Search

CN-122019891-A - Scenic spot personalized AI navigation system and method

CN122019891ACN 122019891 ACN122019891 ACN 122019891ACN-122019891-A

Abstract

The invention provides a scenic spot personalized AI navigation system and method, the method comprises the steps of collecting movement factors, displacement scales and interaction marks of tourists in a navigation process, constructing a tourist behavior feature vector, generating corresponding rhythm scores based on the tourist behavior feature vector and combining with continuity constraints based on historical displacement scales, mapping the rhythm scores into discrete navigation rhythm states, inquiring a preset strategy template set according to the rhythm scores and the navigation rhythm states, generating corresponding navigation strategy instructions, and selecting and outputting matched navigation content versions from a local content library based on the navigation strategy instructions through preset mapping rules. The invention constructs a floor-type, configurable and expandable intelligent navigation control scheme, and provides a technical path which is more in line with actual operation conditions for scenic spot personalized navigation service.

Inventors

  • LAI JUN
  • LIU XIAOHUA
  • Li Nengpan

Assignees

  • 广州翼然科技股份有限公司

Dates

Publication Date
20260512
Application Date
20260128

Claims (10)

  1. 1. A scenic spot personalized AI navigation method, the method comprising: The method comprises the steps of collecting a movement factor, a displacement scale and an interaction mark of a tourist in a tour process, and constructing a tourist behavior feature vector, wherein the movement factor is used for representing the movement intensity level of the tourist, the displacement scale is used for representing the spatial variation amplitude of the tourist, and the interaction mark is used for representing whether the tourist has active interaction behavior or not; Based on the tourist behavior feature vector, generating a corresponding rhythm score by combining with a continuity constraint based on a history displacement scale and mapping the rhythm score into a discrete tour rhythm state, wherein the rhythm score is used for reflecting the slight difference of the attention degree of the tourist in the current state; Inquiring a preset strategy template set according to the rhythm score and the tour rhythm state to generate a corresponding tour strategy instruction, wherein the tour strategy instruction comprises the tour rhythm state and a strategy intensity factor, and the strategy intensity factor group is used for indicating the intensity of tour execution under the corresponding tour rhythm state; And selecting and outputting the matched navigation content version from the local content library through a preset mapping rule based on the navigation strategy instruction.
  2. 2. The scenic spot personalized AI guide method according to claim 1, wherein the displacement scale is obtained based on current position coordinates respectively obtained by two continuous time windows and a spatial distance between two position points; the movement factor is obtained by calculation based on the displacement scale, a corresponding time window and a preset reference value; The interactive mark is generated by an event recording mechanism in the navigation application, when tourists have navigation related operations for any time in the current time window, the interactive mark is endowed with effective value, and if no navigation interactive behavior occurs, the interactive mark is kept as initial value.
  3. 3. The scenic spot personalization AI navigation method of claim 2, wherein the navigation-related operation includes clicking on a scenic spot introduction entry or triggering a voice explanation.
  4. 4. The method of claim 1, wherein the cadence score is obtained by moving factor terms, displacement scale terms, interactive mark terms, and a continuous constraint weight sum for suppressing displacement mutation, and then by logic function compression.
  5. 5. The scenic spot personalized AI navigation method of claim 4, wherein the cadence score maps to a discrete tour cadence state, in particular: A hierarchical threshold decision is employed based on a predefined tempo status label and when the interaction flag has active interactions, the threshold will move down accordingly so that the same tempo score is more prone to be decided as the status of the output navigation content.
  6. 6. The scenic spot personalization AI navigation method of claim 4, wherein the movement factor term is obtained based on the movement factor calculation, the displacement scale term is obtained based on the displacement scale calculation, and the interaction mark term is generated based on the interaction mark calculation.
  7. 7. The scenic spot personalized AI navigation method according to claim 1, wherein the querying a preset policy template set according to the rhythm score and the tour rhythm state generates a corresponding navigation policy instruction, specifically: determining an available navigation strategy type set according to the corresponding navigation rhythm state; Calculating a strategy intensity factor according to the offset of the rhythm score relative to a preset reference threshold value of the current tour rhythm state, wherein the calculation mode of the strategy intensity factor is that the rhythm score is divided by the preset regulating interval width after subtracting the reference threshold value of the current rhythm state, and the result is limited between 0 and 1; and combining the tour rhythm state with the strategy intensity factor to generate a tour strategy instruction.
  8. 8. The scenic spot personalized AI navigation method according to claim 1, wherein the selecting and outputting the matched navigation content version from the local content library based on the navigation policy instruction through a preset mapping rule specifically comprises: screening a guiding content version set applicable to a content library according to the guiding rhythm state in the guiding strategy instruction; selecting a specific content hierarchy from the navigation content version set according to the policy intensity factor in the navigation policy instruction; And loading the resources of the selected content hierarchy and outputting the resources in the form of voice, graphics context or multimedia according to the terminal capability.
  9. 9. The scenic spot personalization AI navigation method of claim 8, wherein the selecting a particular content hierarchy from the navigation content version set is: and multiplying the strategy intensity factor by the total number of available content levels in the corresponding tour rhythm state by one, and then rounding down to obtain the corresponding content level index.
  10. 10. A scenic spot personalized AI navigation system, the system comprising: The system comprises a tourist real-time behavior feature construction module, a tourist behavior feature vector generation module and a tourist real-time behavior feature analysis module, wherein the tourist real-time behavior feature construction module is used for acquiring a movement factor, a displacement scale and an interaction mark of the tourist in the tourist process and constructing the tourist behavior feature vector, the movement factor is used for representing the movement intensity level of the tourist, the displacement scale is used for representing the spatial variation amplitude of the tourist, and the interaction mark is used for representing whether the active interaction of the tourist occurs or not; The system comprises a tourist rhythm state construction module, a tourist rhythm state mapping module and a control module, wherein the tourist rhythm state construction module is used for generating corresponding rhythm scores based on the characteristic vector of the tourist behavior and combining with continuity constraint based on a history displacement scale and mapping the rhythm scores into discrete tourist rhythm states; The system comprises a navigation strategy instruction construction module, a navigation strategy instruction generation module and a navigation strategy instruction generation module, wherein the navigation strategy instruction construction module is used for inquiring a preset strategy template set according to the rhythm score and the navigation rhythm state to generate a corresponding navigation strategy instruction, and the navigation strategy instruction comprises the navigation rhythm state and a strategy intensity factor; And the navigation strategy instruction output module is used for selecting and outputting the matched navigation content version from the local content library through a preset mapping rule based on the navigation strategy instruction.

Description

Scenic spot personalized AI navigation system and method Technical Field The invention belongs to the field of personalized AI navigation, and particularly relates to a scenic spot personalized AI navigation system and method. Background In the existing scenic spot navigation service, the navigation mode still mainly depends on manual explanation, voice navigation equipment for fixing content or mobile terminal application based on position triggering. Such approaches are common in practical applications with problems of stiffness in response, single tempo and insufficient adaptation. The manual explanation is limited by human resources and time arrangement, large-scale tourists are difficult to cover and continuous service is provided, traditional voice guiding equipment mostly adopts a preset sequence or timing playing mode, the explanation time and content depth cannot be adjusted according to the actual tourist behaviors, and the guiding application based on positioning usually only takes the position as a triggering condition although the convenience of information acquisition is improved, and different tourist states possibly presented by the tourists at the same position, such as rapid passing, short stay or long residence, are ignored, so that the guiding content is not matched with the actual requirements of the tourists. Meanwhile, the real scenic spot environment often has the conditions of complex space structure, more shielding, crowds, and the like, the positioning data is easy to generate short-time fluctuation, and the stability problem of driving navigation under a single triggering condition is further amplified. Under the background that tourists form increasingly diversified, the difference of different tourists in the information receiving rhythm, focus and tour mode is more obvious, and the existing tour guide system lacks an overall technical scheme capable of stably understanding the current tour state of the tourists on the premise of not depending on complex sensing equipment and controlling the tour guide content output mode according to the current tour state. Therefore, how to effectively match the navigation rhythm with the tourist status by using the behavior information acquired by the terminal under the condition of the real scenic spot is still a key problem which is not solved in the prior art. Disclosure of Invention The invention aims to provide a scenic spot personalized AI navigation system and method, which solve the problems. In order to achieve the above object, in a first aspect of the present invention, there is provided a scenic spot personalized AI-navigation method, the method comprising the steps of: The method comprises the steps of collecting a movement factor, a displacement scale and an interaction mark of a tourist in a tour process, and constructing a tourist behavior feature vector, wherein the movement factor is used for representing the movement intensity level of the tourist, the displacement scale is used for representing the spatial variation amplitude of the tourist, and the interaction mark is used for representing whether the tourist has active interaction behavior or not; Based on the tourist behavior feature vector, generating a corresponding rhythm score by combining with a continuity constraint based on a history displacement scale and mapping the rhythm score into a discrete tour rhythm state, wherein the rhythm score is used for reflecting the slight difference of the attention degree of the tourist in the current state; Inquiring a preset strategy template set according to the rhythm score and the tour rhythm state to generate a corresponding tour strategy instruction, wherein the tour strategy instruction comprises the tour rhythm state and a strategy intensity factor, and the strategy intensity factor group is used for indicating the intensity of tour execution under the corresponding tour rhythm state; And selecting and outputting the matched navigation content version from the local content library through a preset mapping rule based on the navigation strategy instruction. Further, the displacement scale is obtained by calculation based on the current position coordinates respectively acquired by two continuous time windows and the space distance between two position points; the movement factor is obtained by calculation based on the displacement scale, a corresponding time window and a preset reference value; The interactive mark is generated by an event recording mechanism in the navigation application, when tourists have navigation related operations for any time in the current time window, the interactive mark is endowed with effective value, and if no navigation interactive behavior occurs, the interactive mark is kept as initial value. Further, the navigation-related operation includes clicking on a sight introduction entry or triggering a voice explanation. Further, the rhythm score is obtained by weighting and summing a moving