fundamentally solving the fundamental hidden danger of traditional information technology, i.e. false information, counterfeit information, and error information caused by information asymmetry, which in turn becomes a starting point for fraud and network attacks, and the problems of poor accuracy, long time consumption, poor credibility, and poor strain capacity in complex and comprehensive applications, and thus can truly become an information infrastructure for large-area comprehensive management and a basic infrastructure for digital economy. The present application is different from the technical scheme of the blockchain technology which still adopts independent calculation of each node, determines the result, and focuses on the storage of original data, and the present application focuses on the peer-to-peer collaborative calculation between node devices, so that each node device can adjust its own data processing model (i.e. the algorithm of the calculation result data) and parameters when processing data, and the adjustment is the feedback of all node devices to their own adjustment, so that the calculation of all node devices becomes a whole, and each node device no longer completes the calculation independently, but all node devices complete the calculation together. After the data processing model of the node device is adjusted, the adjustment will have an impact on the next data processing.

[0087] The node device is provided with a data acquisition device (in specific implementation, one or more of an image acquisition device, an audio acquisition device, a temperature measurement device, a vibration frequency perception device, a laser radar, a chemical sensor, and an electromagnetic induction device), and an operation module. The data acquisition device includes at least one type of sensor for acquiring different corresponding types of perception data. The operation module calculates the result data based on a data processing model. The node devices arranged at different collection positions (i.e. located at different physical installation positions) collect at least one point sample of the target, and the point sample is the perception data of the corresponding sensor type. Based on this, the multiple node devices in the peer-to-peer network perform collaborative calculation without obtaining the identity information of the target, determine that each unique target is itself, realize non-specific feature recognition, and realize position recognition of tourists, walking devices, or moving objects.

[0088] Specifically, taking a certain node device as a current node device, combining the data transmission of its previous node device and subsequent node device (in the present application, the previous node device and the subsequent node device are only used to describe the previous and subsequent relationship with the current node device in the current calculation and data transmission process, and do not mean that they have a certain previous and subsequent relationship and priority relationship), the corresponding current node device receives the result data output by other node devices (including the previous node device), and the subsequent node device receives the result data output by other node devices (including the current node device). For the current node device, the collected sensing data is combined with the result data from other node devices (including the previous node device) to calculate the result data of the current node device and send it to other node devices (including the subsequent node device). Similarly, the working process of the subsequent node device is the same as that of the current node device, and the previous node device also receives the result data of the previous node device of the previous node device and performs the same working process as the current node device; that is, the node devices in the peer-to-peer network perform the same working process. Further, the node devices in the peer-to-peer network perform collaborative calculation with the collection of sensing data and the calculation of result data. Among them, the result data output by a certain node device is only received by the subsequent node device and used as input, and the result data of the subsequent node device will cover the result data of the previous node device (including the aforementioned certain node device).

[0089] In the peer-to-peer network, all events are processed synchronously, and it is not necessary to explicitly generate a stage result output such as what event is found and what the specific content of the event is; in the peer-to-peer network, only the sensing of the sensor and the response of the corresponding execution device are explicit, and other intermediate processes are processed simultaneously by collaborative calculation, that is, in the running process of the present application, the intermediate process of the discovery of the event is not sensitive, and the corresponding execution device automatically responds to the execution with the collaborative calculation and the acquisition of the result data of the node device.

[0090] In order to ensure that the data source and the calculation process are further credible, in the present application, all node devices encrypt the result data calculated by them based on an encryption consensus mechanism to obtain encrypted results, and then send the encrypted results to other node devices. The encryption consensus mechanism includes one or more consensus mechanisms, and different consensus mechanisms correspond to changes in the encryption algorithm structure and parameters of the node device.

[0091] The node devices communicate with each other in standard-sized data packets, i.e., result data or calculation results. In the present application, the node devices of the peer-to-peer network are similar to human neurons, and each neuron does not transmit specific data directly describing external events. Similarly, the node devices do not output raw data, but process the raw data obtained by the connected sensors and data acquisition devices into standard-sized data packets (i.e., result data or calculation results, similar to neural impulses of neurons) according to the data processing model (similar to the biological characteristics of nerve cells) of the node device. The information contained in a single data packet is insufficient to restore any event and target information, and a determined result can only be obtained by joint cooperative calculation of the calculation results, multi-dimensional data matrix elements and physical space and facility correspondence on the entire peer-to-peer network. The cooperative calculation has little dependence on the data output by a small number of node devices, and the cooperative calculation simultaneously processes all the demands received or initiated by the node devices. It is a super multi-dimensional related information cooperative verification calculation, which can fundamentally change the nature of traditional information security sensitive to single point.

[0092] In order to ensure the integrity of the data and the effective execution of the cooperative calculation, in the present application, the QoS mechanism is deployed in the peer-to-peer network, and the QoS mechanism is used to ensure the transmission quality of the result data between the node devices.

[0093] In specific implementation, the networking mode of the peer-to-peer network is one or a combination of 4G mode, 5G mode or MESH mode, so as to adapt to different application scenarios, comprehensively consider the implementation feasibility and cost, and realize the optimal solution. Among them, the MESH mode is based on LTE standard and communicates in the physical layer of LTE standard; data is carried by customized frame structure and interaction is carried out by using special wireless communication protocol. The frame structure is customized for peer-to-peer network calculation, and the special wireless communication protocol developed for urban group peer-to-peer network calculation is used, which can further improve the security and reliability. Moreover, the wireless algorithm fully adapts to the multi-path channel environment controlled based on the consensus mechanism required by the peer-to-peer network calculation. The communication distance in the city is 100 meters to 10 kilometers, and in the case of using an omnidirectional antenna in the wild, the high-efficiency transmission of 120 kilometers can also be realized. In the present embodiment, the communication distance of the Mesh network is: the distance between the indoor node devices is 50-150 meters, the distance between the outdoor node devices is 50 meters-120 kilometers, and the number of node devices that can be accessed by each node is 65535. In addition, when networking in 4G mode or 5G mode, the communication distance is unlimited, and the number of node devices that can be accessed depends on the computing chip computing power and the communication delay.

[0094] In the peer-to-peer network, for a certain point sample of a to-be-identified object, the result data transmitted from the node device collecting the point sample to other node devices can enable subsequent node devices to adjust the perceptual attention according to the characteristics of the point sample (it is not necessary to include the characteristics of the point sample in the result data, but the characteristics of the point sample participate in the calculation of the previous node device, so that the result data of the previous node device can enable the subsequent node device to achieve the effect of adjusting the perceptual attention in the calculation when the data processing model of the subsequent node device is input); or, the characteristics of the point sample are reported for the subsequent node device to adjust the perceptual attention (the characteristics of the point sample are directly expressed in the result data). If the subsequent other node devices do not detect the characteristics of the point sample, but can determine from the characteristics of other point samples that the undetected characteristics of the point sample still belong to the to-be-identified object, the undetected characteristics of the point sample are still expressed in the result data of the current node device and transmitted to other node devices. For example, the previous node device perceives the color on the to-be-identified object A, and the current node device does not perceive the color on the to-be-identified object A, but from the perception data of other node devices, it can be determined that there is the to-be-identified object A in addition to other to-be-identified objects, so the undetected color on the to-be-identified object A is still expressed in the result data of the current node device.

[0095] In the embodiment, the method for reporting the characteristics of the point sample for the subsequent node device to adjust the perceptual attention is: adjusting the parameters of the data processing model of the subsequent node device according to the characteristics of the point sample provided by the previous node device, so that the subsequent node device improves the computing power for identifying the characteristics of the point sample; or, the subsequent node device uses a perceptual attention model to match the received characteristics of the point sample to adjust the computing power.

[0096] Wherein, the "characteristics" above are different from the "feature recognition" in the prior art, and the "feature recognition" in the prior art generally refers to information capable of determining the identity of the target, while the "characteristics" of the present application represent a kind of perceived perception data belonging to the to-be-identified object, such as coordinates, colors on the to-be-identified object, etc., and the "non-specific feature recognition" of the to-be-identified object cannot be directly completed by single-point perception "characteristics".

[0097] In this embodiment, the method for reporting the characteristics of the point sample for subsequent node device to adjust the perceptual attention is: for the result data provided by the previous node device to express the characteristics of the point sample (in this application, usually not the characteristics of the point sample itself, but the characteristics of the point sample are expressed in the result data), or the characteristics of the point sample (i.e. the characteristics of the point sample itself) adjust the parameters of the data processing model of the subsequent node device, so that the subsequent node device improves the computing power for identifying the characteristics of the point sample; or the subsequent node device uses the perceptual attention model to match the received characteristics of the point sample or the result data expressing the characteristics of the point sample for computing power adjustment.

[0098] When the node device processes the result data output by several previous node devices, based on the data processing model, when the to-be-identified objects described by the several previous node devices can be determined as the same target through certain common point sample characteristics, the point sample characteristics and other information described by each node device are combined into the same target. For example, the point sample characteristics in the physical space that almost completely overlap at the same time can be determined as the same target.

[0099] When the node device receives the result data, it indicates that the mark used by the current node device to identify the to-be-identified object before the current time receiving the result data is different from the mark used by other node devices to identify the to-be-identified object, and the mark allocated by other node devices for the to-be-identified object is updated, the mark used by the current node device to identify the to-be-identified object before the current time receiving the result data is converted. Specifically, the method for converting the mark used by the current node device to identify the to-be-identified object before the current time receiving the result data is:

[0100] Replace the mark used by the current node device to identify the to-be-identified object before the current time receiving the result data with the latest mark allocated by other node devices for the to-be-identified object; this is a relatively simple implementation provided by the present application.

[0101] Or, record the conversion relationship between the mark used by the current node device to identify the to-be-identified object before the current time receiving the result data and the updated mark allocated by other node devices for the to-be-identified object, and convert when the current time receiving the result data of the current node device is needed; this is a relatively complex implementation provided by the present application.

[0102] Or, the node device deploys a conversion model to convert the marks of multiple to-be-identified objects according to the input original data or result data; this is a more complex implementation provided by the present application.

[0103] In the present application, in order to improve the effectiveness of "non-specific feature recognition", if the feature values of one or more point samples collected by the node device at different collection positions meet the preset similar condition or are determined by a specific model to have relevance reaching a threshold, and are unique at each collection position, it is determined that the point samples of the same kind at different collection positions have relevance.

[0104] On the other hand, for one or more point samples collected by the node device at different collection positions, if the node device at different collection positions collects the same space field, when there is only a unique object to be identified in the space field, or the collected point samples can correctly point to one of the multiple objects to be identified, for a certain object to be identified, the one or more point samples collected by the node device at different collection positions have relevance.

[0105] In the present application, the data collection device of the node device includes one or a combination of image acquisition device, electromagnetic induction device, temperature measurement device, vibration frequency sensing device and laser radar. The data collected by the above devices (i.e. one or a combination of image acquisition device, electromagnetic induction device, temperature measurement device, vibration frequency sensing device) and the three-dimensional point cloud collected by the laser radar or the point cloud generated according to the images collected by multiple image acquisition devices are jointly calculated to obtain three-dimensional points with data. The image color, contour, line, reflectivity, motion trend, electromagnetic feature, temperature, temperature change trend, vibration frequency, vibration frequency change trend based on two-dimensional perception are taken as additional attributes of the corresponding three-dimensional points to form three-dimensional point cloud with attributes. The electromagnetic induction, temperature law, vibration frequency change characteristics, motion correlation (different motion correlations of different materials such as ropes and cloth), and reflectivity are combined to determine the correspondence between each region of the three-dimensional point cloud with attributes and each part or associated part of the 3D appearance of the object to be identified. The present embodiment utilizes the attributes of the three-dimensional point cloud with attributes and their correlation to determine the relationship between each point, the correspondence between each region to which each relevant point belongs and each part or associated part of the 3D appearance of the object to be identified, which can more accurately determine the point sample features belonging to the object to be identified and improve the efficiency and accuracy of "non-specific feature recognition".

[0106] In the process of "non-specific feature recognition", the identity information of the object to be recognized can also be obtained as necessary. Specifically, when it is determined that the identity information of the object to be recognized needs to be obtained, an identity information obtaining command is triggered, the identity information obtaining command is used as one of the inputs for the calculation of the result data of the node device, and the identity information of the object to be recognized is obtained by driving the node device connected to the barrier-free data collection condition in the peer-to-peer network that can obtain the identity information of the object to be recognized to respond to the corresponding result data. The acquisition of the identity information is also the result of collaborative calculation, that is, the acquisition of the identity information is triggered by the determination of the need to obtain the identity information, rather than by an additional trigger through a specific request command. Based on the present application, if the permission calculation is triggered by a request command, in most cases, the identity information does not need to be obtained to complete the calculation, and only when it is found that the identity information cannot be obtained to complete the permission calculation, the determination of the need to obtain the identity information is generated according to the implementation requirements. For example, through collaborative calculation, it is known that the identity information of a certain person exists in a code scanning registration system in several places, a express pick-up registration system in several places, or a consumption registration system in several places, and the person has authorization or query permission according to law, then the peer-to-peer network can drive the node device connected to these systems in a barrier-free data collection manner, and the relevant information obtained by the node device is sent to the peer-to-peer network to obtain the information comparison and provide accurate identity information. Based on this, the present application can also minimize the possibility of tampering with the identity of a certain system.

[0107] Specifically, the peer-to-peer network verifies the identity information of the object to be recognized, and then determines the permission thereof; wherein the node device capable of obtaining the identity information in the peer-to-peer network can not provide the identity information (according to the implementation requirements, the identity information can also be provided), and only according to the verification requirement of the identity information in the received result data, the verification result is expressed in the result data of the node device. That is, in the present application, in the case that the node device capable of obtaining the identity information does not provide the identity information, only according to the verification requirement of the identity information in the received result data, the verification result is expressed in the result data of the node device.

[0108] When the node device capable of obtaining the identity information in the peer-to-peer network does not provide the identity information, the information source device providing the identity information is driven to establish an encrypted file transmission channel between the input terminal of the node device requiring the identity information or to establish an encrypted information transmission channel in other network communication mode; the identity information is used as one of the inputs of the node device.

[0109] When necessary, in order to meet the needs of other traditional calculation modes for raw data, such as the storage of evidence required by traditional evidence, in the present embodiment, the node setting can increase the setting of data storage device for the storage of raw data sensed by the sensor.

[0110] In implementation, the node device can further have a power supply device with functions such as leakage protection. The node device can further have various communication interfaces, including fiber optic interfaces, wireless communication interfaces, etc. The node device can also have a data interface for external storage devices. The node device can be powered by solar energy or mains electricity. The node device can be installed on a pole such as a street lamp (without a cross arm, wrapped around the main pole, or integrated into the lampshade) when implemented outdoors. In areas without poles, the node device can be wall-mounted or integrated into a ceiling when implemented indoors.

[0111] When the present application is implemented indoors and outdoors, the node device as an artificial intelligence facility installed in public space can serve as urban digital economic infrastructure for 24-hour uninterrupted seamless coverage. Through collaborative computing across node devices, vehicle identity recognition accuracy can reach nearly 100% at any location within the coverage area, with location recognition accuracy related to sensor accuracy.

[0112] In the architecture of the peer-to-peer computing network, all node devices are of the same type and function, and each node device adjusts its data processing model in real time and dynamically according to the consensus mechanism of the entire network. The raw data collected by the data collection devices (including sensors, cameras, etc.) connected to each node device are processed and encrypted by the node device according to its data processing model to generate byte-level processing and encryption results (i.e., result data), which will be sent to other node devices (the computing and encryption results output by other node devices received by the current node device also belong to one of the raw data collected by the current node device). Therefore, the effect of the raw data sensed by each sensor will be propagated among a massive number of peer node devices in a power-of-order level. If each node device sends its result data to 100 surrounding node devices, after four units of time, there will be hundreds of millions of node devices affected by the event sensed by the sensor. In this computing mode, information is relatively symmetric, immune to tampering and forgery, fundamentally solving the fundamental problems of traditional information technology, i.e., false information, forged information, and incorrect information caused by information asymmetry, which then becomes a starting point for fraud and cyber attacks, and complex and comprehensive applications with long cycles, poor precision, and poor adaptability, etc. The present application thus truly becomes an information infrastructure for large-area comprehensive management and a digital economic infrastructure.

[0113] In the present application, the walking device is provided with at least one execution device for executing corresponding functions. In the present embodiment, the execution device of the walking device includes but is not limited to a moving component, a braking component, a steering component, an energy component, a voice playing device, a projection device, a light device, a video playing device, a wireless communication module, and a special host; wherein the special host is used to support the augmented reality glasses and functional props worn by tourists. Each execution device is connected to one or more node devices to join a peer-to-peer network, and the multiple node devices in the peer-to-peer network perform collaborative computing to realize the response to the corresponding execution device. In order to avoid hijacking, the present application can also use multiple node devices to cooperatively control the execution device, and further improve the immunity to hijacking attacks.

[0114] The interaction platform submits a request command to the peer-to-peer network. In the present application, the response to the request command includes "demand - execution", "request - answer" or other different cases. When the result data calculated by one or more node devices in the peer-to-peer network matches the request command, the result corresponding to the request command is represented in the result data output by the one or more node devices, according to the output of the preset condition or the pre-deployed program or the data processing model deployed on the node device. If it is determined based on the collaborative computing that the current node device needs to respond to the request command, the current node device will send an instruction to the execution device connected to the current node device according to the calculated result data, to control the execution device to complete the response action; that is, the case of "demand - execution".

[0115] Based on the collaborative computing of the peer-to-peer network, the execution device can be one of the node devices. When the result data calculated by the execution device can perform related operations on the request command during the collaborative computing, the execution device completes the response to the request command. In the present application, the result corresponding to the request command is represented in the result data, and the result corresponding to the request command includes the response demand of the corresponding execution device. The execution device receives the result data output by the connected node device, and if a specific element in the result data indicates that the execution device needs to respond. Among them, responding to the request command is often completed by multiple execution devices on multiple node devices, such as a node connected display playing images, b node connected loudspeaker playing sound, etc. Or, the result data is one of the inputs of the data processing model of the node device, and the calculation determines that the corresponding execution device needs to respond, and then the execution device performs the corresponding operation.

[0116] When the execution device needs to respond, the execution device combines the received result data output by other node devices to calculate its own result data, and controls the execution components on the execution device to perform corresponding operations through the obtained result data. In the present application, the execution device does not need to determine whether it needs to respond first, but combines the received result data output by other node devices with the perception data collected by its own sensors, inputs the data processing model of the execution device, and outputs the result data to determine whether the execution components of the execution device act and what action they take.

[0117] In the present embodiment, the execution device is a node device connected to a specific function execution component, and the execution feedback information of the execution component of the execution device is fed back to the execution device to participate in the calculation of the subsequent result data of the execution device.

[0118] In the present application, since the execution device can be one of the node devices, the execution device responds to the execution based on the calculation results obtained through collaborative calculation, which is highly efficient and avoids illegal responses such as false execution or non-execution during execution caused by network attacks. To avoid hijacking, the present application can also use multiple node devices to cooperatively control the execution device, further improving the immunity to hijacking attacks.

[0119] In the peer-to-peer network, the result data calculated and output by the node device can be implemented as a state represented by the perception data (i.e. original data), which can be represented by a state value, and then the node device does not need to store and transmit the original data. In the present embodiment, the data or elements in the multi-dimensional matrix are related to the installation position, attributes, etc. of each node device, and then when transmitting the result data, the actual transmission is the transcoded result of the multi-group parameters. The so-called multi-dimensional matrix is actually a combination of multi-group parameters, for example, the path of a certain target is a-b-c-d, and the physical position of the abcd node device is fixed, so the sequence of abcd can be represented by a character or similar concept when transmitting the multi-group parameters.

[0120] Based on the technical characteristics of the peer-to-peer network, it can be applied to various use scenarios that provide targeted services or control for a certain target or event. Based on this, since the result data transmitted between node devices is the processing result of information, not the information itself, the collected original data (i.e. perception data) can not be stored, the node device only receives the calculation results output by other node devices, and sends its own calculation results, and the information contained in a single calculation result is insufficient to restore any event and target information, which must be jointly calculated through the calculation results on the entire peer-to-peer network, the multi-dimensional data matrix elements, and the corresponding relationship between the physical space and facilities to obtain a definite result. Collaborative calculation has less dependence on the information transmitted by a small number of node devices, and thus can fundamentally change the nature of traditional information security sensitivity.

[0121] In the present application, the state evolution of the previous node device output result data is embodied in the result data output by each node device, and then, based on the result data received by the current node device, the behavior, attribute, state or event of the target when perceived by the previous node device can be obtained by reverse deduction. For example, when the position of the target a at 15 minutes ago needs to be found, the position corresponding to the node device perceiving the target a at the current time can be obtained, and the position of the target a can be inferred; then, according to the transmission path of the result data, the position of the target a at 15 minutes ago (determined by the node device perceiving the target a) can be deduced, and the node device does not need to store the original data of the target a, that is, based on the present application, the target a can be found without identifying the original data, and the original data of the target a at the time to be found can be obtained from the storage device connected to the node device when necessary.

[0122] In the present application, the execution device receives the result data output by other node devices, and the principle is that when the request command needs to be responded by the corresponding execution device, if the result data calculated by one or more node devices can determine the execution device that needs to be responded, the corresponding execution device is added to the node list for transmitting the current result data, and the result data is transmitted to the execution device or the node device connected to the execution device, wherein the corresponding execution device is added to the node list for transmitting the result data according to the preset condition or algorithm output, model output; or the execution device receives the result data output by other node devices in a layer-by-layer transmission manner. In the process of cooperative calculation of the peer-to-peer network, the node device also calculates the node list that needs to receive the result data when calculating the result data each time, and according to the current result data, one or more execution devices that need to be added are determined and added to the node list, and the execution device or the node device connected to the execution device is directly used as the subsequent node device of the next layer for directly receiving the current result data, so as to cross the normal layer-by-layer transmission and change the peer-to-peer network into a three-dimensional architecture. For example, the result data of the current node device can explicitly know that the evidence needs to be provided to the public security, and if the normal layer-by-layer transmission method is used, the result data of the current node device needs to be transmitted to the node device corresponding to the public security for at least one or more layers; if the node device corresponding to the public security is added to the node list, the node device corresponding to the public security can directly receive the result data of the current node device in the next layer transmission, thereby greatly shortening the disposal time and improving the response ability. The present application adopts the peer-to-peer network, and therefore this temporary construction is exactly the advantage of the present application, and the traditional informationization layer-by-layer aggregation architecture cannot bear the complex calculation demand brought by this temporary construction network.

[0123] The above-described embodiments are merely illustrative of the present application and are not to be used as limitations. Any changes, modifications, and the like of the above-described embodiments made on the basis of the technical idea of the present application will fall within the scope of the present application.

Claims

1. A method of guiding, characterized by, Identity recognition is performed on the tourists, and position information of the tourists is obtained; the interactive platform obtains route requirements of the tourists through interaction between the artificial intelligence guide and the tourists; based on the identity recognition result of the tourists, the walking device is matched with the tourists, the walking device obtains real-time position information and route requirements of the matched tourists, and moves to meet the tourists, real-time optimal navigation routes are calculated based on the real-time position information and route requirements of the current tourists and the position information and route requirements of other tourists, and the current tourists are interactively guided; Non-specific feature recognition and position recognition are performed on the tourists by using a peer-to-peer network; The peer-to-peer network includes a plurality of node devices, and there is no primary and secondary relationship between all the node devices; the node devices are provided with data acquisition devices and operation modules, the data acquisition devices include at least one type of sensor for collecting different corresponding types of sensing data; the node devices arranged at different collection positions collect at least one point sample of the tourists, and the point sample is sensing data corresponding to the type of sensor; For a certain node device, the collected sensing data is processed to obtain result data, and the result data is propagated to other node devices; other node devices receiving the result data take the result data as one of the collected original data, and the result data influences the result data of other node devices; based on this, without obtaining the identity information of the tourists, the plurality of node devices in the peer-to-peer network perform cooperative calculation to determine that each unique tourist is himself / herself, realize non-specific feature recognition, and perform position recognition on the tourists.

2. The method of claim 1, wherein, The position information of the walking device is obtained, and the walking route of the walking device moving to meet the tourists is calculated in real time in combination with the real-time position information of the matched tourists.

3. The method of claim 1, wherein, Action data, expression data and facial expression data of the tourists are obtained, a pre-judgment model pre-trained through machine learning is used to pre-judge the next action of the tourists, including staying, playing, resting, accelerating walking, decelerating walking or running, so as to generate a walking device control instruction to control the walking device to adjust the motion posture, action and speed of the walking device; if the next action of the tourist is not accurately pre-judged, the control instruction is corrected based on the real-time action data of the tourist, and the surrounding personnel density, environmental information, playing content, interactive content, action data, expression data and facial expression data of the tourist are taken as training samples and added to a sample library for further training and adjustment of the pre-judgment model.

4. The method of claim 1 or 3, wherein, After the walking device meets the tourist, the distance between the walking device and the tourist is judged, and environmental factors between the walking device and the tourist are combined; if the distance between the walking device and the tourist is too large, or the environmental factors form a line-of-sight obstruction or a walking obstacle between the walking device and the tourist, the walking device waits for the tourist to approach, or the tourist approaches to eliminate the line-of-sight obstruction; or the walking device stops moving until the line-of-sight obstruction or the walking obstacle is eliminated. Or before the line-of-sight obstruction or the walking obstacle is formed, the walking device waits for the tourist to approach to a distance at which the line-of-sight obstruction cannot be formed.

5. The method of claim 4, wherein, The environmental factors include moving objects and fixed objects. When the moving object forms a visual obstruction between the walking device and the visitor, the walking device waits for the visitor to approach until the visual obstruction is eliminated, or the walking device stops moving until the visual obstruction is eliminated; When the moving object forms a walking obstacle between the walking device and the visitor, the walking device stops moving until the visual obstruction is eliminated; When the fixed object forms a visual obstruction between the walking device and the visitor, the walking device waits for the visitor to approach until the visual obstruction is eliminated; Before the moving object forms a visual obstruction or a walking obstacle between the walking device and the visitor, the walking device waits for the visitor to approach to a distance where the visual obstruction cannot be formed; Before the fixed object forms a visual obstruction between the walking device and the visitor, the walking device waits for the visitor to approach to a distance where the visual obstruction cannot be formed.

6. The method of claim 5, wherein, When the moving object moves between the walking device and the visitor, forming a visual obstruction or a walking obstacle; when the walking device moves and the visitor is located on both sides of the fixed object, and the fixed object forms a visual blind area for the visitor's visual area towards the walking device, a visual obstruction is formed.

7. The method of claim 1, wherein, If the visitor changes the route requirement by interacting with the artificial intelligence guide, the walking device obtains the latest route requirement, and calculates the optimal navigation route in real time based on the real-time position information of the visitor and the latest route requirement.

8. The method of claim 1, wherein, Alternatively, the interaction platform does not directly obtain the route requirement of the visitor; the interaction platform customizes the route requirement for the current visitor based on the interaction information between the artificial intelligence guide and the visitor, including the selection of scenic spots, the visiting order of scenic spots, and the estimated playing time of each scenic spot.

9. The method of claim 1 or 8, wherein, According to the route requirements of all visitors, combined with the real-time position information, route requirements of all visitors, and the current queue number, the queue number of the scenic spot contained in the route requirement of the current visitor at the estimated arrival time of the current visitor is calculated; If the estimated waiting time corresponding to the queue number of the current visitor at the estimated arrival time is greater than the waiting time threshold, the visitor is proposed a route requirement change suggestion to adjust the visiting order of the scenic spots, for guiding the visitor to other scenic spots with an estimated waiting time less than the original next scenic spot.

10. The method of claim 9, wherein, Alternatively, if the estimated waiting time corresponding to the queue number of the current visitor at the estimated arrival time is greater than the waiting time threshold, the interaction platform actively modifies the route requirement, and the artificial intelligence guide adjusts the guide words, changes the connection words connecting the original next scenic spot to the connection words corresponding to the next scenic spot of the latest route requirement, and guides the visitor to the next scenic spot of the latest route requirement.

11. The method of claim 10, wherein, The guide words and the connection words are generated in real time according to the personality characteristics, interest data, knowledge background, and tourism experience of the visitor.

12. The method of claim 9, wherein, The route demand change suggestion is made based on the combination of multiple tourists, and the navigation route corresponding to the route demand is optimized. Specifically, according to the authorization of the tourists, part or all of the tourist information is obtained, including the tourist social platform public information, the tourist age, the residence region, the work type, the educational background, and the travel history obtained from the government platform and the enterprise platform, the language habit of the tourist interaction, and the interaction content. A route adjustment model trained in advance through machine learning is used to determine the most likely accepted route adjustment scheme of the multiple tourists, and a route demand change suggestion for the multiple tourists is generated.

13. The method of claim 12, wherein, After the first round of interaction between the tourists and the interactive platform about the route demand change suggestion, for the tourists who accept the route demand change suggestion and the tourists who express complete rejection of the route demand change suggestion, no route demand change suggestion is actively proposed within the preset time or the best time determined by the route adjustment model based on the aforementioned obtained tourist information. In the second round of interaction, the feedback opinions of the tourists who do not accept the route demand change suggestion, the accepted route demand change suggestion, and the route demand change suggestion expressed as complete rejection are taken as the parameters of the navigation route optimization, and the queuing situation of each scenic spot, the regional density, the road density, and the tourist information are combined. On the basis of the tourists who disagree with the previous route demand change suggestion but do not express complete rejection of the route demand change suggestion, a batch of tourists are selected as the target of the navigation route optimization of this round, and the best persuasion statements are generated for each tourist to persuade the tourist to accept the route demand change suggestion.

14. The method of claim 11 or 12, wherein, The personality characteristics, interest data, knowledge background, and travel experience of the tourists, or the tourist information, are obtained through interaction with the artificial intelligence guide, or are obtained by browsing the publicly available online information of the tourists, or are obtained through the authorization of the tourists in an accessible data interaction manner.

15. The method of claim 5, wherein, A peer-to-peer network is used to identify non-specific features and locations of targets, including tourists, walking devices, or moving objects.

16. The method of claim 15, wherein, The current node device receives the result data output by other node devices; for the current node device, the collected perception data is combined with the result data from other node devices to calculate the result data of the current node device, and the result data is sent to other node devices; The node devices in the peer-to-peer network perform collaborative calculation as the perception data is collected and the result data is calculated.

17. The method of claim 15, wherein, In the peer-to-peer network, for a certain point sample of a target, the result data transmitted from the node device collecting the point sample to other node devices is adjusted by the subsequent node devices according to the characteristics of the point sample, or the characteristics of the point sample are reported to the subsequent node devices to adjust the perception attention; if the subsequent other node devices do not detect the characteristics of the point sample, but from the characteristics of other point samples it can be determined that the undetected characteristics of the point sample still belong to the target, then the undetected characteristics of the point sample are still expressed in the result data of the current node device and transmitted to other node devices.

18. The method of claim 17, wherein, The method for reporting the characteristics of the point sample for subsequent node device to adjust the perceptual attention is: adjusting the parameters of the data processing model of the subsequent node device according to the result data expressing the characteristics of the point sample provided by the previous node device or the characteristics of the point sample, so that the subsequent node device improves the computing power for identifying the characteristics of the point sample; or the subsequent node device uses the perceptual attention model to match the received characteristics of the point sample or the result data expressing the characteristics of the point sample to adjust the computing power.

19. The method of claim 18, wherein, When the node device processes the result data output by several previous node devices, based on the data processing model, when the targets described by the several previous node devices can be determined as the same target through certain common point sample characteristics, the point sample characteristics and other information described by each node device are combined into the same target.

20. The method of claim 19, wherein, When the result data received by the node device shows that the mark used by the current node device to identify the target before the current time of receiving the result data is different from the mark used by other node devices to identify the target, and the mark allocated by other node devices for the target is updated, the mark used by the current node device to identify the target before the current time of receiving the result data is converted.

21. The method of claim 19, wherein, The method for converting the mark used by the current node device to identify the target before the current time of receiving the result data is: replacing the mark used by the current node device to identify the target before the current time of receiving the result data with the latest mark allocated by other node devices for the target; or, recording the conversion relationship between the mark used by the current node device to identify the target before the current time of receiving the result data and the updated mark allocated by other node devices for the target, and converting when the result data currently received by the current node device is needed; or, the node device deploys a conversion model to convert the marks of multiple targets according to the input original data or result data.

22. The method of claim 17, wherein, For one or more point samples collected by node devices at different collection positions in sequence, if the characteristic values of a certain point sample or multiple point samples at different collection positions respectively meet the preset proximity condition or are determined by a certain model to have relevance reaching a threshold, and are unique at each collection position, it is determined that the point sample of the certain type at different collection positions has relevance.

23. The method of claim 17, wherein, For one or more point samples collected by node devices at different collection positions simultaneously, if the node devices at different collection positions collect the same space field, when there is only a unique target in the space field, or the collected point samples can correctly point to one of the multiple targets to which they belong, for a certain target, the one or more point samples collected by the node devices at different collection positions have relevance.

24. The method of claim 23, wherein, The data acquisition device of the node device comprises one or a combination of image acquisition device, electromagnetic induction device, temperature measuring device, vibration frequency sensing device and laser radar. The data collected by the above devices and the three-dimensional point cloud collected by the laser radar or the point cloud generated according to the images collected by the plurality of image acquisition devices are jointly calculated to obtain a three-dimensional point with data. The image color, contour, line, reflectivity, motion trend, electromagnetic feature, temperature, temperature change trend, vibration frequency and vibration frequency change trend based on two-dimensional sensing are taken as additional attributes of the corresponding three-dimensional point to form a three-dimensional point cloud with attributes. The correspondence between each part or associated part of the 3D appearance of the target and each region of the three-dimensional point cloud with attributes is determined in combination with the change characteristics of electromagnetic induction, temperature, vibration frequency, motion correlation and reflectivity.

25. The method of claim 15, wherein, When the identity information of the target needs to be acquired, an identity information acquisition command is triggered, the identity information acquisition command is taken as one of the inputs for the calculation of the result data of the node device, and the identity information of the target is acquired by driving the node device connected in the peer-to-peer network to respond to the corresponding result data under the condition that the node device can obtain the identity information of the target.

26. The method of claim 25, wherein, The peer-to-peer network determines the permission of the target by checking the authenticity of the identity information of the target, wherein the node device capable of acquiring the identity information in the peer-to-peer network does not provide the identity information, but only expresses the checking result of the authenticity of the identity information in the result data of the node device according to the checking requirement of the authenticity of the identity information in the received result data.

27. The method of claim 26, wherein, The node device capable of acquiring the identity information in the peer-to-peer network does not provide the identity information, and the information source device providing the identity information is driven to establish an encrypted file transmission channel between the input terminal of the node device requiring the identity information and the information source device or an encrypted information transmission channel in other network communication mode; the identity information is taken as one of the inputs of the node device.

28. The method of claim 15, wherein, The data acquisition device comprises one or more of image acquisition device, audio acquisition device, temperature measuring device, vibration frequency sensing device, laser radar, chemical sensor and electromagnetic induction device.

29. The method of claim 15, wherein, The walking device is provided with at least one execution device; each execution device is connected to the peer-to-peer network through one or more node devices; the interaction platform submits a request command to the peer-to-peer network, and if it is determined based on collaborative calculation that the current node device needs to respond to the request command, the current node device will send an instruction to the execution device connected thereto according to the calculated result data to control the execution device to complete the response action.

30. The method of claim 29, wherein, The result representation corresponding to the request command is the result data; the execution device receives the result data output by the connected node device, and if a specific element in the result data indicates that the execution device needs to respond, or the result data is taken as one of the inputs of the data processing model of the node device, it is determined that the corresponding execution device needs to respond, and then the execution device performs the corresponding operation.

31. The method of claim 30, wherein, When the execution device needs to respond, the execution device combines the received result data output by other node devices, calculates its own result data, and controls the execution components on the execution device to perform corresponding operations through the obtained result data.

32. The method of claim 31, wherein, The execution device receives the result data output by other node devices, and the principle is that when the request command needs a corresponding execution device to respond, if the result data obtained by one or more node devices can determine the execution device that needs to respond, the corresponding execution device is added to the node list that delivers the current result data, and the one or more node devices directly deliver the result data to the execution device or the node device connected to the execution device. Alternatively, the execution device receives the result data output by other node devices in a layer-by-layer delivery manner.

33. The method of claim 32, wherein, According to the preset condition or algorithm output, model output, the corresponding execution device is added to the node list that delivers the result data.

34. The method of claim 30, wherein, The execution device is a node device connected to a specific function execution component, and the execution feedback information of the execution component of the execution device is fed back to the execution device to participate in the calculation of the subsequent result data of the execution device.

35. The method of claim 29, wherein, The execution device of the walking device includes but is not limited to a motion component, a brake component, a steering component, an energy component, a voice playing device, a projection device, a light device, a video playing device, a wireless communication module, and a special host; wherein the special host is used to support the augmented reality glasses and functional props worn by tourists.