Electric vehicle upstairs detection system based on community multi-monitoring linkage

Abstract

The invention discloses an electric vehicle upstairs detection system based on community multi-monitoring linkage, and relates to a safety detection system. According to the system, a plurality of high-definition cameras are used for monitoring and covering key positions such as community entrances, corridors and elevators. A YOLOv5 algorithm is adopted in the core, the electric vehicle is accurately recognized, and point cloud data are generated to improve the tracking precision of the target. Through an MQTT protocol, the system transmits recognized data to the data processing module, then the SORT algorithm is used for carrying out detailed analysis and real-time tracking on the motion trail of the electric vehicle, a multi-level alarm mechanism is introduced into the system, and when it is detected that the electric vehicle enters a corridor or an elevator, the system will inform community security guards in real time, and the safety of the electric vehicle is guaranteed. The system also has the functions of event recording and big data analysis, helps to optimize a management strategy by recording and analyzing historical data, and improves the overall efficiency of the system. The system is suitable for residence communities, commercial complexes, office buildings and other places.

Description

Technical Field

[0001] The present invention relates to a safety detection system, in particular to an electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area. Background Art

[0002] The fire risk caused by electric vehicles charging indoors during elevator ascents and descents has become a significant social issue. When charging electric vehicles indoors, fires are highly likely to occur due to inconsistent charging equipment quality and substandard charging environments. Once a fire occurs, it spreads rapidly, producing large amounts of toxic smoke, posing a significant threat to residents' lives. Furthermore, electric vehicles are large, and when pushed upstairs, they can easily block evacuation routes such as hallways and elevators, severely hindering residents' escape in emergencies. Furthermore, the frequent entry and exit of electric vehicles in elevators can damage them, increasing maintenance costs and impacting residents' ability to use them. Given these significant safety risks, an effective electric vehicle upstairs detection system is urgently needed to address these issues. Summary of the Invention

[0003] The present invention provides an electric vehicle going upstairs detection system based on multi-monitoring linkage in a residential area. The system aims to address the fire risks and safety hazards caused by electric vehicles charging upstairs.

[0004] The purpose of the present invention is achieved through the following technical solutions: An electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area, the system includes the following components: Core processing controller (PC): As the main control unit of the cell, it is responsible for real-time detection and aggregation of data streams from various sensors and devices, and displays the data on the data platform; Edge processor (RK3588): A high-performance AI edge computing chip with 6T computing power, connected to a depth camera, used to identify electric vehicles entering the community at each gate; Depth camera: Senses and records the size and shape of electric vehicles, focusing on monitoring vehicles that are prone to climbing stairs and pose safety hazards; Temperature and humidity detection module and air quality sensor: responsible for detecting the environmental information of the community and judging whether there is a fire or other dangerous situation at a certain point through changes in temperature, humidity and air quality; Sound and light alarm module: monitors fire and other dangerous information in real time and connects to the community master control unit through the Internet of Things (IoT) to achieve real-time communication and alarm prompts to warn surrounding people; Relay module: It uses 4-way relays with optocoupler isolation, which are controlled by the ESP32 control unit through relays, temperature and humidity detection modules, and sound and light alarm modules; OLED display module: uses a 2.8-inch TFT LCD screen to display the ambient temperature and humidity, air quality information, and time and date in the community in real time; RKNN3588 AI computing platform: As a visual algorithm platform, it detects the various states of the electric vehicle and sends the information to the ESP32 control unit. The ESP32 control unit controls various sensors based on the received information.

[0005] In the electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area, the core processing controller (PC computer) processes and analyzes data from each edge device in real time for centralized management and display.

[0006] The electric vehicle upstairs detection system based on multi-monitoring linkage in a residential area, the edge processor (RK3588) has 6T computing power, can quickly process image data captured by the depth camera, and identify electric vehicles through AI algorithms.

[0007] In the electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area, the depth camera can accurately identify the size and shape of the electric vehicle and transmit the feature data to the edge processor for further analysis.

[0008] The electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area, the temperature and humidity detection module can continuously monitor the environmental parameters in the residential area, determine whether there is a fire or other emergency situation through temperature changes, and transmit the information to the core processing controller.

[0009] The electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area, the sound and light alarm module is linked with the main control unit of the residential area through real-time monitoring and alarm functions, and immediately sends out sound and light alarms when a dangerous situation is detected to alert surrounding people.

[0010] In the electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area, the relay module improves the stability and safety of the system through an optical coupling isolation design, and the operation of various sensors and alarm modules is controlled by the ESP32 control unit.

[0011] The electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area, the OLED display module can not only display the ambient temperature, humidity and air quality, but also display the time and date, which is convenient for residential area managers and residents to check.

[0012] In the electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area, the RK3588 AI computing platform analyzes the status of the electric vehicle and controls the ESP32 unit to manage the sensor in real time, ensuring the efficient operation of the system.

[0013] The advantages and effects of the present invention are: The system of the present invention has excellent scalability and can be flexibly adjusted according to the scale and management needs of the community. For example, the number of cameras and their coverage can be increased to further enhance the system's monitoring capabilities. Furthermore, the system can be integrated with other community security systems, such as access control and fire protection systems, to form a more intelligent and comprehensive security management network.

[0014] From a prospective perspective, this system is not only suitable for residential communities but can also be extended to public spaces such as commercial complexes, schools, and hospitals. Market demand for this system is expected to continue to grow, especially given the growing number of electric vehicles. Through continuous optimization and upgrades, this system is expected to become a crucial component of future smart community security management, providing stronger protection for residents' lives. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Schematic diagram of the overall system structure of the present invention. DETAILED DESCRIPTION

[0016] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments: 1. System structure design The core of the system is a multi-monitoring network covering every key node in the residential complex. Specifically, the network includes multiple high-definition cameras at the residential entrance, in the corridors, and inside elevators. Each camera is equipped with a high-performance image sensor that provides clear images in all lighting conditions. Furthermore, the cameras feature intelligent analysis capabilities, enabling real-time capture and identification of dynamic conditions within the residential complex. Through these cameras, residential managers can monitor the movements of electric vehicles in real time.

[0017] At the hardware level, the system achieves comprehensive, all-around surveillance through surveillance cameras installed at strategic locations. Equipped with high-resolution lenses, these cameras capture clear images even in complex lighting conditions and transmit this information in real time to the system's edge computing nodes. These edge computing nodes utilize embedded GPUs (such as NVIDIA Jetson) to process image data in real time, effectively reducing system latency and improving response speed.

[0018] 2. Key technologies and algorithms YOLOv5 (You Only Look Once, version 5) is widely used as an advanced object detection algorithm in surveillance systems. This algorithm can detect electric vehicles with high accuracy and real-time performance. The YOLOv5 algorithm divides the input image into multiple grids and predicts the bounding box and class probability of the object within each grid, thereby achieving efficient recognition of electric vehicles.

[0019] Once an electric vehicle is detected, the system generates corresponding point cloud data. Point cloud data is a three-dimensional data representation that accurately reflects the shape, size, and position of the electric vehicle. This information is crucial for subsequent target tracking and behavioral analysis. After the point cloud data is generated, the system transmits the data to the edge processing module via the MQTT protocol. MQTT is a lightweight publish / subscribe messaging protocol that is suitable for low-bandwidth, high-latency, or unstable network environments, making it ideal for use in community monitoring systems.

[0020] In its data analysis and processing module, the system utilizes the SORT (Simple Online and Realtime Tracking) deep learning algorithm. The SORT algorithm integrates target detection results and point cloud data to accurately track electric vehicles. This algorithm analyzes the vehicle's trajectory in real time and determines whether it enters a hallway or elevator. If the system detects an electric vehicle ascending a staircase, it immediately issues an alarm.

[0021] 3. Alarm mechanism and information transmission The system incorporates a multi-tiered early warning and response mechanism. When an electric vehicle is identified and confirmed to have entered a hallway or elevator, the system immediately triggers an alarm. This alarm signal not only generates audible and visual alerts at the monitoring center but also transmits real-time alert information to security personnel's mobile devices, such as smartphones or intercoms, via wireless networks. This allows security personnel to be immediately notified of the situation and take swift action, even when they are outside the monitoring room.

[0022] Furthermore, to improve its responsiveness and flexibility, the system also possesses a degree of self-learning capability. By collecting and analyzing past recognition data, the system continuously optimizes the parameters of the YOLOv5 and SORT algorithms, thereby improving the accuracy and speed of scooter recognition. For example, through deep learning technology, the system can gradually adapt to scooters of different models, colors, and lighting conditions, reducing false positives and missed negatives.

[0023] 4. Data Recording and Analysis In addition to real-time monitoring and alarms, the system also features event logging and report generation. Whenever the system detects an electric vehicle entering a hallway or elevator, it automatically generates a detailed event report. This report includes information such as the time and location of the vehicle's identification, its trajectory, and the alarm handling process. All data is securely stored in the system database for subsequent query and analysis. This data allows community managers to better understand electric vehicle usage and develop targeted management measures.

[0024] Over the long term, the system will be able to analyze big data to uncover patterns in electric vehicle usage within a residential complex, such as peak hours and common routes. This information can not only be used to optimize the layout and strategy of the monitoring system but also provide valuable decision-making support for residential complex managers.

Claims

1. An electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area, characterized in that: The system includes the following components: Core processing controller (PC): As the main control unit of the community, it is responsible for real-time detection, aggregation of data streams from various sensors and devices, and displaying the data on the data platform; Edge processor (RK3588): A high-computing AI edge computing chip with 6T computing power, connected to a depth camera, used to identify electric vehicles entering the community at each gate of the community; Depth camera: senses and records the size and shape of electric vehicles, focusing on monitoring vehicles that are easy to go upstairs and have safety hazards; Temperature and humidity detection module and air quality sensor: responsible for detecting the environmental information of the community, judging whether there is a fire or other dangerous situation at a certain point through changes in temperature, humidity and air quality; Sound and light alarm module: real-time monitoring of dangerous information such as fire, and connecting to the community main control unit through the Internet of Things (IoT) to achieve real-time communication and alarm prompts, warning the surrounding people; Relay module: It uses 4 relays with optocoupler isolation, which are controlled by the ESP32 control unit through relays, temperature and humidity detection modules, and sound and light alarm modules; OLED display module: uses a 2.8-inch TFT LCD screen to display the ambient temperature and humidity, air quality information, and time and date in the community in real time; RKNN3588 AI computing platform: As a visual algorithm platform, after detecting various states of electric vehicles, the information is sent to the esp32 control unit, which controls various sensors based on the received information.

2. The electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area according to claim 1 is characterized in that: The core processing controller (PC) processes and analyzes data from each edge device in real time, and performs centralized management and display.

3. The electric vehicle stair-going detection system based on community multi-monitoring linkage according to claim 1 is characterized in that: The edge processor (RK 3588) has 6T computing power, which can quickly process image data captured by the depth camera and identify electric vehicles through AI algorithms.

4. The electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area according to claim 1 is characterized in that: The depth camera can accurately identify the size and shape of the electric vehicle and transmit the feature data to the edge processor for further analysis.

5. The electric vehicle stair-going detection system based on community multi-monitoring linkage according to claim 1 is characterized in that: The temperature and humidity detection module can continuously monitor the environmental parameters in the community, determine whether there is a fire or other emergency situation through temperature changes, and transmit the information to the core processing controller.

6. The electric vehicle stair-going detection system based on community multi-monitoring linkage according to claim 1 is characterized in that: The sound and light alarm module is linked with the main control unit of the community through real-time monitoring and alarm functions, and immediately sends out sound and light alarms when a dangerous situation is detected to alert people around.

7. The electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area according to claim 1 is characterized in that: The relay module improves the stability and safety of the system through the optical coupling isolation design, and the operation of various sensors and alarm modules is controlled by the ESP32 control unit.

8. The electric vehicle stair-going detection system based on multi-monitoring linkage in a residential area according to claim 1 is characterized in that: The OLED display module can not only display the ambient temperature, humidity and air quality, but also display the time and date, which is convenient for community managers and residents to view.

9. The electric vehicle stair-going detection system based on community multi-monitoring linkage according to claim 1 is characterized in that: The RK3588 AI computing platform analyzes the status of the electric vehicle and controls the esp32 unit to manage the sensors in real time to ensure the efficient operation of the system.

Images