Closed area vehicle management method, device and system

Abstract

The invention relates to the technical field of vehicle management, in particular to a closed area vehicle management method, device and system, the method is applied to a control unit in a closed area vehicle management system, and the system further comprises a spherical camera unit and a multi-mode sensing unit; the method comprises the following steps: acquiring image data acquired by the spherical camera unit and multi-modal sensing data acquired by the multi-modal sensing unit; through fusion of the spherical camera unit and multi-modal sensing data, high-precision sensing of target vehicle and parking space information in a closed area is realized, and the problems of low recognition precision and easy environmental interference caused by dependence on a single camera or sensor traditionally are effectively overcome. In combination with the pre-configuration rule and vehicle real-time position analysis, the system can more accurately and stably judge the vehicle state, generate interpretable management data and feed back the management data to the remote terminal, so that the reliability, the intelligent level and the operation efficiency of closed area vehicle management are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle management, in particular to a closed area vehicle management method, device and system. BACKGROUND

[0002] Under the background of rapid development of intelligent transportation system and smart city construction, vehicle management and navigation in closed areas are increasingly valued. Traditional vehicle management systems mainly rely on manual scheduling or single sensor technology, which is difficult to realize dynamic path planning and real-time monitoring of parking space status in complex environments. Especially in the absence of prior map information, how to efficiently and accurately guide vehicles to the target location becomes a technical difficulty.

[0003] Traditional parking detection methods rely on cameras or single type sensors, which have low recognition accuracy and are greatly affected by the environment. In order to improve the efficiency and automation level of vehicle scheduling in closed areas, it is necessary to build an intelligent management system integrating multiple sensing devices to realize comprehensive monitoring of vehicle behavior and fine management of parking resources. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a closed area vehicle management method, device and system.

[0005] In a first aspect, the present application provides a closed area vehicle management method, which is applied to a control unit in a closed area vehicle management system. The system also includes a spherical camera unit and a multi-modal sensing unit. The method comprises: acquiring image data collected by the spherical camera unit and multi-modal sensing data collected by the multi-modal sensing unit; acquiring current environmental parameters, which at least include one of illumination, weather conditions and visibility; based on the current environmental parameters, dynamically calculating the fusion weight of each type of sensing data in the multi-modal sensing unit; according to the fusion weight, fusing the image data and the multi-modal sensing data to construct the association relationship between the target vehicle and the target parking space; combining the preconfigured rules corresponding to the target parking space and the position data of the target vehicle, generating management data and feeding back to the remote terminal.

[0006] In combination with the first aspect, the step of dynamically calculating the fusion weight of each type of sensing data in the multi-modal sensing unit based on the current environmental parameters comprises: performing data cleaning and standardization processing on the acquired current environmental parameters to obtain processed current environmental parameters; inputting the preprocessed environmental parameters into a dynamic weight calculation model to output the initial weight of each type of sensing data; smooth the calculated initial weights to suppress weight mutation; normalize the smoothed weights to obtain fusion weights of the sensing data; wherein the sum of the fusion weights of the sensing data is 1.

[0007] With reference to the first aspect, the step of fusing the image data and the multi-modal sensing data according to the fusion weights to construct the association between the target vehicle and the target parking space comprises: using the fusion weights to perform weighted fusion on the multi-modal sensing data from the plurality of geomagnetic sensors, the infrared sensors and the ultrasonic sensors, and calculating a comprehensive occupancy confidence of each parking space; according to the comprehensive occupancy confidence, screening candidate target parking spaces with a confidence higher than a threshold from all the parking spaces; triggering the spherical camera unit to rotate to align with the target parking space to identify a target vehicle on the target parking space and a recognition result corresponding to the target vehicle; the recognition result comprises position data of the target vehicle, vehicle features and vehicle identification; binding the position data of the target vehicle, the vehicle features and the vehicle identification with identification information of the target parking space to construct the association between the target vehicle and the target parking space.

[0008] With reference to the first aspect, before the step of fusing the image data and the multi-modal sensing data according to the fusion weights to construct the association between the target vehicle and the target parking space, the method further comprises: judging whether the current environmental parameters satisfy a fusion strategy switching condition according to a preset rule or model; if yes, selecting a target fusion strategy from a plurality of pre-stored fusion strategies based on the current environment and the running state of the multi-modal sensing unit; switching from the current fusion strategy to the target fusion strategy in a smooth switching manner.

[0009] With reference to the first aspect, the step of switching from the current fusion strategy to the target fusion strategy in a smooth switching manner comprises: gradually adjusting a mixing ratio of the current fusion strategy and the target fusion strategy within a preset time window until the target fusion strategy is completely transitioned to.

[0010] With reference to the first aspect, the step of switching from the current fusion strategy to the target fusion strategy in a smooth switching manner comprises: running the current fusion strategy and the target fusion strategy in parallel during the switching period; comparing output results of the two strategies, and if the output of the target fusion strategy is stable and meets expectations, stopping running the current fusion strategy and completely switching to the target fusion strategy.

[0011] With reference to the first aspect, the method further comprises: In response to the path planning request, a path starting point and an ending point in the request are acquired; In combination with a preset map of the closed area, the starting point and the ending point, a plurality of planning data are outputted; Based on one or more optimization targets in the request, multi-objective optimization is performed on the plurality of planning data to obtain a target planning path.

[0012] In combination with a preset map of the closed area, the starting point and the ending point, a plurality of planning data are outputted; Acquire the current position data and the current driving path of the target vehicle; Acquire the path overlap rate of the sub-path between the current driving path and the path starting point to the current position data in the target planning path; If the path overlap rate is less than a preset threshold, a warning prompt information is generated and fed back.

[0013] Secondly, the application embodiment also provides a closed area vehicle management device for a control unit in a closed area vehicle management system, and the system further includes a spherical camera unit and a multi-modal sensing unit; the device includes: A first acquisition module is configured to acquire image data collected by the spherical camera unit and multi-modal sensing data collected by the multi-modal sensing unit; A second acquisition module is configured to acquire a current environment parameter, and the current environment parameter at least includes one of illumination, weather condition and visibility; A dynamic calculation module is configured to dynamically calculate a fusion weight of each type of sensing data in the multi-modal sensing unit based on the current environment parameter; An association module is configured to fuse the image data and the multi-modal sensing data according to the fusion weight, and construct an association relationship between the target vehicle and the target parking space; A generation module is configured to generate management data in combination with a preconfigured rule corresponding to the target parking space and position data of the target vehicle, and feed back the management data to a remote terminal.

[0014] Thirdly, the application embodiment also provides a closed area vehicle management system, and the system includes: The spherical camera unit includes a first type of spherical camera and a second type of spherical camera; the first type of spherical camera is configured to collect images of vehicles entering the closed area; and the second type of spherical camera is configured to collect images of parking spaces in the closed area; The multi-modal sensing unit includes a geomagnetic sensor, an infrared sensor and an ultrasonic sensor; The control unit is configured to receive the collected data sent by the spherical camera unit and the multi-modal sensing unit, so as to identify target vehicle information on a target parking space in the closed area; The spherical camera unit and the multi-modal sensing unit are respectively in communication connection with the control unit, and the control unit is configured to execute the method as described above.

[0015] The application provides a closed area vehicle management method, device and system, the method is applied to a control unit in a closed area vehicle management system, and the system further comprises a spherical camera unit and a multi-modal sensing unit; the method comprises the following steps: acquiring image data collected by the spherical camera unit and multi-modal sensing data collected by the multi-modal sensing unit; acquiring a current environment parameter, the current environment parameter at least comprising one of illumination, weather condition and visibility; dynamically calculating a fusion weight of various types of sensing data in the multi-modal sensing unit based on the current environment parameter; fusing the image data and the multi-modal sensing data according to the fusion weight, and constructing an association relationship between a target vehicle and a target parking space; combining a preconfigured rule corresponding to the target parking space and position data of the target vehicle, generating management data and feeding back to a remote terminal.

[0016] The closed area vehicle management method provided by the application dynamically calculates the fusion weight of multi-modal sensing data based on real-time acquired environment parameters such as illumination, weather and visibility, then fuses the image data and the multi-modal sensing data according to the dynamic weight, constructs an accurate association relationship between the target parking space and the target vehicle, finally generates a management decision combining the preconfigured rule and feeds back to the terminal, and through the introduction of the environment perception and dynamic adaptive fusion mechanism, the accuracy and stability of vehicle management in a complex environment are significantly improved.

[0017] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application will be realized and achieved by the structure particularly pointed out in the description, claims and drawings.

[0018] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the following preferred embodiments are specifically described below, and the accompanying drawings are described in detail as follows. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0020] Figure 1 A closed area vehicle management method flow chart provided by the embodiments of the present application; Figure 2Another flow chart of a vehicle management method for an enclosed area according to an embodiment of the present application is provided. Figure 3 A structure diagram of a vehicle management device for an enclosed area according to an embodiment of the present application is provided. Figure 4 An electronic device structure diagram according to an embodiment of the present application is provided.

[0021] Reference signs: 10 - first acquisition module, 20 - second acquisition module, 30 - dynamic calculation module, 40 - association module, 50 - generation module. 130 - processor, 131 - memory, 132 - bus, 133 - communication interface. DETAILED DESCRIPTION

[0022] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be described below in detail with reference to the accompanying drawings. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0023] To facilitate the understanding of the present embodiment, the application scenario and design idea of the present application embodiment will be briefly introduced first.

[0024] Traditional parking space detection methods mostly rely on cameras or single type sensors, and have problems such as low recognition accuracy and great influence from environment.

[0025] Based on this, the present application provides a vehicle management method, device and system for an enclosed area.

[0026] Embodiment 1 The present application provides a vehicle management method for an enclosed area, which is applied to a control unit in a vehicle management system for an enclosed area. The system further includes a spherical camera unit and a multi-modal sensor unit.

[0027] In combination with Figure 1 As shown in the figure, the method comprises: S110, acquiring image data collected by the spherical camera unit and multi-modal sensor data collected by the multi-modal sensor unit.

[0028] S120, acquiring current environment parameters, which at least include one of illumination, weather condition and visibility.

[0029] S130, dynamically calculating fusion weights of various types of sensor data in the multi-modal sensor unit based on the current environment parameters.

[0030] S140, according to the fusion weight, fusing the image data and the multi-modal sensor data, and constructing the association relationship between the target vehicle and the target parking space.

[0031] S150, combining the pre-configured rule corresponding to the target parking space and the position data of the target vehicle, generating management data and feeding back to the remote terminal.

[0032] The method provided in the application overcomes the sensing limitations of the traditional scheme in a harsh environment by acquiring environmental parameters such as illumination, weather, and visibility in real time, dynamically calculating the fusion weight of multi-modal sensor data based on the environmental parameters, and further adopting the collaborative working mode of "sensing preliminary screening + visual confirmation" that can be realized by fusing image data and multi-modal sensor data according to the dynamic weight, thereby ensuring the accurate association between the target vehicle and the target parking space, optimizing the resource utilization, and forming a complete closed loop from environmental perception, data fusion to decision management, thereby providing a more intelligent, efficient, and reliable vehicle management solution for the closed area.

[0033] In step S110, the spherical camera unit includes a first type of spherical camera and a second type of spherical camera, each of which is used to capture image information within a field of view. When a vehicle is detected to enter or leave, image acquisition is triggered, and the acquired image is transmitted to the control unit. The control unit processes the image, including license plate recognition, vehicle type recognition, etc. The field of view of different spherical cameras may have some overlap. By increasing the number of spherical cameras, the closed area can be divided into multiple sub-areas to improve the sensitivity of monitoring and management.

[0034] In this embodiment, the first type of spherical camera is used as the main camera to capture images of the vehicle and transmit them to the control unit, which processes the images to identify vehicle information (including at least license plate code, vehicle color, vehicle type information, etc.); the second type of spherical camera is used as an auxiliary camera to capture scene images within the closed area and send them to the control unit, which processes the images to obtain scene information within the sub-area (including at least illumination parameters, position data of each parking space in the sub-area, object contour data in the parking space, etc.).

[0035] The shell of each ball machine includes a bottom shell and a side shell. The bottom shell has a groove opened in the top surface in the circumferential direction. The ball machine body is arranged on the bottom shell. The side shell surrounds the ball machine body and is inserted into the groove at the bottom. The bottom shell and the side shell jointly form the external protection structure of the ball machine. The design of the bottom shell and the side shell needs to consider the waterproof and dustproof performance (such as IP66 level), heat dissipation performance (high temperature 65℃), low temperature resistance (minus 40℃), etc. The installation of the ball machine usually needs to use a special support. First, the ball machine support is fixed at the required position, and then the ball machine body is installed on the support. For outdoor installation, waterproof measures need to be taken, such as wrapping raw material belts at the cover and the support. In addition, the installation structure of the ball machine also includes fastening components such as clamps, rivet screws and set screws, which are used to ensure the stable connection between the ball machine body and the support. Specifically, the cylindrical protrusion on the ball machine body is inserted into the fastening component, and then the set screw is abutted with the cylindrical protrusion, and the clamp is clamped around the cylindrical protrusion, thereby improving the installation strength and reliability.

[0036] Further, a top cover is further included, which is buckled above the side shell, can effectively prevent rain, dust and the like from entering the interior of the equipment, thereby improving the reliability of the equipment, especially in complex outdoor environments; through close cooperation with the side shell, the top cover can further strengthen the sealing of the entire ball machine shell, ensuring the working stability of the internal electronic components; the design of the top cover can also improve the overall mechanical strength of the ball machine, making it more capable of withstanding external physical impact or vibration. In the embodiment, the bottom shell, the side shell and the top cover adopt a modular clamping slot or buckling structure, which is convenient for installation and disassembly, and provides convenience for later maintenance and debugging.

[0037] Further, the bottom shell is usually designed with a heat dissipation structure, and the top cover takes into account the sealing and ventilation needs to balance the temperature control problem in the operation of the equipment.

[0038] The current environment parameters in step S120 at least include one of illumination, weather condition and visibility. For the illumination parameter, direct measurement is performed by the illumination sensor (such as a photoresistor, a photodiode) arranged inside and outside the machine room, and auxiliary analysis is performed by using the image data collected by the second type of spherical camera, and the illumination information is indirectly obtained by using the image brightness analysis algorithm. For example, the illumination sensor transmits real-time illumination intensity data to the data acquisition module in a wired (such as Modbus, RS485) or wireless (such as Zigbee, Wi-Fi, LoRa) manner, and the data acquisition module uploads the processed data to the backend service (for example, through MQTT, HTTP and the like). The data format of the illumination parameter is: real-time illumination intensity value (unit: Lux), including timestamp, sensor ID, machine room ID and the like.

[0039] The acquisition of weather data is achieved by calling a third-party weather API (such as the China Meteorological Administration Open Platform), while supporting the deployment of small weather stations in specific areas to meet high-precision requirements. A backend service (e.g., a separate microservice or a scheduled task) calls the weather API at regular intervals (e.g., every 5-15 minutes) to obtain the latest weather information for the location of the machine room. The data format is: weather conditions (sunny, cloudy, cloudy, rain, snow, etc.), temperature (°C), humidity (%RH), wind force (level), wind direction, air pressure (hPa), etc.

[0040] The visibility data is directly measured by a professional visibility sensor (such as a transmission visibility meter) and supplemented by the visibility data provided by the weather API. The data format is: visibility value (unit: meters or kilometers), including timestamp, sensor ID, etc.

[0041] It can be understood that the data is stored after step S120, specifically, the collected real-time environmental parameter data is stored in a time series database (such as InfluxDB) for historical query, trend analysis and large data storage; the current latest environmental parameter value is cached to an in-memory database (such as Redis) for quick and low-latency reading during fusion.

[0042] In combination with the first aspect, step S130 includes: S131, data cleaning and standardization processing is performed on the obtained current environmental parameter to obtain a processed current environmental parameter.

[0043] The preprocessing specifically includes: Data cleaning: identify and remove abnormal values or missing values caused by sensor failure or transmission errors; Data standardization / normalization: unify different dimensional environmental parameters (such as light intensity, temperature, visibility) to the same numerical range (e.g., 0-1) for uniform processing by subsequent models or rules.

[0044] Data smoothing processing: smooth the data with large fluctuations to reduce noise influence.

[0045] S132, input the preprocessed environmental parameter into a dynamic weight calculation model to output the initial weights of various types of sensor data.

[0046] The pre-processed environmental parameters are input into a dynamic weight calculation model. In this embodiment, the model supports two calculation methods based on a rule engine and a machine learning model: the rule engine calculates by preset environmental parameter threshold rules (such as increasing the weight of the infrared sensor when the light intensity is <100 Lux); the machine learning model learns the complex nonlinear relationship between environmental parameters and optimal weights by a trained random forest or neural network, and outputs the initial weights of various types of sensor data.

[0047] Method one: based on a rule engine, because environmental parameters can affect data fusion results (such as strong light may cause optical sensor saturation or glare, weak light may cause image recognition accuracy to decrease; bad weather such as rain, snow, and fog can seriously affect the performance of optical sensors (cameras, laser radars), while having less impact on radar sensors; low visibility directly affects the effective detection distance and accuracy of optical sensors), therefore, the mapping relationship between preset environmental parameters and sensor weights is calculated. For example: When the light intensity is < a first threshold value, the weight of the spherical camera is configured to be reduced and the weight of the infrared sensor is configured to be increased; When the weather condition is rain, snow, or fog, the weights of the spherical camera and the laser radar are configured to be reduced and the weights of the millimeter wave radar and the ultrasonic sensor are configured to be increased; When the visibility is < a second threshold value, the weight of the spherical camera is configured to be reduced and the weights of the radar type sensors and the geomagnetic sensor are configured to be increased.

[0048] Method two: based on a machine learning model The environmental parameters are input into a trained weight prediction model, and the model directly outputs the optimal weight vector corresponding to various types of sensor data; the weight prediction model can be one of a neural network, a random forest, or a fuzzy logic system.

[0049] It can be understood that different types of sensors perform differently under different environmental conditions, and environmental noise, sensor accuracy, and environmental dynamic changes can affect the data quality of multi-modal sensors. For example, the image of the camera can be affected by the light condition, while the data of the laser radar can be affected by the rain and snow weather.

[0050] Therefore, the current environmental parameters (such as temperature, humidity, and light intensity) are first obtained, and these environmental parameters provide rich information, which helps to improve the perception and decision-making capabilities of the system, so as to configure weights for each type to match more weights for the type of data that is affected by the current environmental conditions. For example, in the case of low light intensity at night, the infrared sensor data is affected the least, the geomagnetic sensor data is affected the second least, and the ultrasonic sensor data is affected the most, and the corresponding weight configuration is W 红外 =0.5, W 地磁 =0.3, and W 超声=0.2; in the case of low light intensity during the day, the geomagnetic sensor is affected the least, the infrared sensor data and the ultrasonic sensor data are affected similarly and are higher than the geomagnetic sensor data, and the corresponding weight configuration is W 地磁 =0.4, W 红外 =0.3, W 超声 =0.3. It can be understood that the above only gives an example of weight adjustment based on light intensity, and the specific weight configuration can be adjusted according to actual needs or updated based on a pre-constructed deep learning model combined with other environmental parameters (such as temperature, humidity, wind level, noise level, etc.), and here it is only an example and is not limited.

[0051] In order to realize dynamic adaptation, in this embodiment, a plurality of fusion strategies are stored so as to switch the fusion strategy under different environmental conditions and adapt to different application scenarios.

[0052] S133, smoothing the calculated initial weight to suppress weight mutation.

[0053] Smoothing the initial weight output by the dynamic weight calculation model is a key link to ensure the stability of sensor fusion. As an implementable way, an exponential weighted moving average algorithm is used to smooth and correct the initial weight calculated by introducing historical weight data. Specifically, the algorithm assigns different decay factors to the weights at different times, and the recent weight obtains a higher weight, and the historical weight decays exponentially over time. This processing method effectively suppresses the weight mutation caused by the instantaneous fluctuation of environmental parameters, avoids the dramatic jump of the fusion result, and at the same time maintains the response ability of the system to environmental changes. By adjusting the smoothing coefficient, the system can achieve the best balance between response speed and stability.

[0054] S134, normalizing the smoothed weight to obtain the fusion weight of each sensor data; wherein the sum of the fusion weights of each sensor data is 1.

[0055] In step S134, the normalized weight after smoothing is normalized. The normalization process ensures that the sum of the fusion weights of all sensors is strictly equal to 1 by dividing the smoothed weight of each sensor by the sum of all weights. This mathematical processing guarantees the rigor of subsequent weighted average fusion algorithms and prevents fusion result deviation caused by the sum of the weights not being 1. At the same time, the normalization process also makes the weights of different sensors comparable, laying a mathematical foundation for effective fusion of multi-sensor data. After these two steps of fine processing, the system finally obtains a set of sensor fusion weights that reflect the current environmental characteristics and maintain temporal stability.

[0056] In combination with the first aspect, step S140 includes: S141, using the fusion weight, the multi-modal sensor data from the multiple geomagnetic sensors, infrared sensors and ultrasonic sensors are weighted and fused to calculate the comprehensive occupancy confidence of each parking space.

[0057] The dynamic fusion weight calculated by step S130, and the raw data of the geomagnetic sensor (detecting metal objects), infrared sensor (sensing heat sources) and ultrasonic sensor (measuring distance) deployed near each parking space, are used to calculate the comprehensive occupancy confidence of the parking space.

[0058] It can be understood that the comprehensive occupancy confidence is not simply considered as "there is a signal, there is a car", but is intelligently fused. For example: the geomagnetic sensor reading changes significantly (corresponding to weight W1), the infrared sensor detects a biological heat source (corresponding to weight W2), and the ultrasonic sensor distance measurement shows that there is an object in front (corresponding to weight W3), wherein the comprehensive occupancy confidence = (geomagnetic data x W1) + (infrared data x W2) + (ultrasonic data x W3). Thus, a value between 0 and 1 is output, indicating the probability that the parking space is occupied by a vehicle.

[0059] It can be understood that the weights of different sensors in different environments are dynamically adjusted. For example, the weight (W2) of the infrared sensor may be increased at night; the weight (W1) of the sensor less affected by bad weather (such as geomagnetic) will be increased. Through the comprehensive occupancy confidence, the dynamically adjusted weight advantage can be converted into accurate perception capability.

[0060] It can be understood that spatial position calibration is required before actual application, which includes spatial position calibration (at least including spatial position calibration of each ball machine, spatial position calibration of each sensor and relative position calibration), time synchronization calibration and parameter calibration.

[0061] Specifically, in the spatial position calibration process, first, the equipment to be calibrated (such as ball machine or specific sensor) and its functional positioning in the overall system are determined. For example, for an infrared camera used for environmental perception, the relative relationship between it and the ground coordinate system and other sensors needs to be determined.

[0062] Subsequently, a calibration method is selected, for example, for a ball machine calibration process, a calibration board or a scene with known feature points are used to calculate the intrinsic parameters (such as focal length, distortion coefficient) and extrinsic parameters (such as position and attitude relative to the world coordinate system) of the ball machine through geometric transformation for calibration; for the sensor calibration process, the spatial conversion relationship between sensors can be calculated by synchronously collecting multi-sensor data and using commonly observed features (such as fixed reference points or dynamically moving targets); for the relative position calibration process, the position relationship model of the devices in the unified coordinate system can be established by measuring the physical distance or angle between them. Spatial position conversion can also be performed through online calibration or coordinate conversion model. In the time synchronization calibration process, the timestamps of the sensor data can be compared with the reference time source, and time compensation can be performed when the difference is greater than the threshold value, so as to align the timestamps. The above is only an example and is not limited, and can be selected according to actual use habits and needs.

[0063] Through the above steps, accurate judgment of the parking space occupation situation can be realized, and reliable data support can be provided for the intelligent parking lot management system in combination with the license plate recognition technology.

[0064] In S142, candidate target parking spaces with a confidence higher than a threshold value are selected from all parking spaces according to the comprehensive occupation confidence.

[0065] It can be understood that the parking space screening mechanism based on confidence is executed, and the specific implementation process is as follows: 1. Threshold setting mechanism: The system adopts a dynamically configurable confidence threshold, which can be determined in two ways: Static threshold: a preset fixed value (such as 0.65), suitable for stable environment scenarios; Dynamic threshold: adjusted in real time according to environmental parameters, such as appropriately reducing the threshold in rainy and foggy weather to improve system sensitivity.

[0066] 2. Graded screening strategy: First level: screening parking spaces with a confidence greater than 0.8 as high-confidence targets; Second level: screening parking spaces with a confidence of 0.6-0.8 as candidate targets; Third level: parking spaces with a confidence less than 0.6 are temporarily excluded from the candidate range.

[0067] 3. Spatial context verification: In the screening process, the system also considers spatial correlation: The confidence of adjacent parking spaces is analyzed to exclude false positives caused by sensor crosstalk; The rationality of the detection result is verified in combination with the parking space topology.

[0068] 4. Introduce time dimension verification, conduct timing continuity check: Compare historical confidence data to ensure the timing continuity of detection results; Mark the parking space with confidence mutation for special marking for subsequent key verification.

[0069] 5. Output standardization: After screening, the system generates a structured candidate target parking space list, including the following metadata: parking space unique identifier, comprehensive occupancy confidence, sensor contribution analysis, environmental parameter snapshot, and timestamp information.

[0070] Through this multi-dimensional and multi-level screening mechanism, the system can effectively control the false positive rate while ensuring the detection rate, providing high-quality input for the subsequent visual confirmation link, thereby optimizing the efficiency and accuracy of the entire recognition process. This design is particularly suitable for large-scale parking deployment scenarios, achieving the best balance between computing resources and recognition accuracy.

[0071] S143, acquire multi-modal sensor data of the area corresponding to each candidate parking space in the image data.

[0072] S144, fuse multi-modal sensor data to screen target parking spaces with vehicles in each parking space.

[0073] It can be understood that after determining the candidate parking space in step S142, the candidate parking space with high occupancy confidence screened in step S142 is subjected to directional data acquisition. First, according to the unique identifier of each candidate parking space, all sensors deployed in the physical area of the specific parking space are retrieved and called from the registration information of the candidate parking space. Then, the latest readings of these sensors are acquired to form a multi-modal data package for the parking space. This data package usually includes: magnetic field strength data of the geomagnetic sensor, thermal radiation intensity data of the infrared sensor, ranging data of the ultrasonic sensor, and target point cloud data of the radar sensor (if equipped). Subsequently, in step S144, the secondary verification and final confirmation of the candidate parking space are performed. Specifically, the multi-modal data belonging to the same candidate parking space acquired in step S143 is fused at the feature level. Instead of simple weighted averaging, a multi-dimensional feature vector is constructed, such as [magnetic change, thermal radiation intensity, ultrasonic ranging value]. A preset fusion strategy (such as a preset rule or model) is used to determine whether the parking space indeed has a vehicle, thereby screening the parking space with a vehicle and marking it as the final "target parking space".

[0074] S145, trigger the spherical camera unit to rotate and align the target parking space to identify the target vehicle on the target parking space and obtain the acquisition data recognition result corresponding to the target vehicle. The recognition result includes the position data of the target vehicle, the vehicle features, and the vehicle identification.

[0075] Further, the spherical camera unit is controlled to rotate and align with the target parking space to collect images of the target vehicle and perform image recognition, so as to obtain a recognition result.

[0076] In S146, the position data, vehicle features and vehicle identification of the target vehicle are bound with the identification information of the target parking space to construct an association relationship between the target vehicle and the target parking space.

[0077] In combination with the first aspect, before S140, the method further includes: In S1400, it is determined whether the current environment parameter meets the fusion strategy switching condition according to a preset rule or model.

[0078] If yes, S1401-S1402 are executed; if no, S1403 is executed.

[0079] In S1401, a target fusion strategy is selected from a plurality of pre-stored fusion strategies based on the current environment and the running state of the multi-modal sensing unit.

[0080] In S1402, the current fusion strategy is switched to the target fusion strategy in a smooth switching manner.

[0081] In S1403, the current fusion strategy is taken as the target fusion strategy.

[0082] It can be understood that a plurality of fusion strategies are pre-stored in the database, and the switching and adjustment of the fusion strategies can be performed based on actual application scenarios. In each fusion strategy, a preset rule or model is provided, and an execution condition corresponding to the execution of the preset rule or model for data fusion is provided.

[0083] For example, in step S1400, environmental parameters are continuously monitored, and it is determined whether a strategy switch is needed through any of the following methods: First, rule-based judgment: preset environmental parameter threshold conditions, such as "light intensity continuously below 50 Lux for more than 10 seconds" or "visibility below 500 meters and the weather is rainy"; Second, model-based prediction: input the environmental parameters into a pre-trained switching decision model and output a confidence score for strategy switching; Third, system status assessment: combine sensor health status data, and forcefully trigger strategy switching when a key sensor fails. If the fusion strategy switching conditions are met, it means that the current fusion strategy is not applicable, so the strategy is changed, and steps S1401-S1402 are executed. The steps for selecting the target fusion strategy include: first, environment adaptability analysis: evaluating the theoretical performance of each pre-stored strategy under the current environmental parameters; second, resource utilization assessment: considering the computational complexity of each strategy and the current system load; third, historical performance review: querying the historical execution performance records of each strategy under similar environmental conditions; and fourth, strategy priority ranking: outputting a list of candidate strategies ranked by comprehensive score, and selecting the optimal strategy as the target strategy. Then, a smooth switch is performed after determining the appropriate target fusion strategy.

[0084] In conjunction with the first aspect, step S1402 includes: S1402A, within a preset time window, gradually adjusts the mixing ratio of the current fusion strategy and the target fusion strategy until it fully transitions to the target fusion strategy.

[0085] Understandably, within a preset transition time window, a mixing coefficient α is set that linearly changes from 0 to 1. In each processing cycle, the system weights and fuses the output of the current fusion strategy with the output of the target fusion strategy according to the mixing coefficient at the current moment, resulting in the final output for that cycle. Specifically, the final output = (1-μ) × current strategy output + μ × target strategy output. As the processing cycle progresses, the mixing coefficient μ gradually increases from 0 to 1, thus achieving a smooth and disturbance-free transition from the current fusion strategy to the target fusion strategy.

[0086] In conjunction with the first aspect, step S1402 includes: S1402B, during the handover period, the current fusion strategy and the target fusion strategy are run in parallel; S1402C: Compare the output results of the current fusion strategy and the target fusion strategy. If the output of the target fusion strategy is stable and meets expectations, stop running the current fusion strategy and switch completely to the target fusion strategy.

[0087] When the output of the target fusion strategy differs from the output of the current strategy by less than a preset tolerance over N consecutive periods, and its own fluctuation range is below a stability threshold, the target fusion strategy is determined to be stable and in line with expectations. At this point, the system stops running the current fusion strategy and completely switches to the target fusion strategy as the sole output source.

[0088] Combining the first aspect, combining Figure 2 As shown, the method also includes: S210, in response to a route planning request, retrieves the start and end points of the requested route.

[0089] S220 combines a preset map of the closed area with the starting and ending points to output multiple planning data.

[0090] S230: Based on one or more optimization objectives in the request, perform multi-objective optimization on multiple planning data to obtain the target planning path.

[0091] When a route planning request is received from a user or an upper-layer application, the request is first parsed to extract key information, including the route's start and end points. The start and end points can be specific coordinates (such as latitude and longitude, or specific node numbers on an indoor map). Alternatively, if no explicit start point is provided, the current location is used as the default. Furthermore, various input methods are supported, such as voice input, text input, and map clicks.

[0092] Subsequently, based on pre-set map information (usually a topology map with obstacles, boundaries, and passages), and the provided start and end points, path planning algorithms (such as Dijkstra's algorithm, RRT, etc.) are used to search for feasible paths in the map, generating multiple possible path schemes, i.e., planning data. These planning data can include different dimensions such as path length, estimated time, energy consumption, and obstacle avoidance requirements. Furthermore, dynamic environmental factors (such as moving obstacles and real-time traffic conditions) can be considered to adjust the order of the planning data.

[0093] Finally, based on the user's optimization objectives (such as shortest path, lowest energy consumption, least time, maximum safety, etc.), a multi-objective optimization algorithm (such as NSGA-II, MOEA / D, etc.) is used to perform a trade-off analysis, selecting the optimal one or more planning paths from multiple candidate planning data as the final output. A weighting mechanism can be used to allow users to prioritize a specific objective.

[0094] Existing path planning methods typically rely on globally known maps for calculations, making them ill-suited for real-world scenarios involving frequent path changes and unclear road structures within closed areas. The path planning process presented in this application not only achieves effective path finding within closed areas without prior map information but also enhances the intelligence and adaptability of path selection through a multi-objective optimization strategy. This makes it particularly suitable for various practical applications such as navigation systems in complex environments, autonomous driving scenarios, and logistics delivery.

[0095] In conjunction with the first aspect, after step S230, the following also includes: S240: Obtain the current location data and current driving path of the target vehicle.

[0096] S250: Obtain the path overlap rate of the sub-paths between the current driving path and the path starting point to the current location in the target planned road.

[0097] S260, if the path overlap rate is less than the preset threshold, generate and provide a warning message.

[0098] High-precision location data is obtained through vehicle positioning systems (such as GPS, Beidou, UWB, etc.) to acquire the target vehicle's location information (such as GPS coordinates, inertial navigation data, etc.) in real time. Combined with map matching algorithms, the vehicle's actual trajectory is matched with roads or paths in the map to obtain the current driving path.

[0099] Subsequently, the sub-path from the starting point to the current position in the target planned path is extracted, and the sub-path is compared with the vehicle's current actual driving path. The overlap rate between the two is calculated using geometric similarity algorithms or trajectory matching algorithms (such as DTW, LCSS, Hausdorff distance, etc.) to calculate the degree of overlap between the vehicle's current driving path and the original planned path in the already driven segments, in order to determine whether there is a deviation from the planned path.

[0100] When a significant deviation is detected between the vehicle's actual driving path and the planned path (for example, a reasonable path overlap rate threshold is 80%, and anything below this threshold is considered a path deviation), feedback can be provided through voice prompts, visual pop-ups, buzzers, etc., to promptly issue warnings to the user or control system, prevent accidental entry into unexpected areas or improve safety, and can also be linked to the replanning path function to provide new optimal path suggestions.

[0101] If an effective deviation detection mechanism is lacking during vehicle path execution, the vehicle may mistakenly enter unexpected areas, affecting the reliability and safety of the navigation system. This application's embodiment, through monitoring capabilities during path execution, enables the entire path planning system to not only possess intelligent optimization capabilities but also dynamically evaluate and provide feedback on the execution process. This makes it suitable for applications requiring high path execution accuracy, such as autonomous driving, park logistics transportation, and robot navigation.

[0102] Secondly, embodiments of this application also provide a closed-area vehicle management method, applied to a control unit in a closed-area vehicle management system, the system further including a spherical camera unit and a multimodal sensing unit; combined with Figure 3 As shown, the device includes: a first acquisition module 10, a second acquisition module 20, a dynamic calculation module, an association module 40, and a generation module 50.

[0103] The first acquisition module 10 is used to acquire image data collected by the spherical camera unit and multimodal sensing data collected by the multimodal sensing unit.

[0104] The second acquisition module 20 is used to acquire current environmental parameters, which include at least one of the following: illumination, weather conditions, and visibility.

[0105] The dynamic calculation module 30 is used to dynamically calculate the fusion weights of various types of sensor data in the multimodal sensing unit based on the current environmental parameters.

[0106] The association module 40 is used to construct the association relationship between the target vehicle and the target parking space by fusing image data and multimodal sensor data according to the fusion weight.

[0107] The generation module 50 is used to combine the pre-configured rules corresponding to the target parking space and the location data of the target vehicle to generate management data and feed it back to the remote terminal.

[0108] Thirdly, embodiments of this application provide a closed-area vehicle management system, which includes: a spherical camera unit, a multimodal sensing unit, and a control unit.

[0109] The spherical camera unit includes a first type of spherical camera and a second type of spherical camera; the first type of spherical camera is used to acquire images of vehicles entering the closed area; the second type of spherical camera is used to acquire images of parking spaces in the closed area.

[0110] The multimodal sensing unit includes a geomagnetic sensor, an infrared sensor, and an ultrasonic sensor; The control unit is used to receive the collected data sent by the spherical camera unit and the multimodal sensing unit to identify the target vehicle information in the target parking space within the closed area. The spherical camera unit and the multimodal sensing unit are respectively connected to the control unit for communication, and the control unit is used to execute the above method.

[0111] This application achieves all-round perception of vehicles by combining a spherical camera with multi-dimensional sensors, improving the accuracy of data collection. The coupled multi-modal sensing unit can not only identify vehicle information and monitor the vehicle's operating status and parking space usage in real time, but also work with path planning algorithms to realize functions such as automatic guidance and dynamic allocation of parking spaces. When anomalies are detected, early warning prompts are generated in a timely manner.

[0112] In addition, the sensor type and camera distribution can be adjusted according to the needs of different enclosed areas to adapt to diverse management scenarios.

[0113] The geomagnetic sensor is buried in the ground below the parking space (usually about 5-10 cm below ground level) to detect the presence and departure of vehicles, and to determine whether the parking space is occupied by changes in the magnetic field. Preferably, one geomagnetic sensor is buried in the center of each parking space or near the wheel to improve detection accuracy.

[0114] Infrared sensors are typically installed on pillars, light poles, or ceilings above or to the side of parking spaces, generally at a height of 2-3 meters above the ground. They are used to detect whether a vehicle has entered or left the parking space, forming a "light curtain" sensing zone. This helps determine the occupancy status of the parking space, and is especially suitable for nighttime or low-light environments. Preferably, infrared sensors are installed in pairs at the front and rear ends of the parking space, forming a transmission and reception channel. When a vehicle passes by, a signal change is triggered. Alternatively, a single-point infrared proximity sensor can be used to detect if an object is approaching.

[0115] Ultrasonic sensors are typically installed directly above parking spaces (such as on garage ceilings) or on the pillars at the front of the parking space, at a height of approximately 2-2.5 meters above the ground. They are used to accurately measure the distance between the vehicle and the sensor, determining whether the vehicle is fully parked. This helps prevent vehicles from not being fully parked or exceeding the parking space boundaries. Preferably, one ultrasonic probe is installed at the top of each parking space to detect vertically downwards whether a vehicle is parked there. Ultrasonic sensors can also be installed at the entrance to determine whether a vehicle has fully entered or left the designated area.

[0116] In conjunction with the third aspect, the outer shell of the spherical camera includes a bottom shell and a side shell. The top surface of the bottom shell has a groove circumferentially formed, and the bottom of the side shell is inserted into the groove. A top cover is connected above the side shell.

[0117] Fourthly, embodiments of this application provide an electronic device, combined with Figure 4 As shown, the electronic device includes a memory 131 and a processor 130. The memory 131 stores a computer program, and the processor 130 runs the computer program to make the electronic device perform the above-described method.

[0118] Furthermore, combined Figure 4 The electronic device shown also includes a bus 132 and a communication interface 133, with the processor 130, the communication interface 133 and the memory 131 connected via the bus 132.

[0119] The memory 131 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 133 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 132 may be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0120] Processor 130 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 130 or by instructions in software form. Processor 130 may be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 131, and processor 130 reads the information in memory 131 and, in conjunction with its hardware, completes the steps of the method described in the foregoing embodiments.

[0121] Fourthly, embodiments of this application provide a readable storage medium storing computer program instructions, which are read and executed by a processor to perform the above-described method.

[0122] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and apparatus described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0123] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0124] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0125] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0126] Finally, it should be noted that the above embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for vehicle management in a closed area, characterized in that, A control unit applied in a closed-area vehicle management system, the system further comprising a spherical camera unit and a multimodal sensing unit; the method comprising: Acquire image data collected by the spherical camera unit and multimodal sensing data collected by the multimodal sensing unit; Obtain current environmental parameters, which include at least one of illumination, weather conditions, and visibility; Based on the current environmental parameters, the fusion weights of various types of sensor data in the multimodal sensing unit are dynamically calculated; Based on the fusion weights, the image data and multimodal sensor data are fused to construct the association between the target vehicle and the target parking space; By combining the pre-configured rules corresponding to the target parking space and the location data of the target vehicle, management data is generated and fed back to the remote terminal.

2. The method according to claim 1, characterized in that, The step of dynamically calculating the fusion weights of various types of sensor data in the multimodal sensing unit based on the current environmental parameters includes: The acquired current environmental parameters are cleaned and standardized to obtain the processed current environmental parameters; The preprocessed environmental parameters are input into the dynamic weight calculation model, and the initial weights of various sensor data are output. The calculated initial weights are smoothed to suppress abrupt changes in weights; The smoothed weights are normalized to obtain the fusion weights of each of the sensor data; wherein the sum of the fusion weights of each of the sensor data is 1.

3. The method according to claim 2, characterized in that, The step of fusing the image data and multimodal sensing data according to the fusion weights to construct the association between the target vehicle and the target parking space includes: Using the fusion weights, multimodal sensing data from multiple geomagnetic sensors, infrared sensors, and ultrasonic sensors are weighted and fused to calculate the overall occupancy confidence of each parking space. Based on the comprehensive occupancy confidence level, candidate target parking spaces with a confidence level higher than the threshold are selected from all the parking spaces; The spherical camera unit is triggered to rotate and align with the target parking space to identify the target vehicle in the target parking space and the corresponding identification result; the identification result includes the location data of the target vehicle, vehicle features, and vehicle identification. The location data, vehicle characteristics, and vehicle identification of the target vehicle are bound to the identification information of the target parking space to establish the association between the target vehicle and the target parking space.

4. The method according to claim 3, characterized in that, Before the step of fusing the image data and multimodal sensing data according to the fusion weights to construct the association between the target vehicle and the target parking space, the method further includes: Based on preset rules or models, determine whether the current environmental parameters meet the conditions for switching the fusion strategy; If so, based on the current environment and the operating state of the multimodal sensing unit, a target fusion strategy is selected from multiple pre-stored fusion strategies; The current fusion strategy is switched to the target fusion strategy in a smooth switching manner.

5. The method according to claim 4, characterized in that, The steps for smoothly switching from the current fusion strategy to the target fusion strategy include: Within a preset time window, the mixing ratio of the current fusion strategy and the target fusion strategy is gradually adjusted until the target fusion strategy is fully transitioned to.

6. The method according to claim 4, characterized in that, The steps for smoothly switching from the current fusion strategy to the target fusion strategy include: During the handover, the current fusion strategy and the target fusion strategy are run in parallel. Compare the output results of the two strategies. If the output of the target fusion strategy is stable and meets expectations, stop running the current fusion strategy and switch completely to the target fusion strategy.

7. The method according to claim 1, characterized in that, The method further includes: In response to a route planning request, retrieve the start and end points of the route in the request; By combining the preset map of the closed area, the starting point, and the ending point, multiple planning data are output; Based on one or more optimization objectives in the request, multi-objective optimization is performed on the multiple planning data to obtain the target planning path.

8. The method according to claim 7, characterized in that, After the step of performing multi-objective optimization on the multiple planning data based on one or more optimization objectives in the request to obtain the target planning path, the method further includes: Obtain the target vehicle's current location data and current driving path; Obtain the path overlap rate of the sub-paths between the current driving path and the data from the starting point of the path to the current position in the target planned path; If the path overlap rate is less than a preset threshold, an early warning message is generated and fed back.

9. A vehicle management device for a closed area, characterized in that, A control unit applied in a closed-area vehicle management system, the system further including a spherical camera unit and a multimodal sensing unit; the device includes: The first acquisition module is used to acquire image data collected by the spherical camera unit and multimodal sensing data collected by the multimodal sensing unit; The second step is to obtain current environmental parameters, which include at least one of illumination, weather conditions, and visibility. The dynamic calculation module is used to dynamically calculate the fusion weights of various types of sensor data in the multimodal sensing unit based on the current environmental parameters. The association module is used to fuse the image data and multimodal sensing data according to the fusion weight to construct the association relationship between the target vehicle and the target parking space; The generation module is used to combine the pre-configured rules corresponding to the target parking space and the location data of the target vehicle to generate management data and feed it back to the remote terminal.

10. A closed-area vehicle management system, characterized in that, The system includes: The spherical camera unit includes a first type of spherical camera and a second type of spherical camera; the first type of spherical camera is used to acquire images of vehicles entering the enclosed area; the second type of spherical camera is used to acquire images of parking spaces in the enclosed area. The multimodal sensing unit includes a geomagnetic sensor, an infrared sensor, and an ultrasonic sensor; The control unit is used to receive the collected data sent by the spherical camera unit and the multimodal sensing unit to identify the target vehicle information in the target parking space within the closed area; The spherical camera unit and the multimodal sensing unit are respectively communicatively connected to the control unit, which is used to execute the method as described in any one of claims 1-8.

Images