Parking management system based on generative AI

Abstract

The invention relates to the technical field of parking management systems, and discloses a generative AI-based parking management system, which comprises an LLM (Logical Language Model), a GenAI assistant module, a memory module, an information input processing module, an execution module and a strategy making module, the GenAI assistant module is used for calling the capability of the LLM to carry out input information processing, strategy making and response generation; according to the invention, by deploying the GenAI assistant module, when an automatic system has an abnormal condition, the GenAI assistant module intervenes in communication with a vehicle owner and processes the abnormal condition, so that various requirements in operation management of the parking lot are met, and complete unmanned management is realized; through the arrangement of the manual coach module, auxiliary decision making can be carried out on the deployed GenAI assistant module, and accumulated data can be used for fine tuning training of an LLM large language model, so that smooth import and stable operation of the GenAI module are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of parking management systems, and in particular to a parking management system based on generative AI. Background Art

[0002] Traditional smart parking lots mainly use license plates to identify vehicle information, and after confirming the vehicle's access rights, the gate is opened to release. The vehicle's access rights are determined by the management program based on local database data combined with cloud server data records and other information. Vehicle drivers can pay by IC card, cash, mobile payment, etc. However, due to the complexity of the system, manual intervention is required in some cases, such as payment failure, license plate recognition failure, system failure and other reasons. At present, manual intervention is mainly achieved by setting up an interactive terminal next to the gate, and talking to the on-duty personnel through voice. The on-duty personnel solve the problem remotely or arrange personnel to solve it on site. In some abnormal cases, manual intervention is still required. Therefore, a parking management system based on generative AI is proposed. Summary of the invention

[0003] The purpose of the present invention is to provide a parking management system based on generative AI to solve the problem raised in the above background technology that the current manual intervention is mainly achieved by setting up an interactive terminal next to the gate, talking to the on-duty personnel in voice form, and the on-duty personnel solve the problem remotely or arrange personnel to solve it on site. In some abnormal situations, manual intervention is still required.

[0004] To achieve the above-mentioned object, the present invention provides the following technical solutions: a parking management system based on generative AI, comprising an LLM large language model, a GenAI assistant module, a memory module, an information input processing module, an execution module and a strategy formulation module;

[0005] The LLM large language model is used for event analysis, logical judgment, decision generation, and generation of interactive text information;

[0006] The GenAI assistant module is used to call the capabilities of the LLM large language model to process input information, formulate strategies, and generate responses;

[0007] The memory module is used to store the processed historical events as context;

[0008] The information input processing module is responsible for the input processing of two types of information;

[0009] The execution module is a component used to execute specific tasks and functions;

[0010] The strategy formulation module is used for situation analysis, priority control, task decomposition, strategy selection, strategy evaluation, strategy optimization and strategy synthesis.

[0011] Preferably, the above-mentioned LLM large language model is connected to the GenAI assistant module, and the GenAI assistant module is respectively connected to the memory module, information input processing module, execution module and strategy formulation module.

[0012] Preferably, the system further comprises a manual coaching module, an execution unit, a case processing unit and a local corpus unit;

[0013] The manual coach module is used to manually review and judge the request prepared by the GenAI assistant module, and can modify or reject the request of the GenAI assistant module;

[0014] The execution unit is used to execute the operations of the GenAI assistant module reviewed by the manual coach module;

[0015] The case processing unit is used for the manual coaching module to assist the GenAI assistant module in processing the completed events and generating cases;

[0016] The local corpus unit stores processed case information for subsequent model retraining.

[0017] Preferably, the above-mentioned GenAI assistant module is connected to the manual coaching module, the execution unit and the case processing unit respectively, the case processing unit is connected to the local corpus unit, and the local corpus unit is connected to the LLM large language model.

[0018] Preferably, the memory module comprises a short-term memory storage unit and a long-term memory storage unit;

[0019] The short-term memory storage unit is responsible for storing and processing temporary information;

[0020] The long-term memory storage unit is responsible for the storage and processing of long-term information.

[0021] Preferably, the short-term memory storage unit is connected to the GenAI assistant module, and the long-term memory storage unit is connected to the GenAI assistant module.

[0022] Preferably, the above-mentioned components include a gate control unit, a camera, a display screen, a voice device, an alarm and a mobile communication unit.

[0023] Preferably, the information input processing module comprises a user input unit and a situation awareness module;

[0024] The user input unit is used for interaction of user information;

[0025] The situation awareness module is used to collect and generate situation awareness information.

[0026] Preferably, the user input unit is connected to the GenAI assistant module, and the situation awareness module is connected to the GenAI assistant module.

[0027] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0028] 1. The present invention deploys the GenAI assistant module. When an abnormal situation occurs in the automation system, the GenAI assistant module intervenes to communicate with the car owner and handle the abnormal situation, thereby solving various needs in the operation and management of the parking lot and realizing completely unmanned management.

[0029] 2. Through the setting of the artificial coach module, the present invention can assist the decision-making of the deployed GenAI assistant module, and the accumulated data can be used for fine-tuning training of the LLM large language model, thereby realizing the smooth introduction and stable operation of the GenAI module. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 Schematic diagram of the GenAI assistant module of the present invention;

[0032] Figure 2 It is a schematic diagram of the artificial coaching module of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the conditions under which this application can be implemented. Therefore, they have no substantive technical significance. Any structural modification, change in proportional relationship or adjustment of size should still fall within the scope of the technical content disclosed in this application without affecting the effects and purposes that can be achieved by this application.

[0035] Example

[0036] See also Figure 1-2 The present invention provides a technical solution: a parking management system based on generative AI, including an LLM large language model, a GenAI assistant module, a memory module, an information input processing module, an execution module and a strategy formulation module; by deploying the GenAI assistant module, when an abnormal situation occurs in the automation system, the GenAI assistant module intervenes to communicate with the car owner and handle the abnormal situation, thereby solving various needs in the operation and management of the parking lot and realizing completely unmanned management;

[0037] The GenAI assistant module uses the LLM large language model to perform event analysis and intelligent decision-making, and can call other modules to resolve events, such as: judging that payment is successful: directly open the gate to release the vehicle; judging that license plate recognition fails, re-identify the license plate, and release the vehicle if it meets the square condition; if it does not meet the release condition, inform the reason; judging that mechanical failure occurs, contact staff to come to the scene, and remind the owner to enter or exit from other exits;

[0038] LLM large language model, used for event analysis, logical judgment, decision generation, and interactive text information generation;

[0039] The GenAI assistant module is used to call on the capabilities of the LLM large language model to process input information, formulate strategies, and generate responses;

[0040] A memory module, used to save processed historical events as context;

[0041] The information input processing module is responsible for the input processing of two types of information;

[0042] Execution modules are components used to perform specific tasks and functions;

[0043] Strategy formulation module for situation analysis, priority control, task decomposition, strategy selection, strategy evaluation, strategy optimization and strategy synthesis;

[0044] The LLM large language model is connected to the GenAI assistant module, and the GenAI assistant module is respectively connected to the memory module, information input processing module, execution module and strategy formulation module.

[0045] The system also includes a human coaching module, an execution unit, a case processing unit, and a local corpus unit;

[0046] The manual coach module is used to manually review and judge the requests prepared by the GenAI assistant module, and can modify or reject the requests of the GenAI assistant module;

[0047] An execution unit, which is used to execute the operations of the GenAI assistant module that have been reviewed by the human coach module;

[0048] The case processing unit is used by the manual coach module to assist the GenAI assistant module in processing completed events and generating cases;

[0049] The local corpus unit stores processed case information for subsequent model retraining.

[0050] The GenAI assistant module is connected to the manual coach module, the execution unit, and the case processing unit respectively. The case processing unit is connected to the local corpus unit, and the local corpus unit is connected to the LLM large language model. Through the setting of the manual coach module, the deployed GenAI assistant module can be assisted in decision-making, and the accumulated data can be used for fine-tuning training of the LLM large language model, thereby realizing the smooth introduction and stable operation of the GenAI module.

[0051] Functions of the manual coaching module:

[0052] When the manual coach module is activated, the operation request prepared by the GenAI assistant module is presented to the manual coach through the manual coach module. The manual coach is responsible for reviewing and judging the request prepared by the AI. The manual coach can modify or reject the request of the GenAI assistant module. Only the operation of the GenAI assistant module reviewed by the manual coach will continue to be executed.

[0053] After the operation request of the GenAI assistant module is executed, the effect is evaluated by the manual coaching module or other modules. If the evaluation is passed, it is stored in the local corpus unit, which is used for subsequent model retraining;

[0054] After accumulating a certain amount of local corpus, the LLM large language model is retrained using the local corpus to optimize the local adaptability of the model and improve the response capabilities of the GenAI assistant module;

[0055] The manual coach module can be enabled or disabled, depending on whether the LLM large language model and the GenAI assistant module have the required processing capabilities for local scenarios. When the manual coach module is disabled, the GenAI assistant module will handle events independently without relying on the manual coach.

[0056] The memory module includes a short-term memory storage unit and a long-term memory storage unit;

[0057] Short-term memory storage unit, responsible for the storage and processing of temporary information;

[0058] Long-term memory storage unit is responsible for the storage and processing of long-term information.

[0059] The short-term memory storage unit is connected to the GenAI assistant module, and the long-term memory storage unit is connected to the GenAI assistant module.

[0060] The components include a gate control unit, camera, display, voice equipment, alarm and mobile communication unit; the GenAI assistant module can perform intelligent situational awareness of the parking lot operation status, and promptly issue alarms or open or close the parking lot gate. If a fire is sensed, the GenAI assistant module will automatically open the gate to facilitate the escape of vehicles and personnel or the entry of rescue vehicles.

[0061] The information input processing module includes a user input unit and a situation awareness module;

[0062] A user input unit, used for interaction of user information;

[0063] The situation awareness module is used to collect and generate situation awareness information.

[0064] The user input unit is connected to the GenAI assistant module, and the situation awareness module is connected to the GenAI assistant module.

[0065] In summary, the present invention deploys the GenAI assistant module. When an abnormal situation occurs in the automation system, the GenAI assistant module intervenes to communicate with the car owner and handle the abnormality, thereby solving various needs in the operation and management of the parking lot and realizing completely unmanned management. Through the setting of the artificial coach module, the deployed GenAI assistant module can be assisted in decision-making, and the accumulated data can be used for fine-tuning training of the LLM large language model, thereby realizing the smooth introduction and stable operation of the GenAI module.

[0066] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present invention may be combined and / or combined in various ways, even if such combinations and / or combinations are not explicitly described in the present invention. In particular, the features described in the various embodiments and / or claims of the present invention may be combined and / or combined in various ways without departing from the spirit and teachings of the present invention. All of these combinations and / or combinations fall within the scope of the present invention.

Claims

1. A parking management system based on generative AI, characterized by: It includes the LLM large language model, GenAI assistant module, memory module, information input processing module, execution module and strategy formulation module; The LLM large language model is used for event analysis, logical judgment, decision generation, and generation of interactive text information; The GenAI assistant module is used to call the capabilities of the LLM large language model to process input information, formulate strategies, and generate responses; The memory module is used to store the processed historical events as context; The information input processing module is responsible for the input processing of two types of information; The execution module is a component used to execute specific tasks and functions; The strategy formulation module is used for situation analysis, priority control, task decomposition, strategy selection, strategy evaluation, strategy optimization and strategy synthesis.

2. A parking management system based on generative AI according to claim 1, characterized in that: The LLM large language model is connected to the GenAI assistant module, and the GenAI assistant module is respectively connected to the memory module, information input processing module, execution module and strategy formulation module.

3. The parking management system based on generative AI according to claim 1, characterized in that: The system also includes a manual coaching module, an execution unit, a case processing unit, and a local corpus unit; The manual coach module is used to manually review and judge the request prepared by the GenAI assistant module, and can modify or reject the request of the GenAI assistant module; The execution unit is used to execute the operations of the GenAI assistant module reviewed by the manual coach module; The case processing unit is used for the manual coaching module to assist the GenAI assistant module in processing the completed events and generating cases; The local corpus unit stores processed case information for subsequent model retraining.

4. A parking management system based on generative AI according to claim 3, characterized in that: The GenAI assistant module is connected to the manual coaching module, the execution unit and the case processing unit respectively, the case processing unit is connected to the local corpus unit, and the local corpus unit is connected to the LLM large language model.

5. The parking management system based on generative AI according to claim 1, characterized in that: The memory module includes a short-term memory storage unit and a long-term memory storage unit; The short-term memory storage unit is responsible for storing and processing temporary information; The long-term memory storage unit is responsible for the storage and processing of long-term information.

6. A parking management system based on generative AI according to claim 5, characterized in that: The short-term memory storage unit is connected to the GenAI assistant module, and the long-term memory storage unit is connected to the GenAI assistant module.

7. A parking management system based on generative AI according to claim 1, characterized in that: The components include a gate control unit, a camera, a display screen, a voice device, an alarm and a mobile communication unit.

8. The parking management system based on generative AI according to claim 1, characterized in that: The information input processing module includes a user input unit and a situation awareness module; The user input unit is used for interaction of user information; The situation awareness module is used to collect and generate situation awareness information.

9. A parking management system based on generative AI according to claim 8, characterized in that: The user input unit is connected to the GenAI assistant module, and the situation awareness module is connected to the GenAI assistant module.

Images