A radar-based luminaire control method and system

Abstract

A radar-based lamp control method and system, comprising: respectively through the radar sensing module integrated in multiple types of lamps with different installation heights, emitting radar signals to the indoor monitoring area and receiving echo signals reflected by target objects; processing each echo signal to extract the motion speed feature, the center of gravity height change feature, the vertical direction acceleration feature and the point cloud spatial distribution feature of the target object; time sequence alignment, coordinate system and weighted fusion processing of echo data from lamps with different installation heights, judging whether the target object is in an abnormal state such as falling based on the fusion feature parameters; and outputting lamp control instructions according to the judgment result. The height stratification characteristics of ceiling lamps, table lamps and floor lamps naturally existing in indoor space are utilized to construct a three-dimensional multi-angle radar sensing network, realizing reliable detection of indoor human abnormal state without image acquisition, without wearing equipment and zero incremental deployment cost.

Description

Technical Field

[0001] This invention relates to the field of intelligent sensing and lighting control technology, specifically to a method and system that integrates a radar sensing module into indoor lighting fixtures, utilizes the hierarchical characteristics of various types of lighting fixtures in indoor spaces to construct a three-dimensional radar sensing network, and realizes indoor human status detection and lighting fixture linkage control. Background Technology

[0002] With the accelerating aging of the population and the increasing number of people living alone, real-time monitoring of abnormal indoor human conditions, especially the automatic detection and alarm of falls, has become an important demand in the fields of smart homes and health management. How to achieve reliable perception of indoor human conditions without infringing on user privacy or requiring users to actively wear devices is a core problem that needs to be solved in this field.

[0003] Existing technologies mainly include the following types of solutions:

[0004] Vision-based camera recognition solutions: These solutions involve deploying cameras indoors to collect image or video data, then using computer vision algorithms to analyze human posture and identify abnormal states such as falls. While this approach provides relatively intuitive detection results, it suffers from the following significant drawbacks: The continuous collection of image data by cameras involves highly sensitive personal privacy information such as user facial features, body shape, and behavioral habits, facing both legal compliance risks and low user acceptance; the reliability of visual recognition decreases significantly in low-light environments (such as at night or in corridors); furthermore, the large volume of image data places high demands on storage and transmission bandwidth.

[0005] Wearable device-based detection solutions collect human motion data and make fall detection by requiring users to wear devices such as accelerometers, gyroscopes, or specific smart bracelets. The limitations of this solution are: it requires users to continuously and actively wear the device, which leads to poor compliance among the elderly, people with mobility impairments, or those with cognitive decline; users often remove the device during sleep or bathing, resulting in detection blind spots; the device requires regular charging and maintenance, increasing the burden on users; and solutions relying solely on wearable devices cannot cover scenarios where abnormalities occur when the user is not wearing the device.

[0006] A dedicated deployment scheme based on independent radar sensors involves deploying radar sensors as standalone devices indoors, using radar echo signals to sense human movement. This scheme addresses privacy concerns to some extent because the radar only outputs physical data of the target object's motion, without involving biometric information such as images. However, this scheme still has the following problems: independent radar sensors are non-standard configuration devices in indoor environments, requiring additional planning of installation locations and angles, increasing deployment costs and construction complexity; the appearance of independent devices is visually jarring and clashes with home décor, limiting user acceptance; a single radar sensor installed at a fixed height and angle provides only a single dimension of human state perception, posing a risk of false alarms and missed alarms in fall detection due to limited observation angles; and when multiple sensors are deployed collaboratively, the location planning and debugging are complex and difficult for ordinary users to complete independently.

[0007] In summary, existing technologies struggle to simultaneously address the four dimensions of privacy protection, seamless use, low deployment costs, and reliable detection, and a systematic solution capable of meeting all of these requirements has yet to be developed. Summary of the Invention

[0008] To address the technical problems of existing technologies, such as visual recognition solutions infringing on user privacy, wearable device solutions having poor user compliance, high deployment costs and low appearance integration of independent radar, and radar with a single installation angle being prone to false alarms and missed alarms due to its limited sensing dimension, this invention provides a radar-based lighting control method and system. It utilizes the inherent high-layer characteristics of ceiling lights, table lamps, and floor lamps in indoor spaces to construct a three-dimensional, multi-angle radar sensing network and fuses multi-source echo data to improve the reliability of indoor human abnormality detection.

[0009] To achieve the above objectives, the first aspect of the present invention provides a radar-based lighting control method, comprising the following steps:

[0010] S1. Radar signals are transmitted to the indoor monitoring area through radar sensing modules integrated in ceiling lights, table lamps at table height, and floor lamps near the ground, respectively, and radar echo signals reflected by target objects within the monitoring area are received. Specifically, the radar sensing module integrated in the ceiling lights radiates radar signals to the monitoring area from a top-down angle, the radar sensing module integrated in the table lamps radiates radar signals to the monitoring area from a side or oblique angle, and the radar sensing module integrated in the floor lamps radiates radar signals to the monitoring area from a horizontal or upward angle.

[0011] S2. Perform signal processing on the radar echo signals received by each of the radar sensing modules to extract feature parameters reflecting the motion state of the target object. The feature parameters include at least the height change feature of the target object's center of gravity relative to the ground.

[0012] S3. The echo data from the radar sensing modules of the ceiling light, the table lamp and the floor light are sequentially processed by time alignment, coordinate system and weighted fusion to obtain the fusion feature parameters of the target object, and the target object is determined to be in an abnormal state based on the fusion feature parameters.

[0013] S4. Based on the judgment result, output the corresponding lighting control command.

[0014] Furthermore, the radar sensing module integrated in each of the aforementioned lamps adopts at least one of frequency-modulated continuous wave radar, ultra-wideband radar, or millimeter-wave radar.

[0015] Further, the feature parameters in step S2 also include at least one of the following: the motion velocity feature of the target object, the acceleration feature of the target object along the vertical direction, and the spatial distribution feature of the radar reflection point cloud of the target object; wherein, the motion velocity feature is obtained by analyzing the Doppler frequency shift of the radar echo signal; the height change feature of the center of gravity relative to the ground is obtained by analyzing the echo energy distribution from different range gates, and calculating the height estimate of the center of gravity of the target object and its height change curve over time by combining the installation height and radiation angle parameters of each radar sensing module; the vertical acceleration feature is obtained by performing second-order difference calculation on the height change curve; the spatial distribution feature of the point cloud is obtained by jointly processing the radar echo signal in the range dimension, azimuth dimension, and Doppler dimension to obtain the three-dimensional point cloud distribution of the target object, and extracting at least one morphological parameter among the projected area of ​​the three-dimensional point cloud in the horizontal plane, the height of the point cloud center of gravity, and the aspect ratio of the point cloud.

[0016] Furthermore, the step S3 of determining whether the target object is in an abnormal state includes: determining whether the change in the height of the target object's center of gravity within a preset time window T1 exceeds a preset height threshold H. th Furthermore, if the height change direction is from high to low, then it is further determined whether the target object remains stationary or in a low-speed motion state within a preset static judgment time T2 after the height change occurs. The low-speed motion state is defined as the absolute value of the target object's motion speed being less than a preset speed threshold V. th If the target object remains stationary or in a low-speed motion state for the preset static judgment time T2, it is determined that the target object has fallen and a fall alarm signal is generated.

[0017] Further, the weighted fusion processing in step S3 includes: assigning a first weight, a second weight, and a third weight to the top-view echo data of the top light, the side-view echo data of the table lamp, and the horizontal echo data of the ground lamp according to their contribution to the judgment of the abnormal state; and performing a weighted summation of the feature parameters corresponding to each echo data to obtain the fusion feature parameters.

[0018] Further, the coordinate system processing in step S3 includes: establishing an absolute coordinate system with the ground center of the monitoring area as the origin; constructing a rotation matrix and translation vector from the local coordinate system of each radar sensing module to the absolute coordinate system based on the installation height, radiation angle, and relative position parameters of each radar sensing module; and using the rotation matrix and translation vector to transform the target position information described by each radar sensing module in its local coordinate system to the absolute coordinate system.

[0019] Furthermore, the extraction of feature parameters in step S2 and the judgment of abnormal states in step S3 are both performed in the signal processing unit of the lamp; the method only outputs state identification information representing the state of the target object to the external system, and does not transmit the original radar echo data to the external system.

[0020] Furthermore, the method also includes: acquiring in real time the quantity and movement pattern information of target objects within the monitoring area; when multiple moving targets exist simultaneously within the monitoring area, performing target separation and tracking on the echo data of each moving target; and dynamically adjusting the preset height threshold H according to the preset movement capability level of the monitored object. th And the preset static judgment duration T2.

[0021] A second aspect of the present invention provides a radar-based lighting control system, comprising a first radar light fixture, a second radar light fixture, a third radar light fixture, a data fusion processing unit, and a control unit; the first radar light fixture is installed on the indoor ceiling, comprising a ceiling light body and a first radar sensing module integrated within the ceiling light body, the first radar sensing module transmitting radar signals to the monitoring area from a top-down angle and receiving echo signals; the second radar light fixture is installed at a lower height than the first radar light fixture, comprising a table lamp body and a second radar sensing module integrated within the table lamp body, the second radar sensing module transmitting and receiving signals from a side-view or oblique-view angle. The echo signal; the third radar light fixture is installed at a lower height than the second radar light fixture, including the ground light body and the third radar sensing module integrated in the ground light body. The third radar sensing module transmits and receives echo signals at a horizontal or upward angle; the data fusion processing unit is communicatively connected to the three radar sensing modules respectively, performs time sequence alignment, coordinate system and weighted fusion processing on each echo data, extracts the height change characteristics of the target object's center of gravity relative to the ground and obtains fused feature parameters, and judges the abnormal state accordingly; the control unit is connected to the data fusion processing unit, and generates and issues light fixture control commands according to the judgment results.

[0022] By adopting the above technical solution, the present invention has the following beneficial effects:

[0023] Firstly, radar sensing technology essentially only collects physical data such as the target object's speed, distance, acceleration, and spatial distribution, without generating images, voiceprints, or other biometric information that can identify an individual. This eliminates the possibility of privacy violations from the source of data collection, giving it an essential advantage in privacy protection compared to visual recognition solutions.

[0024] Secondly, the detection function is automatically executed by the lamps with integrated radar sensing modules, without requiring the target to actively wear any devices, achieving truly non-intrusive and all-time detection coverage, and eliminating the wearing compliance problem compared to wearable device solutions.

[0025] Thirdly, this invention fully utilizes the fact that ceiling lights, table lamps, and floor lamps are standard features in indoor home environments, integrating radar sensing functionality into existing or new lighting equipment, eliminating the need for additional dedicated sensors and fundamentally reducing the deployment costs of independent devices.

[0026] Fourth, since the three types of lamps are naturally located at different height levels in the indoor space, the ceiling lamp provides a downward viewing angle, the table lamp provides a side viewing angle, and the floor lamp provides a horizontal or upward viewing angle. The three work together to form a three-dimensional multi-angle coverage of the monitoring area. The same change in human body state (especially sudden changes in posture such as falls) can be reflected in the radar echoes from multiple different angles at the same time. The multi-dimensional redundancy verification mechanism significantly improves the reliability of detection and reduces the false alarm and false alarm rates.

[0027] Fifth, after the radar sensing module is embedded in the lamp body, it does not significantly change the appearance of the lamp and can be naturally integrated with various home decoration styles, realizing the invisible deployment of sensing functions. Attached Figure Description

[0028] Figure 1 This is a block diagram of the radar-based lighting control system described in Embodiment 1 of the present invention;

[0029] Figure 2 This is a schematic diagram showing the arrangement of the three types of radar lights in an indoor space according to Embodiment 1 of the present invention, including their installation height and radiation angle.

[0030] Figure 3 This is a flowchart of the radar-based lighting control method described in Embodiment 1 of the present invention;

[0031] The attached figures are labeled as follows:

[0032] 100. First radar light fixture; 110. First radar sensor module; 200. Second radar light fixture; 210. Second radar sensor module; 300. Third radar light fixture; 310. Third radar sensor module; 400. Data fusion processing unit; 500. Control unit; 600. Monitoring area; 700. Target object; 800. External linkage equipment. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0034] It should be noted that the descriptions involving "first," "second," etc., in this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0035] Example 1

[0036] This embodiment uses a bedroom scenario as an example to describe the basic technical solution and complete workflow of the present invention. (Refer to...) Figure 1 The system in this embodiment includes a first radar light 100, a second radar light 200, a third radar light 300, a data fusion processing unit 400, and a control unit 500.

[0037] The first radar light fixture 100 is a ceiling light installed in a bedroom, typically at a height between 2.4 meters and 3.0 meters above the ground. The first radar light fixture 100 integrates a first radar sensing module 110 within its body. The first radar sensing module 110 is installed on the side of the light fixture facing the monitoring area 600, and its radar antenna radiates radar signals towards the monitoring area 600 at a roughly vertical downward viewing angle, with the beam coverage encompassing the main ground area of ​​the monitoring area.

[0038] The second radar lamp 200 is a table lamp placed on a bedside table or desktop in the bedroom, typically installed at a height between 0.7 meters and 1.2 meters from the ground. The lamp body of the second radar lamp 200 integrates a second radar sensing module 210. The radar antenna of the second radar sensing module 210 radiates radar signals towards the monitoring area 600 at a side-view or downward-sloping angle, primarily covering the mid-height cross-sectional area of ​​the monitoring area. It can effectively sense changes in the cross-sectional shape of the target object 700 when it is standing or walking, as well as lateral movement characteristics during the transition from standing to lying down.

[0039] The third radar light fixture 300 is a floor lamp placed near the bedroom floor, typically installed at a height between 0.1 meters and 0.5 meters above the ground. The third radar light fixture 300 integrates a third radar sensing module 310 within its body. The radar antenna of the third radar sensing module 310 radiates radar signals towards the monitoring area 600 at a roughly horizontal or slightly upward angle, primarily covering the lower space of the monitoring area. It can effectively sense the speed characteristics of the target object 700's lower limb movement and the abrupt change in the height of the lower limbs relative to the ground as the target object suddenly falls from a vertical position to the ground.

[0040] Reference Figure 2 The different installation heights of the three types of lamps in the indoor space enable the three radar sensing modules to radiate radar signals to the same monitoring area from three different angles: top-down, side-down, and horizontal. The beam coverage areas of the three overlap in three-dimensional space, eliminating the detection blind spots caused by the limited radar radiation angle in the single-angle deployment scheme, and realizing three-dimensional multi-angle blind-spot-free coverage of the monitoring area.

[0041] In this embodiment, each radar sensing module can be at least one of frequency-modulated continuous wave radar, ultra-wideband radar, or millimeter-wave radar. Preferably, the first radar sensing module 110 in this embodiment uses a millimeter-wave radar operating in the 60GHz band, which has a small antenna size and strong beam directivity, making it advantageous for miniaturized integration within the lamp body.

[0042] The data fusion processing unit 400 is communicatively connected to the first radar sensing module 110, the second radar sensing module 210, and the third radar sensing module 310, respectively. It receives echo data collected by the three radar sensing modules, fuses the data, extracts motion state characteristic parameters of the target object 700, and determines whether the target object 700 is in an abnormal state such as a fall. In this embodiment, the data fusion processing unit 400 is integrated inside the lamp body of the first radar lamp 100, sharing the same processor with the first radar sensing module 110. The second radar sensing module 210 and the third radar sensing module 310 transmit data with the processor via a wireless communication protocol.

[0043] The control unit 500 is connected to the data fusion processing unit 400 and is used to generate corresponding lighting control commands based on the status judgment results of the data fusion processing unit 400, and send the lighting control commands to at least one of the first radar lighting 100, the second radar lighting 200, and the third radar lighting 300, so as to control the corresponding lighting to perform corresponding control actions.

[0044] Reference Figure 3 The method in this embodiment includes the following steps:

[0045] Step S100: Multiple types of lighting fixtures collaboratively transmit radar signals and collect echo data. After system startup, the first radar sensing module 110, the second radar sensing module 210, and the third radar sensing module 310 transmit radar signals to the monitoring area 600 according to a preset timing sequence, and continuously receive radar echo signals from the target object 700 within the monitoring area 600. The transmission timing of the three radar sensing modules is coordinated through a time-division multiplexing mechanism to avoid mutual signal interference. When the target object 700 is not within the monitoring area, each radar sensing module maintains a low-power standby state, and automatically enters full-power operating mode when a moving target is detected in the echo signal.

[0046] Step S200: Echo signal processing and feature parameter extraction. The data fusion processing unit 400 processes the radar echo signals received by each radar sensing module to extract feature parameters reflecting the motion state of the target object 700. The feature parameters include at least the height variation characteristics of the target object 700's center of gravity relative to the ground, and may further include at least one of the following:

[0047] The motion velocity characteristics of the target object 700 are obtained by analyzing the Doppler frequency shift of the echo signal, reflecting the motion velocity components of the target object 700 in each direction within the monitoring area;

[0048] The height variation characteristics of the target object 700's center of gravity are obtained by analyzing the echo energy distribution from different range gates and combining the installation height and radiation angle parameters of each radar sensor module. The estimated height of the target object 700's center of gravity relative to the ground and its time-varying curve are calculated. For example, for the first radar sensor module 110 installed at a top-down angle and at an installation height of H0, the slant range R and elevation angle θ corresponding to the centroid of the target echo energy can be obtained by weighting the echo energy of each range gate. The estimated height of the target object's center of gravity relative to the ground is then calculated using h=H0−R·cosθ, and the height variation curve is obtained by arranging the samples according to the sampling time sequence.

[0049] The acceleration characteristics of the target object 700 in the vertical direction are obtained by performing second-order difference calculation on the above height change curve, reflecting the magnitude and direction of the target object 700's acceleration in the vertical direction;

[0050] The spatial distribution characteristics of the radar reflection point cloud of target object 700 are obtained by jointly processing the echo signal in the range, azimuth, and Doppler dimensions to obtain a three-dimensional point cloud distribution map of target object 700. Then, morphological parameters such as the projected area of ​​the point cloud in the horizontal plane, the centroid height of the point cloud, and the aspect ratio of the point cloud are extracted. These morphological parameters show significant differences between the target object 700 in an upright state and a horizontally lying state, and can serve as auxiliary features for attitude recognition.

[0051] Step S300: Multi-angle echo data fusion processing. The data fusion processing unit 400 performs the following fusion processing on the echo data from radar sensor modules at three different installation heights:

[0052] Step S310: Time Sequence Alignment Processing. Since there is a slight time difference in the sampling times of the three radar sensor modules, the data fusion processing unit 400 first aligns the three echo data streams according to their timestamps, unifying the three data streams to the same time reference to ensure consistency in the time dimension during subsequent fusion processing. As an example, the sampling time of the radar sensor module with the highest sampling rate can be selected as the reference time axis, and the other two data streams can be linearly interpolated according to their timestamps to obtain a data sequence aligned with the reference time axis.

[0053] Step S320: Coordinate System Processing. The data fusion processing unit 400, based on the installation height, radiation angle, and relative position parameters of each radar sensor module, transforms the target position information described by the three echo data in their respective sensor local coordinate systems into an absolute coordinate system with the center of the ground in the monitoring area as the origin. This ensures that the three data streams can consistently describe the same physical spatial location. As an example, let the target position vector in the local coordinate system of a certain radar sensor module be P. local Then its position vector P in the absolute coordinate system world Satisfy P world =R·P local +T, where R is a 3×3 rotation matrix determined based on the radiation angles (elevation and azimuth) of the radar sensor module, and T is a 3×1 translation vector determined based on the installation height of the radar sensor module and its relative position to the center of the ground in the horizontal plane. R and T for each radar sensor module can be determined by a one-time calibration after system installation and stored in the data fusion processing unit 400.

[0054] Step S330: Weighted Fusion Processing. The data fusion processing unit 400 assigns corresponding weights to the three echo data streams based on their contribution to the abnormal state judgment, performing weighted fusion to obtain the fused feature parameters of the target object 700. Specifically, in the fall state recognition task, the top-view echo data collected by the first radar sensing module 110 (top lamp) has the highest sensitivity to changes in the center of gravity height of the target object 700 and is assigned the first weight w1; the horizontal echo data collected by the third radar sensing module 310 (ground lamp) has a relatively high sensitivity to changes in the lower limb speed of the target object 700 and is assigned the third weight w3; the side-view echo data collected by the second radar sensing module 210 (table lamp) provides information on changes in the cross-sectional contour of the target object 700 and is assigned the second weight w2 as an auxiliary verification feature. All three satisfy w1 > w3 > w2 and w1 + w2 + w3 = 1. Preferably, w1 ranges from 0.4 to 0.6, w2 ranges from 0.1 to 0.3, and w3 ranges from 0.2 to 0.4; in a specific example, w1=0.5, w2=0.2, and w3=0.3. The fusion feature parameters can be obtained by weighted summation of each feature parameter and its corresponding weight. For example, the fusion centroid height change ΔH of the target object 700 is ΔH=w1·ΔH1+w2·ΔH2+w3·ΔH3, where ΔH1, ΔH2, and ΔH3 are the centroid height changes calculated based on the three echo data. The specific values ​​of the weights can be adjusted within the above ranges based on system calibration results and the actual detection scenario.

[0055] Step S400: Abnormal State Judgment. Based on the fusion feature parameters obtained in step S300, the data fusion processing unit 400 determines whether the target object 700 has experienced a fall event according to the following logic:

[0056] Step S410: Determine whether the change ΔH in the center-of-gravity height of the target object 700 within the preset time window T1 exceeds the preset height threshold H. th And the direction of change is from high to low (i.e., ΔH < −H). th The preset time window T1 ranges from 0.3 seconds to 1.0 seconds and is used to capture the time characteristic of the rapid descent of a person's center of gravity during a fall. The preset height threshold H... th The value ranges from 0.5 meters to 1.0 meter, and is used to distinguish between the rapid change in center of gravity height caused by a fall and the slow change in height caused by normal bending or squatting movements. In a specific example, T1 = 0.6 seconds, H... th =0.7 meters.

[0057] Step S420: If the judgment condition of step S410 is met, further determine whether the target object 700 remains in a static state or a low-speed movement state within the preset static judgment duration T2 after the height change occurs. The preset static judgment duration T2 ranges from 2 seconds to 5 seconds and is used to distinguish between the situation where the person cannot get up immediately after falling and the normal behavior of consciously sitting down quickly and then getting up immediately. The judgment condition for the low-speed movement state is that the absolute value of the target object 700's movement speed is less than a preset speed threshold V. th The preset speed threshold V th The value ranges from 0.1 m / s to 0.5 m / s, and is used to distinguish the slight residual movements of the human body after a fall from active movements such as normal walking and getting up. In a specific example, T2 = 3 seconds, V... th =0.3 m / s.

[0058] Step S430: If the judgment condition of step S420 is also met, that is, the center of gravity height of the target object 700 undergoes a downward change exceeding the threshold in a short period of time, and the target object 700 continues to be in a low-speed or stationary state after the change, then it is determined that the target object 700 has fallen. The data fusion processing unit 400 generates a fall alarm signal and transmits the fall alarm signal and the timestamp of the fall event to the control unit 500.

[0059] Step S440: If the judgment condition of step S410 or step S420 is not met, it is determined that no fall event has occurred, and the method flow returns to step S100 to continue the loop detection.

[0060] Step S500: Output lighting control commands. After receiving the fall alarm signal, the control unit 500 generates and outputs at least one of the following lighting control commands: controls the first radar light 100, the second radar light 200, and the third radar light 300 to perform flashing alarm actions at a preset frequency to alert surrounding personnel through sound and light; controls the three types of lights to turn on and adjust to the highest brightness level to provide sufficient lighting for the scene that may require rescue; sends an alarm notification containing fall event information and the time of occurrence to a preset terminal device through a wireless communication interface. The preset terminal device may include a family member's mobile phone, a community management platform, or an emergency rescue system; and coordinates other smart devices in the monitoring area to perform preset response actions, such as coordinating to unlock doors or reporting event information to the property management platform.

[0061] In this embodiment, all signal processing steps, feature parameter extraction steps, and abnormal state judgment steps are executed locally within the data fusion processing unit 400, and the raw radar echo data is not transmitted to the outside of the lamp body. The control unit 500 and the external linkage device 800 only receive status identification information representing the state of the target object (e.g., "normal" or "fall alarm"), and do not receive raw echo data or intermediate feature parameters. The status identification information does not contain any biometric features that can identify the individual identity of the target object, thus protecting user privacy from the technical architecture level.

[0062] This embodiment takes a living room scenario as an example to describe a technical solution for achieving blind-spot-free coverage in a large-area monitoring area by increasing the number of similar light fixtures.

[0063] The floor area of ​​a living room is usually larger than that of a bedroom, and the radar beam coverage of a single ceiling light may not be able to cover the entire floor area. In this scenario, the system is configured with multiple ceiling lights that integrate radar sensing modules. The radar beam coverage areas of each ceiling light are adjacent to each other, and the entire floor area is covered through a zoned collaborative approach.

[0064] After extracting local area features, the data fusion processing unit 400 of each ceiling light uploads the feature parameters of the area and the coordinate information of the target object to a unified regional data aggregation node. The regional data aggregation node then performs global fusion of data from multiple ceiling light sub-areas to form a complete state perception of the entire living room monitoring area.

[0065] In the living room setting, table lamps and floor lamps also integrate radar sensing modules in the manner described in Example 1, participating in three-dimensional multi-angle fusion perception to further improve the detection reliability in large-area scenes.

[0066] When the target object 700 moves from one top light sub-region to an adjacent top light sub-region, the regional data aggregation node automatically completes the cross-region handover of target tracking, ensuring the continuity and accuracy of detection and avoiding detection interruption due to the target object being at the boundary of an adjacent sub-region.

[0067] This embodiment describes the adaptive parameter adjustment mechanism of the system in multi-user scenarios or special user group scenarios.

[0068] During operation, the system acquires real-time information on the number and motion patterns of target objects within the monitoring area. When the system detects multiple moving targets simultaneously within the monitoring area, the data fusion processing unit 400 performs independent target separation and tracking processing on the echo data of each target, and independently maintains its motion state characteristic parameters and abnormal state judgment process for each target to avoid false alarms caused by the aliasing of multiple target echoes.

[0069] At the user configuration level, the system allows users to set the mobility level of monitored individuals via an external interface. For elderly individuals with limited mobility or patients in the recovery phase, the system will preset a height threshold H. th The detection sensitivity for falls among this user group can be appropriately reduced, and the preset static judgment duration T2 can be appropriately extended to improve the false negative rate. For example, H can be... th The default height threshold has been reduced from 0.7 meters to 0.5 meters, and the time interval (T2) has been extended from the default 3 seconds to 5 seconds. For active child users, the system will preset a height threshold H. th The static judgment time T2 should be appropriately increased and shortened to avoid false alarms caused by children's normal jumping, squatting and running, and other vigorous movements; as an example, H can be... th Increasing it to 1.0 meter reduces T2 to 2 seconds.

[0070] In addition, when there are no moving targets in the monitoring area for a long time, each radar sensing module automatically switches to a low-power operating mode to reduce the overall power consumption of the system; when a moving target is detected to re-enter the monitoring area, the system automatically returns to the normal power consumption operating mode, achieving a balance between power management and real-time detection.

[0071] This embodiment describes the specific implementation differences of three types of radar sensing modules under different technical specifications.

[0072] In the implementation using frequency-modulated continuous wave radar (FMCW radar), each radar sensing module transmits a continuous wave signal whose frequency varies linearly with time and receives the echo. The target distance and velocity information are simultaneously extracted using the frequency difference (intermediate frequency signal) between the transmitted signal and the echo. FMCW radar features high range resolution and good velocity measurement accuracy, making it suitable for precise measurement of vertical displacement and velocity changes of a target object. In the fall detection application of this invention, it can accurately extract the characteristics of changes in center of gravity height.

[0073] In implementations employing ultra-wideband radar (UWB radar), each radar sensing module emits extremely narrow pulse signals, with a spectral coverage width typically exceeding 500MHz. UWB radar features extremely high range resolution and strong penetration capabilities, enabling it to effectively extract echo characteristics even when the target object is obscured by thin coverings (such as sheets or blankets), making it suitable for continuous detection of the status of targets on the bed in bedroom nighttime use scenarios.

[0074] In the implementation using millimeter-wave radar, each radar sensing module operates in the millimeter-wave frequency band (e.g., 60GHz or 77GHz), which features small antenna size and strong beam directionality, facilitating miniaturized integration within the lamp body. Simultaneously, millimeter-wave radar has high detection sensitivity for minute human movements (such as minute displacements caused by breathing and heartbeat), which can be used in this invention to further confirm whether the target object is still conscious after determining whether the target object has fallen.

[0075] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A radar-based luminaire control method, characterized by, Includes the following steps: S1. Radar signals are transmitted to the indoor monitoring area through the radar sensing modules integrated in the ceiling light installed on the indoor ceiling, the table lamp installed at the height of the tabletop, and the floor lamp installed near the ground, respectively, and radar echo signals reflected by the target objects in the monitoring area are received. The radar sensing module integrated in the ceiling lamp radiates radar signals to the monitoring area from a top-down angle, the radar sensing module integrated in the table lamp radiates radar signals to the monitoring area from a side-view or oblique-view angle, and the radar sensing module integrated in the floor lamp radiates radar signals to the monitoring area from a horizontal or upward-view angle. S2. Perform signal processing on the radar echo signals received by each of the radar sensing modules to extract feature parameters reflecting the motion state of the target object. The feature parameters include at least the height change feature of the target object's center of gravity relative to the ground. S3. The echo data from the radar sensing modules of the ceiling light, the table lamp and the floor light are sequentially processed by time alignment, coordinate system and weighted fusion to obtain the fusion feature parameters of the target object, and the target object is determined to be in an abnormal state based on the fusion feature parameters. S4. Based on the judgment result, output the corresponding lighting control command.

2. The radar-based lighting control method according to claim 1, characterized in that, The radar sensing module integrated in each of the aforementioned lamps adopts at least one of frequency-modulated continuous wave radar, ultra-wideband radar, or millimeter-wave radar.

3. The radar-based lighting control method according to claim 1, characterized in that, The feature parameters in step S2 further include at least one of the following: the target object's motion velocity characteristics, the target object's acceleration characteristics along the vertical direction, and the spatial distribution characteristics of the target object's radar reflection point cloud; wherein, The motion speed characteristics are obtained by analyzing the Doppler frequency shift of the radar echo signal; The height variation characteristics of the center of gravity relative to the ground are obtained by analyzing the echo energy distribution from different distance gates and combining the installation height and radiation angle parameters of each radar sensing module to calculate the height estimate of the target object's center of gravity and its height variation curve over time. The vertical acceleration characteristics are obtained by performing second-order difference calculations on the height change curve; The spatial distribution characteristics of the point cloud are obtained by jointly processing the radar echo signal in the range dimension, azimuth dimension and Doppler dimension to obtain the three-dimensional point cloud distribution of the target object, and at least one morphological parameter among the projected area of ​​the three-dimensional point cloud in the horizontal plane, the centroid height of the point cloud and the aspect ratio of the point cloud is extracted.

4. A radar-based lighting control method according to claim 1 or 3, characterized in that, Step S3, determining whether the target object is in an abnormal state, includes: Determine whether the change in the height of the target object's center of gravity within a preset time window T1 exceeds a preset height threshold, and whether the direction of the height change is from high to low; If so, it is further determined whether the target object remains stationary or in a low-speed motion state within a preset static judgment time T2 after the height change occurs. The low-speed motion state is when the absolute value of the target object's motion speed is less than a preset speed threshold. If the target object remains stationary or in a low-speed motion state for the preset static judgment time T2, it is determined that the target object has fallen and a fall alarm signal is generated. The preset time window T1 ranges from 0.3 seconds to 1.0 seconds, the preset height threshold ranges from 0.5 meters to 1.0 meters, the preset static judgment duration T2 ranges from 2 seconds to 5 seconds, and the preset speed threshold ranges from 0.1 meters / second to 0.5 meters / second.

5. The radar-based lighting control method according to claim 1, characterized in that, The weighted fusion process described in step S3 includes: Based on the contribution of the echo data of the ceiling light, the table lamp, and the ground light to the judgment of the abnormal state, a first weight, a second weight, and a third weight are assigned to the top-view echo data of the ceiling light, the side-view echo data of the table lamp, and the horizontal echo data of the ground light, respectively. The feature parameters corresponding to each echo data are weighted and summed to obtain the fused feature parameters. Wherein, the first weight is greater than the third weight, the third weight is greater than the second weight, and the sum of the first weight, the second weight and the third weight is 1; the value range of the first weight is 0.4 to 0.6, the value range of the second weight is 0.1 to 0.3, and the value range of the third weight is 0.2 to 0.

4.

6. The radar-based lighting control method according to claim 1, characterized in that, The coordinate system processing in step S3 includes: An absolute coordinate system is established with the center of the ground in the monitored area as the origin; Based on the installation height, radiation angle, and relative position parameters of each radar sensing module in the ceiling light, table lamp, and floor lamp, the rotation matrix and translation vector from the local coordinate system to the absolute coordinate system of each radar sensing module are constructed respectively. By using the rotation matrix and translation vector, the target position information described by each radar sensing module in its local coordinate system is transformed to the absolute coordinate system, so that the echo data from each path can consistently describe the same physical spatial position.

7. The radar-based lighting control method according to claim 1, characterized in that, The extraction of feature parameters in step S2 and the judgment of abnormal states in step S3 are both performed in the signal processing unit of the lamp. The method only outputs the state identification information representing the state of the target object to the external system, and does not transmit the original radar echo data to the external system.

8. The radar-based lighting control method according to claim 1, characterized in that, The lighting control command in step S4 includes at least one of the following: Control at least one of the overhead light, the table lamp, and the floor lamp to perform a flashing alarm action at a preset frequency; Control at least one of the ceiling light, the table lamp, and the floor lamp to turn on or adjust to a preset brightness level; Send an alarm notification containing the abnormal status information and the time of its occurrence to a preset terminal device; The system can coordinate with other smart devices within the monitoring area to execute preset response actions.

9. A radar-based lighting control method according to claim 4, characterized in that, Also includes: The system acquires the quantity and motion pattern information of target objects within the monitoring area in real time. When multiple moving targets exist simultaneously within the monitoring area, it performs target separation and tracking on the echo data of each moving target and independently maintains its characteristic parameters and abnormal status judgment for each moving target. Based on the pre-set mobility level of the monitored object, the preset height threshold and the preset static judgment duration T2 are dynamically adjusted; wherein, when the mobility level indicates inconvenience in movement, the preset height threshold is lowered and the preset static judgment duration T2 is extended; when the mobility level indicates active movement, the preset height threshold is raised and the preset static judgment duration T2 is shortened.

10. A radar-based lighting control system, characterized in that, include: The first radar light fixture is installed on the indoor ceiling and includes a ceiling light body and a first radar sensing module integrated in the ceiling light body. The first radar sensing module transmits radar signals to the monitoring area from a top-down angle and receives echo signals. The second radar lamp is installed at a lower height than the first radar lamp. It includes a lamp body and a second radar sensing module integrated into the lamp body. The second radar sensing module transmits radar signals to the monitoring area at a side-view or oblique-view angle and receives echo signals. The third radar light fixture is installed at a lower height than the second radar light fixture. It includes a ground light body and a third radar sensing module integrated in the ground light body. The third radar sensing module transmits radar signals to the monitoring area at a horizontal or upward angle and receives echo signals. The data fusion processing unit is communicatively connected to the first radar sensing module, the second radar sensing module, and the third radar sensing module, respectively. It is used to perform time-series alignment, coordinate system and weighted fusion processing on the echo data collected by each radar sensing module, extract the height change characteristics of the center of gravity of the target object relative to the ground in the monitoring area and obtain fusion feature parameters, and determine whether the target object is in an abnormal state based on the fusion feature parameters. The control unit, connected to the data fusion processing unit, is used to generate lighting control commands based on the judgment result of the data fusion processing unit, and send the lighting control commands to at least one of the first radar lights, the second radar lights, and the third radar lights, respectively, to control the corresponding lights to perform corresponding actions.

Images