A parking monitoring system supporting time-lapse recording and event triggering

Abstract

The application relates to the technical field of vehicle-mounted electronic equipment and security monitoring, and discloses a parking monitoring system supporting time-lapse recording and event triggering, which comprises space sensing nodes and opto-mechanical-electrical execution nodes in communication with each other; a microwave radar module in the space sensing node acquires the relative distance and azimuth angle of a physical target, and a BLE beacon module acquires the polar coordinates of a mobile terminal; a main control chip in the opto-mechanical-electrical execution node maps and aligns the polar coordinates and outputs a cross-authentication state, controls a video control module to switch between time-lapse recording and regular recording states according to the authentication state, and drives a zoom camera to change the field angle and drives the lens of a light projection lamp module to change the light beam state. The authorized state is determined by calculating the spatial Euclidean distance difference to avoid frequent false wake-up, and the target position is oriented and the light is supplemented in synchronization, so that the definition of the night monitoring evidence picture is improved while the system power consumption and storage consumption are reduced.

Description

Technical Field

[0001] This invention relates to the field of vehicle electronic equipment and security monitoring technology, specifically a parking monitoring system that supports time-lapse recording and event triggering. Background Technology

[0002] Vehicle parking monitoring systems are primarily used to record abnormal events in the vicinity of a vehicle while it is parked with the engine off, in order to ensure vehicle safety. Conventional parking monitoring systems mostly rely on basic image motion detection or single-point radar for activation. In complex environments such as public parking lots, the normal movement of pedestrians, stray animals, or surrounding vehicles can easily cause frequent false triggers to the system, resulting in the monitoring equipment being in a high-power recording state for extended periods. This not only rapidly depletes the vehicle's battery power but also occupies a significant amount of storage space, making it difficult to achieve effective monitoring over long periods.

[0003] To reduce false trigger rates, some monitoring devices have incorporated Bluetooth signals from mobile terminals as an aid in identity verification. However, underground parking garages contain numerous metal load-bearing columns and densely parked vehicles, causing multipath reflections of radio waves during propagation. Traditional signal processing methods struggle to remove interference spikes caused by these static metal reflectors, resulting in spatial coordinate deviations in terminal positioning and preventing the system from accurately performing cross-authentication between the target and the vehicle owner.

[0004] Furthermore, lighting conditions in underground parking lots or outdoor environments at night are typically poor. Existing surveillance equipment mostly uses fixed-focus lenses and supplementary lights with fixed illumination angles, which cannot dynamically adjust the optics according to the actual distance and orientation of the physical target. When unauthorized targets are active at a distance or in peripheral areas, a fixed field of view cannot capture clear local features, and fixed supplementary lighting parameters can easily result in insufficient overall brightness or local overexposure, ultimately leading to blurry surveillance footage that cannot provide effective visual evidence for subsequent tracing. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a parking monitoring system that supports time-lapse recording and event triggering. This system solves the problems of excessive power consumption and storage consumption caused by frequent false triggering by irrelevant moving targets in existing parking monitoring systems, as well as low positioning accuracy due to metal multipath reflection in complex parking environments and unclear shooting of distant targets in low light conditions.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a parking monitoring system that supports time-lapse recording and event triggering, comprising a spatial sensing node and an opto-electro-mechanical execution node that communicate with each other.

[0007] The microwave radar module within the spatial perception node acquires the instantaneous relative distance and azimuth of the physical target, while the BLE beacon module acquires the terminal's relative polar coordinates. The main control chip within the opto-mechatronics execution node maps the instantaneous relative distance, azimuth, and terminal relative polar coordinates to a local two-dimensional polar coordinate system and outputs the cross-authentication status. Based on the cross-authentication status, it controls the video control module to switch between time-lapse recording and encoding states, drives the zoom camera to change its optical field of view, and drives the stepper motor within the floodlight module to move the lens relative to the LED beads, thus changing the beam state.

[0008] The spatial sensing node employs a multi-layer printed circuit board stacked structure. The high-frequency microwave antenna layer of the microwave radar module and the radio frequency antenna array layer of the BLE beacon module maintain a predetermined vertical spacing in physical space, with a copper-clad grounding shield layer laid between the two antenna boards. The floodlight module also includes a lead screw drive mechanism and a linear guide rail. The lens is slidably mounted on the linear guide rail, and the output shaft of the stepper motor is mechanically connected to the lead screw drive mechanism.

[0009] When the main control chip receives a signal that the vehicle engine is off and the main power supply is disconnected, it cuts off the backup power supply to the motor servo drive circuit and the floodlight module inside the zoom camera. The image sensor of the zoom camera retains a low-power basic power supply. The video control module controls the zoom camera to acquire images and perform time-lapse recording encoding. When the microwave radar module detects a physical target exceeding the preset distance threshold, it sends a wake-up interrupt signal to the opto-electro-mechanical execution node, and the main control chip restores the backup power supply to the zoom camera and the floodlight module.

[0010] The BLE beacon module acquires phase difference data of signals received from multiple antennas and generates a spatial spectrum peak sequence. The BLE beacon module then uses a preset static metallic reflection spatial map to perform a mask comparison on the spatial spectrum peak sequence. The static metallic reflection spatial map is a set of values ​​containing multiple forbidden azimuth intervals. The BLE beacon module compares the azimuth values ​​of each spatial spectrum peak with the forbidden azimuth intervals, removes static reflection spurious peaks falling within the forbidden azimuth intervals, extracts the peak with the highest energy amplitude from the remaining sequence as the effective direct line-of-sight radial peak, and uses the polar coordinate angle corresponding to the effective direct line-of-sight radial peak as the terminal's relative azimuth angle. Combined with the received signal strength indication attenuation data, the terminal's relative polar coordinates are calculated.

[0011] The spatial sensing node encapsulates the instantaneous relative distance and azimuth of the physical target, the relative polar coordinates of the terminal, and the system timestamp into a single spatial vector data packet matrix. Upon acquiring the instantaneous relative distance and azimuth, the spatial sensing node transmits the spatial vector data packet matrix to the opto-mechatronic execution node via the vehicle-mounted local area communication bus.

[0012] The main control chip continuously acquires multiple frames of mapped and aligned polar coordinate data within a set time window. It calculates the spatial Euclidean distance difference between the instantaneous relative distance and azimuth angle of the physical target and the relative polar coordinates of the terminal within the same time slice as a spatial deviation factor. The main control chip sums the sequence spatial deviation factors generated within the time window and divides this sum by the total number of sampled frames to obtain the average value of the sequence spatial deviation factors. The main control chip compares the average value with a preset tolerance threshold. If the average value is not greater than the preset tolerance threshold, it outputs a cross-authentication status indicating an authorized state; if the average value is greater than the preset tolerance threshold, it outputs a cross-authentication status indicating an unauthorized state.

[0013] When the cross-authentication status is authorized, the main control chip maintains the time-lapse recording encoding status of the video control module. The main control chip outputs a pulse width modulation signal to drive the stepper motor to move the lens closer to the LED, and the floodlight module switches to a wide field-of-view floodlight state.

[0014] When the cross-authentication status is unauthorized, the main control chip controls the video control module to switch to high frame rate regular recording and encoding mode. The main control chip extracts the instantaneous relative distance and azimuth angle as servo tracking parameters, driving the stepper motor to move the lens away from the LED, and the floodlight module switches to a small field-of-view focusing mode. The video control module extracts a photosensitive area mask based on the instantaneous relative distance and azimuth angle and feeds it back to the internal image signal processor. The image signal processor independently calculates the exposure compensation parameters based on the photosensitive area mask and sends them to the zoom camera. The zoom camera calculates the target focal length based on the instantaneous relative distance, maps the target focal length to the target running steps of the internal micro servo motor, and drives the internal zoom optical lens group to translate.

[0015] The zoom camera retrieves the photosensitive area mask captured by the video control module and sets the two-dimensional pixel space covered by the photosensitive area mask as the region of interest (ROI). The zoom camera controls the internal focusing motor to fine-tune the position of the internal focusing optical lens group, and simultaneously calculates the gradient values ​​of the high-frequency components of the image within the ROI to perform autofocus.

[0016] The floodlight module calculates the target illumination beam angle based on the instantaneous relative distance. The microcontroller inside the module consults a lens translation calibration table to convert the target illumination beam angle into a target movement step count, driving a stepper motor to mechanically translate the lens along the light source's emission direction. The floodlight module also calculates the illumination drive pulse width modulation duty cycle based on the instantaneous relative distance, writing the duty cycle value into the comparison register of an internal hardware timer to generate the corresponding pulse width modulation level signal, which controls the on-time ratio of the switching transistor in the internal constant current drive circuit.

[0017] This invention provides a parking monitoring system that supports time-lapse recording and event triggering. It has the following advantages: 1. This invention establishes a cross-authentication mechanism by calculating the spatial Euclidean distance difference between the physical target coordinates obtained by the microwave radar module and the terminal coordinates obtained by the BLE beacon module. The system compares the average value of the calculated spatial deviation factor with a preset tolerance threshold to determine whether the approaching target is a vehicle owner carrying an authorized terminal. When the system is determined to be in an authorized state, it maintains low-power time-lapse recording; when the system is determined to be in an unauthorized state, it switches to high-frame-rate regular recording. This design avoids the invalid and frequent wake-ups caused by stray animals or irrelevant personnel passing by in traditional monitoring systems, and reduces the system's power consumption and storage consumption while ensuring the effectiveness of monitoring.

[0018] 2. When an unauthorized target approaches, the main control chip extracts the relative distance and azimuth angle obtained by the radar as servo tracking parameters, and drives the lens group of the zoom camera to translate and the floodlight module to switch to the focusing state. In conjunction with the video control module, the system extracts the photosensitive area mask according to the target position and independently calculates the exposure compensation parameters. The system can perform directional lighting and precise focusing for targets in specific spatial positions, which solves the problems of blurry shooting and difficulty in identifying features of distant targets at night or in complex lighting conditions, and improves the clarity of surveillance and evidence collection images.

[0019] 3. The BLE beacon module of this invention is equipped with a static metal reflection spatial spectrum. By comparing the azimuth of the spatial spectrum peak generated by the multi-antenna received signals with a preset forbidden azimuth interval, static reflection pseudo-peaks falling within the interval are eliminated. This spatial masking processing method can effectively filter out multipath signal interference generated by other metal vehicles or building columns in environments such as underground parking garages, and extract the true direct line-of-sight radial peak, thereby improving the polar coordinate positioning accuracy of the mobile terminal in complex parking environments. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the overall structure of the parking monitoring system according to an embodiment of the present invention; Figure 2 This is the main flowchart of the control method supporting time-lapse recording and event triggering according to an embodiment of the present invention; Figure 3 This is a comparison data diagram of field-of-view stability during target approach according to an embodiment of the present invention; Figure 4 This is a comparison chart of floodlight illumination driving power and overexposure prevention data in an embodiment of the present invention.

[0021] Among them, 10 is the spatial perception node; 101 is the microwave radar module; 102 is the BLE beacon module; 20 is the opto-electro-mechanical execution node; 201 is the main control chip; 202 is the video control module; 203 is the zoom camera; 204 is the floodlight module; 2041 is the stepper motor; 2042 is the lens; and 2043 is the LED. Detailed Implementation

[0022] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] See attached document Figure 1 The present invention provides a parking monitoring system that supports time-lapse recording and event triggering, including: a spatial sensing node 10 and an opto-electro-mechanical execution node 20.

[0024] The spatial sensing node 10 and the opto-electro-mechanical execution node 20 establish a data connection via an onboard local area communication bus. The spatial sensing node 10 is deployed in a non-metallic shielded area inside the vehicle. The spatial sensing node 10 includes a microwave radar module 101 and a BLE beacon module 102. The microwave radar module 101 is used to acquire and output the instantaneous relative distance and azimuth of physical targets in the onboard local area communication bus. The BLE beacon module 102 contains multiple radio frequency antennas for receiving radio frequency broadcast signals from mobile terminals and outputting phase difference data.

[0025] The opto-electro-mechanical execution node 20 is deployed inside the windshield of the vehicle. The opto-electro-mechanical execution node 20 includes a main control chip 201, a video control module 202, a zoom camera 203, and a floodlight module 204. The main control chip 201 establishes electrical connections with the video control module 202, the zoom camera 203, and the floodlight module 204, respectively.

[0026] The floodlight module 204 internally includes a stepper motor 2041, a lens 2042, an LED chip 2043, a lead screw drive mechanism, and a linear guide rail. The lens 2042 is slidably mounted on the linear guide rail, and the output shaft of the stepper motor 2041 is mechanically connected to the lead screw drive mechanism. The stepper motor 2041 receives instructions from the main control chip 201 and drives the lens 2042 to linearly translate along the optical axis of the LED chip 2043 on the linear guide rail via the lead screw drive mechanism, thereby changing the light emission angle of the floodlight module 204.

[0027] See attached document Figure 2 This invention provides a control method for a parking monitoring system that supports time-lapse recording and event triggering, comprising the following steps: S10, when the main control chip 201 receives the signal that the vehicle engine is off and the main power is disconnected, it activates the hierarchical power gating strategy. The opto-electro-mechanical execution node 20 cuts off the motor servo drive circuit inside the zoom camera 203 and the backup power supply of the floodlight module 204. The image sensor of the zoom camera 203 retains the low-power basic power supply. The video control module 202 drives the zoom camera 203 to acquire images and perform time-lapse video encoding at a preset low frame rate. The microwave radar module 101 enters the polling scanning state, and the BLE beacon module 102 enters the radio frequency broadcasting state. S20, after the mobile terminal enters the communication coverage area of ​​the BLE beacon module 102, the BLE beacon module 102 acquires the phase difference data of the multi-antenna received signals and generates spatial spectrum peaks. The BLE beacon module 102 calls the preset static metal reflection spatial map to perform mask comparison on the spatial spectrum peaks, filters out static reflection pseudo-peaks and extracts the effective direct line-of-sight radial peaks. The BLE beacon module 102 calculates the terminal's relative polar coordinates in combination with the received signal strength indication attenuation data and sends the terminal's relative polar coordinates to the mobile terminal. S30, when the microwave radar module 101 detects that the physical target has exceeded the preset distance threshold, it sends a wake-up interrupt signal to the opto-electro-mechanical execution node 20, and the main control chip 201 restores the backup power supply to the zoom camera 203 and the floodlight module 204. S40, the main control chip 201 synchronously acquires the physical target polar coordinates output by the microwave radar module 101 and the terminal relative polar coordinates output by the BLE beacon module 102, and the main control chip 201 maps the physical target polar coordinates and the terminal relative polar coordinates to the local two-dimensional polar coordinate system. S50, the main control chip 201 continuously collects the mapped polar coordinate data within the set time window, calculates the spatial Euclidean distance difference between the physical target polar coordinates and the terminal relative polar coordinates as the spatial deviation factor, calculates the average value of the sequence spatial deviation factor, and compares the average value with the preset tolerance threshold to output the cross-authentication status. S60, when the cross-authentication state is authorized, the main control chip 201 maintains the time-lapse recording encoding state of the video control module 202, and the main control chip 201 outputs a pulse width modulation signal to drive the stepper motor 2041 of the floodlight module 204 to move the lens 2042 towards the direction closer to the lamp bead 2043, so that the floodlight module 204 is in a large field of view floodlight state. S70, when the cross-authentication state is unauthorized, the main control chip 201 controls the video control module 202 to switch to the high frame rate regular video recording and encoding state. The main control chip 201 extracts the polar coordinates of the physical target as servo tracking parameters, drives the stepper motor 2041 of the floodlight module 204 to move the lens 2042 away from the lamp bead 2043 (to the vicinity of the focal point), so that the floodlight module 204 is in a small field of view focusing state. The main control chip 201 drives the zoom camera 203 to change the optical field of view and lock it to the physical space area corresponding to the polar coordinates of the physical target.

[0028] See attached document Figure 1 It includes: a spatial sensing node 10 and an opto-mechatronic execution node 20. The spatial sensing node 10 and the opto-mechatronic execution node 20 establish a data connection through an on-board local area communication bus.

[0029] To overcome the physical shielding and attenuation problem of radio frequency signals caused by the vehicle's metal frame, the spatial sensing node 10 is deployed in a non-metallic shielded area inside the vehicle. This non-metallic shielded area, in the specific physical structure of the vehicle, refers to spaces below the plastic trim panel of the center console and inside the plastic base of the rearview mirrors—spaces not obstructed by the vehicle's metal panels. The system adopts a split installation structure, physically separating the spatial sensing node 10 from the opto-electro-mechanical actuation node 20 conventionally installed on the windshield. This allows the radio frequency antenna inside the spatial sensing node 10 to acquire complete external spatial signals, avoiding multipath reflection and attenuation when the signal penetrates the vehicle's metal body panels.

[0030] In terms of hardware layout, the spatial sensing node 10 adopts a multi-layer printed circuit board stack-up structure. The spatial sensing node 10 includes a microwave radar module 101 and a BLE beacon module 102. The high-frequency microwave antenna layer of the microwave radar module 101 and the radio frequency antenna array layer of the BLE beacon module 102 maintain a predetermined vertical distance in physical space. The specific value of this vertical distance is determined based on one-quarter to one-half of the wavelength corresponding to the operating frequency of the microwave radar module 101. By laying a copper-clad ground shielding layer between the two antenna boards, near-field crosstalk between electromagnetic waves of different frequency bands is avoided, ensuring that the microwave radar module 101 and the BLE beacon module 102 can operate synchronously and stably.

[0031] Regarding the interconnection hardware architecture between nodes, the on-board local communication bus is specifically manifested as a controller area network bus or a local interconnection network bus. The spatial perception node 10 and the opto-electro-mechanical execution node 20 are physically connected via twisted-pair cables. The twisted-pair cables simultaneously carry the transmission of differential data signals and the distribution of DC power between nodes. For the physical layer circuit design of the controller area network bus transceiver and the selection and arrangement of the twisted-pair cables, those skilled in the art can configure them according to conventional vehicle electrical specifications; the underlying hardware topology is well-known in the field and will not be elaborated upon here.

[0032] The process of establishing a data connection and cooperating between the spatial sensing node 10 and the opto-mechatronics execution node 20 specifically includes the following steps: S101, when the vehicle is powered on and started, the opto-electro-mechanical execution node 20 outputs the basic power supply voltage and bus wake-up level to the space perception node 10 through the vehicle local area communication bus. The power management circuit inside the space perception node 10 performs step-down and voltage regulation on the basic power supply voltage and provides a stable operating voltage for the microwave radar module 101 and the BLE beacon module 102. S102, after the spatial perception node 10 initializes its operating parameters, it sends a synchronization confirmation message containing the device hardware serial number to the opto-electro-mechanical execution node 20 through the vehicle local area communication bus. The opto-electro-mechanical execution node 20 receives the synchronization confirmation message and completes the communication addressing and binding between the nodes. At the same time, the opto-electro-mechanical execution node 20 issues a global clock synchronization command to the spatial perception node 10, so that the local timers of the two nodes are aligned to the same time base. S103, after the system enters the parking monitoring state, when the spatial perception node 10 obtains the valid spatial coordinates of the external target, it encapsulates the coordinate data and the system timestamp into a data frame of a specific format, and continuously sends the data frame to the opto-electro-mechanical execution node 20 through the vehicle local area communication bus.

[0033] The data frames sent by the spatial sensing node 10 carry spatial vector data packets. The main control chip 201 processes the data based on the values ​​in these spatial vector data packets. The data structure of these spatial vector data packets is represented by the following matrix formula: ; In the above formula, Represents a spatial vector data packet matrix for a single transmission; This represents the system timestamp of the currently acquired data; This indicates the instantaneous relative distance of the physical target acquired by the microwave radar module 101; This indicates the azimuth angle of the physical target acquired by the microwave radar module 101; This indicates the relative distance to the mobile terminal obtained by the BLE beacon module 102; This represents the azimuth angle of the mobile terminal acquired by the BLE beacon module 102. Among these, the distance variable... and Typically measured in meters, its effective range is determined by the maximum physical detection limit of the microwave radar and Bluetooth chip used; azimuth variable and The unit is degrees, and the value ranges from 0 to 360 degrees.

[0034] Due to differences in sampling frequencies and hardware resolution time between microwave radar and Bluetooth radio frequency, direct data reading will result in spatial misalignment. The system will combine the two sets of coordinates with the system timestamp aligned through step S102. By forcibly encapsulating them within the same matrix, it can be ensured that the physical target and the mobile terminal being compared are within the same time slice when the main control chip extracts data. When the system is running but the BLE beacon module 102 does not receive any mobile terminal signal, the variables in the matrix... and Preset null identifiers are used for placeholder filling. After receiving the spatial vector data packet matrix, the opto-electro-mechanical execution node 20 performs subsequent coordinate mapping and cross-authentication control based on the data in each dimension of the matrix.

[0035] In the spatial sensing node 10 provided by this invention, the microwave radar module 101 serves as the underlying triggering hardware for physical space target approximation, enabling low-power environmental sensing and initial spatial coordinate acquisition. To meet the requirements for long-term operation in parking monitoring scenarios where vehicle battery power is limited, the microwave radar module 101 employs duty cycle-based low-power operating logic and a bus network wake-up mechanism.

[0036] The underlying radio frequency architecture of the microwave radar module 101 adopts a frequency-modulated continuous wave (FM-CW) system. The microwave radar module 101 integrates one set of transmitting antennas and multiple sets of receiving antennas. It transmits electromagnetic waves whose frequency changes linearly with time and receives the echoes reflected from the target, calculating the difference frequency signal to obtain the relative spatial information of the physical target. For the specific circuit wiring of the voltage-controlled oscillator design, RF front-end mixer circuit, and underlying fast Fourier transform hardware accelerator of the FM-CW radar, those skilled in the art can select and configure them according to the hardware specifications of conventional automotive millimeter-wave radars. Its underlying physical layer transceiver principle is well-known technology in the field and will not be elaborated here.

[0037] The specific operation and processing of the microwave radar module 101 in the parking monitoring system includes the following steps: S201, after the system enters the graded power gating state, the power management chip inside the microwave radar module 101 controls the RF transceiver link to enter a millisecond-level sleep and wake-up alternating polling scanning mode. Within a complete duty cycle, the microwave radar module 101 maintains a set proportion of RF shutdown time to reduce static power consumption, and drives the transmitting antenna to radiate a set of linear frequency modulated pulse signals to the external physical space within the wake-up time window.

[0038] S202, the receiving antenna of the microwave radar module 101 captures the echo signal reflected by the physical target. The radio frequency front-end circuit mixes the received echo signal with the currently transmitted local oscillator signal and outputs a difference frequency signal. The digital signal processor inside the microwave radar module 101 performs sampling and frequency domain transformation calculations on the difference frequency signal and extracts the peak frequency in the spectrum whose amplitude exceeds a preset noise floor threshold.

[0039] The physical basis for ranging and angle measurement of microwave radar module 101 is that the transmitted frequency-modulated pulse signal is delayed after being reflected by an external physical target. In the frequency-modulated continuous wave system, this time delay is manifested as the frequency difference between the received echo and the currently transmitted local oscillator signal, i.e., the difference frequency. At the same time, since different receiving antennas are arranged in physical space, the flight path length of the same echo to different receiving antennas is different, thus generating a phase difference between different receiving channels.

[0040] S203, the digital signal processor analyzes the instantaneous relative distance of the physical target based on the extracted peak frequency, and analyzes the azimuth angle of the physical target based on the phase difference between different receiving antennas at the same moment. The core calculation logic called by the microwave radar module 101 to analyze the instantaneous relative distance and azimuth angle is expressed by the following formula: ; ; In the above formula, This indicates the instantaneous relative distance of the physical target acquired by the microwave radar module 101; The speed constant of electromagnetic waves in air is typically taken as 3 × 10⁻⁶. 8 meters per second; This represents the peak difference frequency extracted after performing a fast Fourier transform on the difference frequency signal; This represents the frequency modulation slope of the transmitted linear frequency modulated pulse signal; This indicates the azimuth angle of the physical target acquired by the microwave radar module 101; Indicates the center wavelength of the transmitted microwave signal; This represents the phase difference data generated when two adjacent receiving antennas capture the same echo signal, and its value ranges from [value range missing]. to Between arcs; This represents the physical distance between two adjacent receiving antennas. The frequency modulation slope is also included. Center wavelength and the physical spacing between antennas These are all hardware constant parameters that are factory-set for the microwave radar module 101.

[0041] S204, the microwave radar module 101 calculates the instantaneous relative distance. The value is compared with a preset distance threshold. The specific value of this preset distance threshold is set based on the physical boundary size of the standard parking space occupied by the parked vehicle, ranging from 1.5 meters to 3.0 meters. When the instantaneous relative distance... When the distance is greater than or equal to the preset distance threshold, the microwave radar module 101 determines that the target is outside the safe distance, actively clears the coordinate data in the current memory register, and returns to step S201 to continue the polling scan that alternates between sleep and wake-up.

[0042] S205, when the instantaneous relative distance When the distance is less than the preset distance threshold, the microwave radar module 101 determines that a physical target is approaching the vehicle. The microwave radar module 101 then sends a wake-up message containing a specific network wake-up identifier to the opto-electro-mechanical execution node 20 via the vehicle-mounted local area communication bus. This message triggers the opto-electro-mechanical execution node 20 to restore power to the corresponding electromechanical module. The microwave radar module 101 then calculates the instantaneous relative distance... With azimuth The data is encapsulated and output to the main control chip 201 according to the preset spatial vector data frame format to support subsequent coordinate mapping alignment and heterogeneous data cross-authentication calculation.

[0043] In the spatial perception node 10 provided by the present invention, the BLE beacon module 102 serves as radio frequency hardware for mobile terminal identification and relative coordinate acquisition, and is used to achieve blind-spot-free vehicle location and data acquisition in complex physical environments.

[0044] The BLE beacon module 102 includes a Bluetooth microcontroller and multiple sets of RF antennas arranged at predetermined physical intervals. For the design of the Bluetooth constant audio spread signal transmission and reception protocol and the in-phase quadrature sampling circuit, those skilled in the art can configure it according to the standard Bluetooth direction finding core specifications. The principles of RF sampling and baseband demodulation are well-known technologies in the field and will not be elaborated upon here.

[0045] To overcome multipath reflection interference to radio frequency signals caused by the vehicle's metal frame, the BLE beacon module 102 introduces a preset mask matching mechanism. The specific processing steps include: S301, when the mobile terminal enters the communication coverage area of ​​the BLE beacon module 102, the BLE beacon module 102 receives the radio frequency broadcast signal sent by the mobile terminal. The baseband processing unit inside the BLE beacon module 102 performs in-phase quadrature sampling on the same broadcast signal received by multiple radio frequency antennas and extracts the phase difference data of the multi-channel signals. Because the propagation path length of the radio waves transmitted by the mobile terminal to each radio frequency antenna is different, the baseband digital signal exhibits a quantifiable phase delay between each receiving channel.

[0046] S302, the BLE beacon module 102 calls a multi-signal classification algorithm to perform covariance matrix eigenvalue decomposition on the extracted phase difference data, generating a spatial spectrum peak sequence containing multiple signal arrival directions. Because the vehicle interior contains numerous fixed metal components such as side pillars and roof sheet metal parts, radio frequency signals, after being reflected by these metal components, will form multiple propagation paths to reach the receiving antenna. This results in multiple energy-concentrated directional peaks appearing in the spatial spectrum generated by the algorithm, making it impossible to directly distinguish between the original direct signal and the reflected interference signal.

[0047] S303, the BLE beacon module 102 calls the preset static metal reflection spatial map in the memory to perform mask comparison on the generated spatial spectrum peak sequence. The static metal reflection spatial map is a set of mask data generated during the equipment manufacturing stage based on the specific body frame structure of the target vehicle. The data structure of this map is a set of values ​​containing multiple forbidden azimuth angle intervals, each forbidden azimuth angle interval corresponding to the physical blocking and reflection range of a fixed metal reflection source on the vehicle. The BLE beacon module 102 compares the azimuth angle values ​​of each spatial spectrum peak obtained at the moment with the above-mentioned forbidden azimuth angle intervals. If the azimuth angle value of a certain set of spatial spectrum peaks falls within any forbidden azimuth angle interval, the BLE beacon module 102 determines it as a static reflection pseudo-peak caused by the metal reflection of the vehicle body and removes it from the sequence. After the removal and filtering, the remaining sequence represents the dynamic reflection path or direct path that is not blocked by fixed metal. Under the constraints of electromagnetic wave propagation characteristics, the unobstructed direct path has the shortest physical propagation distance and the least energy attenuation. The BLE beacon module 102 extracts the peak with the highest energy amplitude in the remaining sequence as the effective direct line-of-sight radial peak, and uses the polar coordinate angle corresponding to the peak as the terminal relative azimuth angle.

[0048] S304, after extracting the relative azimuth angle of the terminal, the BLE beacon module 102 retrieves the received signal strength indication data for that channel. The physical energy of a radio signal attenuates with increasing spatial distance as it propagates through the air. The BLE beacon module 102 calculates the spatial distance of the mobile terminal based on the logarithmic distance path loss model. The specific principle behind the distance calculation by the BLE beacon module 102 is expressed by the following formula: ; In the above formula, This indicates the relative distance to the mobile terminal obtained by the BLE beacon module 102, in meters; This indicates the actual received signal strength when the BLE beacon module 102 receives a valid direct line-of-sight radial peak signal, and its valid value range is between -100 and -30 dBmW. It represents the standard reference received signal strength at a physical distance of 1 meter from the radio frequency antenna. It is a constant parameter calibrated by testing in an anechoic chamber environment, and its value is usually between -50 and -70 dBmW. This represents the environmental path loss index, which is set based on the signal attenuation complexity of parking physical environments such as underground garages, and its value ranges from 2 to 4.

[0049] S305, BLE beacon module 102 will calculate the relative distance The relative azimuth angle of the terminal is combined with the terminal's relative polar coordinates. The BLE beacon module 102 uses the established local radio frequency connection link to send these relative polar coordinates to the mobile terminal, enabling the mobile terminal to update its location and provide vehicle-finding route guidance in environments without satellite positioning signals. Simultaneously, the BLE beacon module 102 sends these relative polar coordinates to the main control chip 201 via the vehicle's local communication bus for subsequent data cross-comparison.

[0050] The opto-electro-mechanical execution node 20 provided by the present invention serves as the core computing and field-of-view execution unit of the parking monitoring system, and its overall structure is separated from the spatial perception node 10.

[0051] The opto-electro-mechanical actuator node 20 is deployed on the inner side of the vehicle's windshield. In specific installation, the housing of the opto-electro-mechanical actuator node 20 is attached to the center of the upper part of the windshield. This installation position ensures that the optical window at the front of the opto-electro-mechanical actuator node 20 faces the external space in front of the vehicle, ensuring that the viewing optical path of the zoom camera 203 and the illumination optical path of the floodlight module 204 are not obstructed by the vehicle's own structure.

[0052] The opto-electro-mechanical execution node 20 integrates a main control chip 201 and electronic components driving a high-power light source, which will generate heat accumulation during long-term parking monitoring. To ensure system stability, the opto-electro-mechanical execution node 20 employs a passive heat dissipation design in its housing structure. The back of the housing of the opto-electro-mechanical execution node 20 is integrally molded from a high thermal conductivity metal material, and multiple sets of parallel heat dissipation fins are arranged on the surface of the back of the housing. The surface of the heat-generating motherboard area inside the housing is covered with a thermally conductive silicone pad. The heat generated by the chip operation is conducted to the metal housing through the thermally conductive silicone pad, and finally exchanged with the air inside the vehicle through the heat dissipation fins. For the die-casting process of the automotive electronic device housing and the specific selection of thermally conductive materials, those skilled in the art can configure them according to the temperature control standards of conventional automotive electronic devices. Its passive heat dissipation structure is well-known technology in the field and will not be described in detail here.

[0053] The opto-electro-mechanical execution node 20 establishes a power supply and data transmission topology among its internal hardware modules. Its underlying interconnection logic specifically includes the following steps: In S401, at the power supply topology level, the opto-electromechanical execution node 20 has an independent power management circuit board. This power management circuit board is connected to the DC constant power input line of the vehicle battery and outputs different levels of operating voltage through a step-down and voltage regulation conversion network. The power management circuit board provides the resident base voltage for the main control chip 201, the video control module 202, and the vehicle local area bus transceiver. Simultaneously, the power management circuit board connects a field-effect transistor (FET) in series as an electronic switch in the power supply circuit of the zoom camera 203 and the floodlight module 204. The general-purpose input / output pins of the main control chip 201 are directly connected to the control gate of this FET. The main control chip 201 controls the conduction and cutoff of the FET by outputting high and low levels, thereby realizing the electromechanical pre-power supply cutoff function in the hierarchical power gating strategy.

[0054] S402, regarding the system's network wake-up mechanism, the receiving pin of the vehicle local area bus transceiver continuously monitors the differential level changes on the bus. When a network wake-up message sent by the spatial perception node 10 appears on the vehicle local area communication bus, the vehicle local area bus transceiver parses the message and outputs a level toggle signal to the wake-up interrupt port of the main control chip 201 through a dedicated suppression pin. After receiving this signal, the main control chip 201 switches from low-power sleep mode to full-speed operation, thereby driving the field-effect transistor on the power supply circuit to conduct and restoring the operating power supply to the electromechanical module.

[0055] In the S403 data transmission topology, the main control chip 201 and the video control module 202 establish a data interaction link via a high-speed serial device interconnect bus to meet the high-bandwidth transmission requirements of video encoding compression instructions and spatial coordinate data. The image sensor array inside the zoom camera 203 is connected to the video control module 202 via a mobile industrial processor interface bus. The raw environmental photoelectric signals acquired by the image sensor array are converted from analog to digital and then continuously transmitted as a raw data stream to the video control module 202 via this interface bus for low-level frame buffering and digital image signal processing.

[0056] In the S404 electromechanical control topology, the main control chip 201 configures the hardware timer peripherals according to the internal program's operating logic, generating multiple pulse width modulation (PWM) signals with adjustable duty cycles. The main control chip 201 outputs these PWM signals to the internal motor drive isolation amplifier circuits of the floodlight module 204 and the zoom camera 203, respectively. The zoom camera 203 is equipped with a dual-channel drive circuit that receives independent PWM signals, thereby achieving precise servo control of the translation of the lens 2042 inside the floodlight module 204, the zoom of the lens in the zoom camera 203, and the focusing motor. This electromechanical control topology provides hardware support for the subsequent implementation of adaptive photoelectric electromechanical feedback in the system.

[0057] In the opto-electro-mechanical execution node 20 provided by the present invention, the main control chip 201 serves as the core of computation and control, and is mainly responsible for the spatial coordinate system alignment and cross-identification computation of heterogeneous sensor data.

[0058] The specific data processing and control process of the main control chip 201 includes the following steps: S501, the main control chip 201 continuously monitors the voltage level of the vehicle's accessory power lines through the vehicle status detection circuit. When the voltage of the accessory power lines drops to a preset threshold and an engine sleep message appears on the vehicle's local area communication bus, the main control chip 201 determines that the vehicle engine is off and the main power supply is disconnected. The main control chip 201 activates a tiered power gating strategy, cutting off the backup power supply to the zoom camera 203 and the floodlight module 204 by controlling the level state of the general-purpose input / output pins, thus enabling the opto-electro-mechanical execution node 20 to enter a low-power operation state.

[0059] In step S502, after the system receives the bus network wake-up signal, the main control chip 201 synchronously extracts the target polar coordinates output by the microwave radar module 101 and the terminal relative polar coordinates output by the BLE beacon module 102 from the spatial vector data packet matrix sent by the spatial sensing node 10. In practical applications, since the spatial sensing node 10 is installed in a non-metallic area inside the vehicle, its origin has a fixed installation offset relative to the center of the vehicle's front end, and there is also a hardware layout gap between the microwave radar antenna and the Bluetooth antenna array. Therefore, when performing coordinate mapping, the main control chip 201 retrieves the pre-stored device installation offset parameters and antenna spacing parameters to construct a spatial translation transformation matrix. Using this spatial translation transformation matrix, the main control chip 201 first converts the originally acquired target polar coordinates and terminal relative polar coordinates into two-dimensional plane coordinates in the local rectangular coordinate system through trigonometric functions for linear translation and alignment. Then, it uses the Pythagorean theorem and inverse trigonometric functions to uniformly and inversely map them to the local two-dimensional polar coordinate system with the vertical axis of the vehicle's front end center as the polar axis, obtaining the aligned target polar coordinates and terminal polar coordinates.

[0060] In S503, the main control chip 201 activates its internal hardware timer to continuously acquire multiple frames of mapped and aligned polar coordinate data within a set time window. The length of this time window is set to between 1 and 3 seconds to filter out random position drift caused by single sampling. Within the time window, the main control chip 201 calculates the spatial Euclidean distance difference between the target polar coordinates and the terminal polar coordinates in the same time slice, defining this distance difference as the spatial deviation factor. The main control chip 201 uses the law of cosines to solve for the straight-line distance between these two coordinate points in a two-dimensional plane, thereby quantifying the degree of difference in their positions. The process of the main control chip 201 calculating the spatial deviation factor is expressed by the following formula: ; In the above formula, Indicates the first The spatial deviation factor is calculated from each sampling frame, and the unit is meters. Indicates the first Radial distance of the aligned target polar coordinates in each sampling frame; Indicates the first The radial distance of the aligned terminal polar coordinates in each sampled frame; Indicates the first The azimuth angle of the aligned target polar coordinates in each sampling frame; Indicates the first The azimuth angle of the aligned terminal polar coordinates in each sampled frame.

[0061] S504, after reaching the set time window cutoff point, the main control chip 201 sums up a series of spatial deviation factors generated within the time window and divides them by the total number of sampling frames to calculate the average value of the sequence spatial deviation factors.

[0062] S505, the main control chip 201 compares the calculated average value with a preset tolerance threshold. This preset tolerance threshold is set based on the limb extension limits of a normal adult holding or carrying a mobile terminal, and its value ranges from 0.5 meters to 1.0 meter. When the calculated average value is less than or equal to the preset tolerance threshold, the main control chip 201 determines that the target detected by the radar and the user holding a legitimate mobile terminal are spatially overlapping. The main control chip 201 outputs a cross-authentication status of "authorized" in its register, indicating that the target approaching the vehicle is the owner of the vehicle carrying an authenticated terminal.

[0063] S506, when the calculated average value is greater than the preset tolerance threshold, or when the terminal relative polar coordinates extracted within the time window remain null, the main control chip 201 determines that there is no legitimate mobile terminal around the target detected by the radar. The main control chip 201 outputs the cross-authentication status as unauthorized in the register, indicating that the target approaching the vehicle is an unauthorized external person or moving object, and relevant evidence collection and warning actions need to be triggered.

[0064] For the direct memory access mechanism for extracting low-level data from the main control chip 201 and the basic configuration of the floating-point arithmetic unit, those skilled in the art can write the code according to the development manual of a conventional vehicle microcontroller. The basic data handling and computing hardware principles are well-known technologies in this field and will not be elaborated here.

[0065] In the opto-electro-mechanical execution node 20 provided by the present invention, the video control module 202 mainly undertakes image signal processing, image coordinate mapping and video compression and storage tasks.

[0066] The video control module 202 integrates an image signal processor and a hardware video encoding engine. For the Bayer array color restoration, automatic exposure adjustment, and H.265-based inter-frame compression algorithm of the image sensor, those skilled in the art can configure it according to conventional digital image processing specifications. Its underlying pixel processing logic is well-known in the field and will not be elaborated upon here.

[0067] After receiving the working state switching command issued by the main control chip 201, the video control module 202 performs the following operation steps: S601, in normal parking monitoring mode, when the main control chip 201 does not output an authorized or unauthorized status, the video control module 202 maintains time-lapse recording mode. The video control module 202 controls the zoom camera 203 via the mobile industrial processor interface bus to extract single-frame images at set intervals, synthesizes these single-frame images into a video file at a preset low frame rate, and writes it to the local storage medium. The set interval is set to a value between 1 second and 5 seconds to reduce static power consumption and storage space consumption while ensuring monitoring continuity.

[0068] S602, when the main control chip 201 outputs an unauthorized status, the video control module 202 receives the status flag via a register interrupt and immediately exits the time-lapse recording mode. The video control module 202 sends a full-speed sampling command to the zoom camera 203, restoring it to the standard video acquisition rate of 30 to 60 frames per second to record the approaching process of the external target.

[0069] S603, to achieve effective visual capture of the approaching target, the video control module 202 retrieves the target's polar coordinates after mapping and alignment by the main control chip 201. The video control module 202 utilizes the pinhole camera imaging model to convert the azimuth and radial distance in the target's polar coordinates into pixel coordinates on the image sensor plane and the estimated pixel width occupied by the target. The lateral offset of the target's spatial azimuth after projection through the lens's optical center is geometrically quantized by multiplying the tangent of the azimuth by the lens's current focal length. This offset is then superimposed with the principal point offset reference of the lens 2042, thus completing the coordinate transformation from real physical space to digital pixel space. The geometric calculation for coordinate mapping performed by the video control module 202 is expressed by the following formula: ; ; In the above formula, This represents the horizontal coordinate of the target in the image pixel coordinate system, in pixels. This indicates the horizontal coordinates of the principal point of the zoom camera 203, which are the optical center parameters calibrated at the factory for the zoom camera 203. Its value is usually half of the total number of horizontal pixels of the image sensor. This represents the equivalent pixel focal length in the horizontal direction corresponding to the current optical lens focal length of the zoom camera 203. Since this embodiment uses a zoom camera, this focal length is a dynamic variable. The main control chip 201 obtains this value in real time by calling a preset focal length calibration lookup table in the memory, based on the pulse width modulation signal currently output to the motor drive circuit or the actual step position of the zoom motor. Its specific value range is determined by the physical zoom ratio of the lens, and typically varies dynamically between several hundred and several thousand pixels. This indicates the azimuth angle of the physical target acquired by the microwave radar module 101; This indicates the estimated pixel width that the target will occupy in the image. This indicates the preset physical reference width of the external target, which is set to a range of 0.5 meters to 0.8 meters when monitoring human targets. This indicates the instantaneous relative distance of the physical target acquired by the microwave radar module 101.

[0070] S604, the video control module 202 calculates the horizontal coordinate value. With the estimated pixel width Then, using the horizontal coordinate values A photosensitive area mask of corresponding width is extracted from the center. The video control module 202 feeds this photosensitive area mask back to its internal image signal processor. The image signal processor independently calculates exposure compensation parameters based on the ambient brightness data corresponding to the pixel array covered by the photosensitive area mask and sends them to the zoom camera 203. This local metering strategy based on radar coordinate mapping enables the zoom camera 203 to prioritize the brightness and contrast of the area near the target in complex lighting environments such as nighttime parking garages, avoiding global exposure inaccuracies caused by strong background light sources or large areas of shadow.

[0071] In the opto-electro-mechanical execution node 20 provided by the present invention, the zoom camera 203 serves as a front-end execution device for visual acquisition, and is used to provide local feature images when an unauthorized target approaches, so as to achieve dynamic field-of-view capture of the target.

[0072] The zoom camera 203 internally includes a zoom optical lens group, a focusing optical lens group, an image sensor array, and a miniature servo motor for driving the translation of the lens 2042. Those skilled in the art can select the appropriate components for the photoelectric conversion structure of the image sensor array, the filter configuration, and the closed-loop control circuit of the miniature servo motor based on conventional vehicle monitoring hardware specifications. The underlying photoelectric conversion and electromechanical drive principles are well-known technologies in the field and will not be elaborated upon here.

[0073] When performing adaptive zoom and focus control, the zoom camera 203 executes the following steps: S701, the main control chip 201 activates the adaptive zoom logic of the zoom camera 203 based on the unauthorized status output by cross-authentication. The control register inside the zoom camera 203 retrieves the radial distance data in the target polar coordinates mapped and aligned by the main control chip 201 via the data bus.

[0074] S702, the zoom camera 203 calculates the required target focal length based on the radial distance data. Since the pixel size of the external target imaged on the image sensor plane is directly proportional to the focal length of the lens 2042 and inversely proportional to the physical distance to the target, in order to ensure that approaching targets at different distances occupy a sufficient proportion of pixels in the final captured image to identify facial or body details, the system needs to dynamically extend or shorten the focal length according to the real-time distance of the target. The logic for the zoom camera 203 to calculate the target focal length is expressed by the following formula: ; In the above formula, This indicates the target focal length that the zoom camera 203 needs to be adjusted to achieve, in equivalent pixels. This indicates the instantaneous relative distance of the physical target acquired by the microwave radar module 101, in meters; This indicates the preset physical reference width of the external target, which ranges from 0.5 meters to 0.8 meters when setting up monitoring for human targets. This represents the preset target pixel width that the target is expected to occupy in the image. In order to ensure that the subject is sharp while retaining some environmental background information, the value of this preset target pixel width is usually set between one-third and one-half of the total horizontal pixel resolution of the image sensor.

[0075] The drive control circuit inside the S703 zoom camera 203 receives the target focal length value and maps it to the target number of steps for the micro servo motor by consulting a focal length calibration table stored in the read-only memory. This focal length calibration table is generated during the device's manufacturing process through standard darkroom testing and records the corresponding mapping values ​​between different mechanical translation positions of the micro servo motor and the equivalent pixel focal length of the image. The drive control circuit outputs a corresponding pulse width modulation signal, driving the micro servo motor to mechanically translate the zoom optical lens group along the optical axis, so that the actual equivalent pixel focal length of the lens reaches the calculated target focal length.

[0076] S704, after completing the translation of the zoom optical lens group, the change in the relative position of the lens group during optical zoom will cause the focal plane to shift, resulting in blurring of the image captured by the image sensor. The zoom camera 203 retrieves the photosensitive area mask captured by the video control module 202 in the previous step, and sets the two-dimensional pixel space covered by the mask as the focus area of ​​interest.

[0077] The S705 zoom camera 203 controls an independent focusing motor to fine-tune the position of the focusing optical lens group, simultaneously executing an autofocus algorithm based on image contrast detection within the region of interest. The image signal processor continuously calculates the gradient values ​​of high-frequency components of the image within this region of interest; when the gradient value reaches its peak, focusing is complete. Using a local region based on target polar coordinate mapping as the focusing reference avoids defocusing issues caused by complex background environments or foreground raindrops at the focusing algorithm level, ensuring that the target features of the approaching vehicle remain in sharp focus.

[0078] In the opto-electro-mechanical execution node 20 provided by the present invention, the floodlight module 204 serves as an auxiliary light source for visual acquisition, and is used to provide directional illumination for the zoom camera 203 at night or in low-light environments.

[0079] The floodlight module 204 internally includes an array of light-emitting diodes (LEDs), a focusing lens group, and a stepper motor 2041 that drives the focusing lens group to translate. For the design of the constant current drive circuit for the LED array and the basic control logic of the stepper motor 2041, those skilled in the art can refer to the hardware specifications of conventional electromechanical lighting devices for configuration. Its light-emitting and mechanical transmission principles are well-known technologies in the field and will not be elaborated upon here.

[0080] When performing adaptive lighting and power control, the floodlight module 204 executes the following steps: In step S801, when the main control chip 201 determines that the ambient light level of the vehicle's environment is lower than a preset threshold, it sends an activation command to the floodlight module 204 to illuminate the LED beads 2043. The specific value of this preset threshold is set according to the conventional low-light standard at night or in an underground parking garage, typically between 10 and 50 lux. The ambient light level data is obtained by the main control chip 201 through calling the global ambient brightness evaluation parameters output by the image signal processor inside the video control module 202, or through acquisition by an independent photoresistor set on the surface of the device casing. At this time, if the cross-authentication state is authorized, the floodlight module 204, in conjunction with the aforementioned wide field-of-view floodlight state, provides welcome lighting for the vehicle owner; if the cross-authentication state is unauthorized, the floodlight module 204 retrieves the radial distance data from the target polar coordinates mapped and aligned by the main control chip 201 via the internal data bus to execute subsequent adaptive spotlight tracking lighting.

[0081] S802, the floodlight module 204 calculates the required target illumination beam angle based on the radial distance data. According to the laws of optical projection, to avoid the dispersion of light source energy and reduce light pollution to the surrounding environment, the spot of the illumination beam should exactly cover the spatial range where the target is located. Since the spot diameter increases with the increase of illumination distance, the floodlight module 204 needs to dynamically adjust the divergence angle of the beam according to the actual distance to the target. The logic of the floodlight module 204 in calculating the target illumination beam angle is expressed by the following formula: ; In the above formula, This indicates the target illumination beam angle that the floodlight module 204 needs to adjust to achieve, in degrees or radians; This indicates the instantaneous relative distance of the physical target acquired by the microwave radar module 101, in meters; This indicates the preset baseline width of the lighting coverage. Considering the redundant space for target movement and the illuminance attenuation at the edge of the light spot, the value of this baseline width of the lighting coverage is usually set between 1.5 meters and 2.5 meters.

[0082] In step S803, the microcontroller inside the floodlight module 204 receives the calculated target illumination beam angle and consults the lens 2042 translation calibration table stored in the memory. This calibration table records the mapping relationship between the mechanical stepping position of the stepper motor 2041 and the beam divergence angle. The microcontroller converts the angle value into the corresponding target running step number and outputs a drive pulse signal to the motor driver chip. The stepper motor 2041 drives the focusing lens group to mechanically translate along the light source emission direction, changing the optical distance between the lens group and the light-emitting diode array, so that the divergence angle of the actual emitted beam reaches the calculated target illumination beam angle.

[0083] S804, while adjusting the beam angle, the floodlight module 204 simultaneously calculates the required illumination drive power. Based on the physical law that illuminance decreases with propagation distance, the illuminance received by the target surface is inversely proportional to the square of the distance to the light source. To ensure that the reflected illuminance approaching the target surface remains within the ideal light sensitivity range required by the zoom camera 203 image sensor, and to avoid overexposure at close range or underexposure at long range, the floodlight module 204 dynamically adjusts the output power of the light-emitting diodes according to the radial distance to the target. The process by which the floodlight module 204 calculates the illumination drive pulse width modulation duty cycle is expressed by the following formula: ; In the above formula, This indicates the duty cycle of the pulse width modulation signal output to the LED array, and its value ranges from 0% to 100%. This represents the power compensation coefficient, which is obtained by calibrating the device at the factory in a standard darkroom environment based on the optimal signal-to-noise ratio of the image sensor, and is a constant parameter. This indicates the instantaneous relative distance of the physical target acquired by the microwave radar module 101, in meters; This represents the basic on-state duty cycle threshold of the floodlight module 204, used to overcome the on-state voltage drop of the hardware drive circuit. Its value is usually set between 5% and 15% based on the current-voltage characteristic curve of the selected light-emitting diode.

[0084] In step S805, the floodlight module 204 writes the calculated duty cycle value into the compare register of the hardware timer, generating a corresponding pulse width modulation level signal. This signal controls the on-time ratio of the switching transistors in the constant current drive circuit, thereby changing the average current flowing through the LED array. Through this linkage control mechanism, the floodlight module 204 can synchronously adjust the illumination beam range and luminous intensity according to the actual distance to the target, providing a stable and reliable supplementary lighting environment for video acquisition.

[0085] Specific application example: Nighttime parking monitoring in underground parking garages Scenario preset: Environmental parameters: The vehicle is parked in an underground garage with no natural light and an ambient illuminance of approximately 15 lux.

[0086] System status: Vehicle engine is off and main power is disconnected. The system is in a tiered power gating state (preset sleep mode). The zoom camera 203 and floodlight are ready for power cut-off. The microwave radar module is polling and scanning. The low-power Bluetooth beacon module is broadcasting radio frequency. The video control module 202 is recording time-lapse video by extracting a single frame every two seconds.

[0087] Authorization status event (vehicle owner approaches): Spatial Awareness and Wake-up: The vehicle owner approaches the vehicle with their bound smartphone (mobile terminal). When the owner is approximately 3.5 meters away from the vehicle, the low-power Bluetooth beacon module extracts the effective direct line-of-sight radial peak value and calculates the terminal's relative polar coordinates. When the owner moves to a distance of 2.8 meters from the vehicle (less than the preset distance threshold of 3 meters), the microwave radar detects the physical target and immediately wakes up the opto-electro-mechanical actuator to restore standby power.

[0088] Cross-authentication: The main control chip 201 aligns the polar coordinates of the radar target with the polar coordinates of the Bluetooth terminal. After continuous acquisition within a two-second time window, the average spatial deviation factor between the two is calculated to be approximately 0.4 meters (less than the preset tolerance threshold of 0.8 meters), and the authorized status is output.

[0089] Actions performed: The video control module 202 maintains time-lapse recording to avoid generating a large number of invalid files; the stepper motor 2041 inside the floodlight module 204 drives the lens 2042 close to the LED bead 2043 to form a wide field of view floodlight state, which serves as a welcome light to illuminate a wide area in front of the car.

[0090] Unauthorized events (stranger / suspicious person approaching): Spatial Awareness and Wake-up: A stranger without an authorized mobile phone approaches the vehicle. The microwave radar detects the target at 2.8 meters and wakes up the opto-electro-mechanical execution node 20. At this time, the low-power Bluetooth beacon does not receive a valid signal (or only receives spurious peaks of metallic reflections from other vehicles in the vicinity, which, after being masked out, have no matching signal).

[0091] Cross-authentication: Due to the lack of effective terminal polar coordinates, the spatial deviation factor is infinite (or the null identifier is continuously output), and the main control chip 201 outputs an unauthorized status.

[0092] Execution actions (adaptive tracking and illumination): Normal high frame rate: The video control module 202 immediately switches to full-speed recording mode at sixty frames per second.

[0093] Adaptive zoom and focus: As the stranger gradually approaches from 2.8 meters to 1.0 meters, the main control chip 201 dynamically calculates the target's focal length based on the radial distance in polar coordinates. The zoom motor continuously drives the lens to ensure that the stranger always occupies about one-third of the pixel width in the image (clearly capturing facial and body features); and completes precise autofocus within the local area mapped by the target's radar coordinates.

[0094] Reference Appendix Figure 3 This diagram illustrates the stability of the field of view as a target approaches. The horizontal axis (relative distance to the target) represents the physical distance between the person and the vehicle, decreasing gradually from left to right, corresponding to the person approaching the car window. The vertical axis (the width of the image pixels occupied by the target) represents the horizontal size of the person in the monitored image, measured in pixels. The hollow triangle with a solid line (traditional fixed-focus lens) shows that as the person approaches, their proportion in the image increases dramatically, easily causing facial features to extend beyond the image boundaries. In contrast, the solid rhombus with a dashed line (adaptive zoom lens of this invention) shows that the system shortens the focal length in real time based on distance data, keeping the person's pixel size in the image consistently around a preset baseline width (stable at approximately 500 pixels in the diagram). The undulating dashed line reflects the dynamic fine-tuning process of the mechanical motor during focusing, achieving stable acquisition of the target image.

[0095] Adaptive lighting: The floodlight lens 2042 shifts away from the LED 2043 (focusing state), and the beam angle dynamically decreases with the distance from the stranger, so that the light spot tightly covers the suspect; at the same time, the pulse width modulation duty cycle decreases proportionally to the square of the distance (the closer the distance, the lower the output power) to prevent overexposure and whitening when shooting the suspect's face at close range.

[0096] Reference Appendix Figure 4 This diagram illustrates a comparison between the floodlight's illumination drive power and overexposure prevention. The horizontal axis represents the physical distance of an external person approaching the vehicle; the vertical axis represents the floodlight's drive duty cycle (i.e., the proportion of output power). The hollow inverted triangle with a solid line (traditional constant brightness supplemental lighting) forms a horizontal straight line, indicating that after an alarm is triggered, the supplemental lighting continuously illuminates at a constant maximum power of approximately 85% duty cycle, easily leading to overexposure of faces at close range and loss of features. The asterisk with a dashed line (adaptive overexposure prevention dimming of this invention) shows that as external persons approach, the system adaptively reduces the output current proportion based on the distance change. This control strategy avoids the whitening effect of strong light at close range and significantly reduces power consumption during parking monitoring.

Claims

1. A parking monitoring system supporting time-lapse recording and event triggering, characterized in that, It includes a spatial sensing node (10) and an opto-electro-mechanical execution node (20) that communicate with each other. The microwave radar module (101) in the spatial sensing node (10) acquires the instantaneous relative distance and azimuth of the physical target, and the BLE beacon module (102) acquires the terminal relative polar coordinates of the mobile terminal. The main control chip (201) in the opto-electro-mechanical execution node (20) maps the instantaneous relative distance, azimuth angle and terminal relative polar coordinates to the local two-dimensional polar coordinate system and outputs the cross-authentication status. According to the cross-authentication status, the video control module (202) switches the time-lapse recording encoding status, drives the zoom camera (203) to change the optical field of view, and drives the stepper motor (2041) in the floodlight module (204) to drive the lens (2042) to translate relative to the lamp bead (2043) to change the beam state.

2. A parking monitoring system supporting time-lapse recording and event triggering according to claim 1, characterized in that, The spatial sensing node (10) adopts a multi-layer printed circuit board stacked structure. The high-frequency microwave antenna layer of the microwave radar module (101) and the radio frequency antenna array layer of the BLE beacon module (102) maintain a set vertical distance in physical space. A copper-clad ground shielding layer is laid between the two antenna boards. The floodlight module (204) also includes a lead screw drive mechanism and a linear guide rail. The lens (2042) is slidably mounted on the linear guide rail. The output shaft of the stepper motor (2041) is mechanically connected to the lead screw drive mechanism.

3. A parking monitoring system supporting time-lapse recording and event triggering according to claim 1, characterized in that, When the main control chip (201) receives the signal that the vehicle engine is off and the main power is disconnected, it cuts off the backup power supply of the motor servo drive circuit and the floodlight module (204) inside the zoom camera (203). The image sensor of the zoom camera (203) retains the low-power basic power supply. The video control module (202) controls the zoom camera (203) to acquire images and perform time-lapse video encoding. When the microwave radar module (101) detects that the physical target has exceeded the preset distance threshold, it sends a wake-up interrupt signal to the opto-electro-mechanical execution node (20), and the main control chip (201) restores the backup power supply to the zoom camera (203) and the floodlight module (204).

4. A parking monitoring system supporting time-lapse recording and event triggering according to claim 1, characterized in that, The BLE beacon module (102) acquires phase difference data of multi-antenna received signals and generates a spatial spectrum peak sequence; The BLE beacon module (102) calls a preset static metal reflection spatial map to perform mask comparison on the spatial spectrum peak sequence; The static metal reflection spatial map is a set of values ​​containing multiple forbidden azimuth intervals; The BLE beacon module (102) compares the azimuth values ​​of each spatial spectrum peak with the forbidden azimuth interval, removes static reflection pseudo-peaks that fall within the forbidden azimuth interval, extracts the peak with the highest energy amplitude in the remaining sequence as the effective direct line-of-sight radial peak, takes the polar coordinate angle corresponding to the effective direct line-of-sight radial peak as the terminal relative azimuth angle, and calculates the terminal relative polar coordinates by combining the received signal strength indication attenuation data.

5. A parking monitoring system supporting time-lapse recording and event triggering according to claim 1, characterized in that, The spatial sensing node (10) forcibly encapsulates the instantaneous relative distance and azimuth of the physical target, the relative polar coordinates of the terminal, and the system timestamp into the same spatial vector data packet matrix; When the spatial sensing node (10) obtains the instantaneous relative distance and azimuth angle, it sends the spatial vector data packet matrix to the opto-electro-mechanical execution node (20) through the vehicle local communication bus.

6. A parking monitoring system supporting time-lapse recording and event triggering according to claim 1, characterized in that, The main control chip (201) continuously collects multiple frames of mapped and aligned polar coordinate data within a set time window, and calculates the spatial Euclidean distance difference between the instantaneous relative distance and azimuth angle of the physical target and the relative polar coordinates of the terminal under the same time slice as the sequence spatial deviation factor. The main control chip (201) sums up the sequence spatial deviation factors generated within the time window and then divides them by the total number of sampling frames to obtain the average value of the sequence spatial deviation factors; The main control chip (201) compares the average value of the sequence spatial deviation factor with the preset tolerance threshold. When the average value of the sequence spatial deviation factor is not greater than the preset tolerance threshold, the cross-authentication status is output as authorized. When the average value of the sequence spatial deviation factor is greater than the preset tolerance threshold, the cross-authentication status is output as unauthorized.

7. A parking monitoring system supporting time-lapse recording and event triggering according to claim 6, characterized in that, When the cross-authentication status is authorized, the main control chip (201) maintains the time-lapse video encoding status of the video control module (202); The main control chip (201) outputs a pulse width modulation signal to drive the stepper motor (2041) to move the lens (2042) closer to the lamp bead (2043), and the floodlight module (204) switches to the wide field of view floodlight state.

8. A parking monitoring system supporting time-lapse recording and event triggering according to claim 6, characterized in that, When the cross-authentication status is unauthorized, the main control chip (201) controls the video control module (202) to switch to high frame rate regular video recording encoding status; The main control chip (201) extracts the instantaneous relative distance and azimuth angle as servo tracking parameters, drives the stepper motor (2041) to move the lens (2042) away from the lamp bead (2043), and the floodlight module (204) switches to a small field of view focusing state; The video control module (202) extracts the photosensitive area mask based on the instantaneous relative distance and azimuth angle and feeds it back to the internal image signal processor. The image signal processor independently calculates the exposure compensation parameters based on the photosensitive area mask and sends them to the zoom camera (203). The zoom camera (203) calculates the target focal length based on the instantaneous relative distance and maps the target focal length to the target running steps of the internal micro servo motor to drive the translation of the internal zoom optical lens group.

9. A parking monitoring system supporting time-lapse recording and event triggering according to claim 8, characterized in that, The zoom camera (203) retrieves the photosensitive area mask captured by the video control module (202) and sets the two-dimensional pixel space covered by the photosensitive area mask as the focus area of ​​interest. The zoom camera (203) controls the internal focusing motor to fine-tune the position of the internal focusing optical lens group, and simultaneously calculates the gradient value of the high-frequency components of the image in the focusing interest area to perform autofocus.

10. A parking monitoring system supporting time-lapse recording and event triggering according to claim 8, characterized in that, The floodlight module (204) calculates the target illumination beam angle based on the instantaneous relative distance; The microcontroller inside the floodlight module (204) consults the translation calibration table of the lens (2042), converts the target illumination beam angle into the target running steps, and drives the stepper motor (2041) to drive the lens (2042) to perform mechanical translation along the light source emission direction; The floodlight module (204) calculates the duty cycle of the lighting drive pulse width modulation based on the instantaneous relative distance, writes the duty cycle value into the comparison register of the internal hardware timer to generate the corresponding pulse width modulation level signal to control the on-time ratio of the switching transistor in the internal constant current drive circuit.

Images