A target presence detection method

Abstract

The application provides a target presence detection method, relates to the technical field of sensing detection, and is applied to a presence detection system comprising a first sensor and a radio frequency detection sensor. The method comprises the following steps: constructing an effective detection area; detecting environmental changes through the first sensor and outputting a trigger signal; when the trigger signal is valid, acquiring a signal feature set output by the radio frequency detection sensor, the signal feature set comprising motion-related signal features, static-related signal features and spatial position-related information; performing target activation determination based on the signal feature set to determine target response; determining a dominant spatial area according to the spatial position-related information and verifying whether the dominant spatial area is located in the effective detection area; and performing fusion determination based on at least the trigger signal, the target activation determination result and the area verification result to obtain a presence determination result and generate a control signal according to the presence determination result. Through the combination of multi-sensor fusion and spatial constraints, false triggering and missed detection can be reduced, and the accuracy and stability of presence detection can be improved.

Description

Technical Field

[0001] This application relates to the field of sensor detection technology, and in particular to a method for detecting the presence of a target. Background Technology

[0002] With the development of intelligent sensing and automatic control technologies, the technology of controlling devices in conjunction with each other based on the presence of a target has been widely applied in smart homes, office spaces, commercial venues, security monitoring, environmental control, and human-computer interaction. These systems typically use sensors to detect the entry, stay, or departure of a human or other target, and then control the opening, closing, or switching of operating modes of external devices accordingly.

[0003] In existing technologies, passive environmental change sensors, such as passive infrared sensors, typically identify human movement by detecting changes in infrared radiation. They are widely used due to their simple structure, low cost, and low power consumption. However, these sensors are primarily sensitive to changes in heat sources or motion, and are easily affected by factors such as hot air currents, thermal convection, temperature fluctuations, changes in sunlight, and pet activity, leading to false triggers. Furthermore, passive infrared sensors often struggle to reliably detect stationary targets or those exhibiting only minor movements, resulting in missed detections. On the other hand, active detection sensors, such as millimeter-wave radar sensors, can detect human movement, subtle movements, and even stationary objects by emitting detection signals and receiving echo information, exhibiting high sensitivity and strong presence detection capabilities. However, these sensors typically have a large default detection range, easily detecting people or objects outside the target area in practical deployments, leading to triggering in non-target areas. They are also sensitive to swaying curtains, airflow disturbances, equipment vibrations, multipath reflections, and false echoes in complex environments, still exhibiting a high probability of false detections.

[0004] Therefore, there is a need for an existence detection technology that can more reliably determine the validity of the target and verify regional constraints within the user's desired spatial range, thereby reducing false triggers and missed detections and improving the applicability and stability of the system in different application environments. Summary of the Invention

[0005] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a target presence detection method that can reduce the probability of false triggering and false judgment, and improve the accuracy, reliability and environmental adaptability of presence detection.

[0006] The target presence detection method according to an embodiment of this application is applied to a presence detection system including a first sensor and a radio frequency detection sensor. The target presence detection method of the embodiment includes: Based on the user-defined detection range parameters, an effective detection area is constructed; The first sensor detects the environment and outputs a valid trigger signal when the triggering conditions are met. When the trigger signal is valid, the signal feature set output by the radio frequency detection sensor in the current detection period is obtained, wherein the signal feature set includes at least motion-related signal features, static-related signal features, and spatial position-related information. Target activation is determined based on the signal feature set to ascertain whether a target response exists within the current detection period. Based on the spatial location-related information, the dominant spatial region corresponding to the target response is determined, and it is determined whether the dominant spatial region is located within the effective detection region to obtain the region verification result; A fusion determination is performed based at least on the trigger signal, the result of the target activation determination, and the region verification result to obtain an existence determination result; Based on the existence determination result, a control signal is generated to control the external output device.

[0007] The target presence detection method according to the embodiments of this application has at least the following beneficial effects: First, an effective detection area is constructed based on the detection range parameters set by the user, thereby limiting the actual response space of the system to the range desired by the user. It no longer relies solely on the large physical detection range of the sensor itself for rough judgment, thus better adapting to application scenarios where the target area is clearly defined and the detection space is limited. Furthermore, the target presence detection method of the embodiments performs preliminary detection of the environment using a first sensor, and then acquires the signal feature set output by the radio frequency detection sensor within the current detection cycle when the trigger condition is met. This allows the system to utilize the first sensor for pre-stage triggering and screening, and also to utilize the motion-related signal features, static-related signal features, and spatial position-related information provided by the radio frequency detection sensor for more in-depth detection and analysis of the target. This helps reduce invalid detection and improves the targeting and stability of the detection process. Meanwhile, the target presence detection method in this embodiment does not rely solely on the strength of a single signal to determine presence. Instead, it first performs target activation determination based on the signal feature set to ascertain the presence of a target response. Then, it determines the dominant spatial region corresponding to the target response based on spatial location-related information and determines whether this dominant spatial region is located within the effective detection area. Finally, it combines at least the trigger signal, the target activation determination result, and the region verification result for a fusion determination to obtain the presence determination result. This multi-source information fusion and spatial constraint-based determination method effectively distinguishes between real human targets and targets outside the detection area, environmental noise, multipath reflections, pet activities, curtain swaying, or airflow disturbances, reducing the probability of false triggers and false judgments, and improving the accuracy, reliability, and environmental adaptability of presence detection. Furthermore, generating control signals based on the presence determination result to control external output devices allows the opening and closing of external output devices to better meet the actual user needs, thereby improving the overall control effect and user experience.

[0008] According to some embodiments of this application, the step of determining target activation based on the signal feature set to ascertain whether a target response exists within the current detection period includes: If the motion-related signal feature is not less than a preset motion determination threshold, or if the static-related signal feature is not less than a preset static determination threshold, then the target response is determined to exist within the current detection period.

[0009] According to some embodiments of this application, the output modes of the radio frequency detection sensor include a discrete distance partition mode and a global distance mode; In the discrete distance partitioning mode, the effective detection area is represented as a set of distance partitions whose numbers do not exceed the effective judgment threshold; the spatial location-related information is represented as several candidate distance partitions and their corresponding signal features; In the global distance mode, the effective detection area is characterized as a continuous spatial range that does not exceed the maximum effective detection distance; the spatial location-related information is represented as the distance to a single target or the area identifier mapped thereto.

[0010] According to some embodiments of this application, the step of determining the dominant spatial region corresponding to the target response based on the spatial location-related information, and determining whether the dominant spatial region is located within the effective detection region to obtain a region verification result includes: In the discrete distance partitioning mode, the representative energy corresponding to each candidate distance partition is determined, and the candidate distance partition with the largest representative energy is selected as the dominant spatial region; wherein, the representative energy is determined by the larger value between the motion-related signal feature and the smoothed static-related signal feature of the corresponding candidate distance partition; In the global distance mode, the spatial region mapped by the single target distance is directly used as the dominant spatial region.

[0011] According to some embodiments of this application, the step of fusing the determination based at least on the trigger signal, the result of the target activation determination, and the region verification result to obtain the existence determination result includes: The presence determination result is valid only if the trigger signal is valid, the target response is determined to exist, and the dominant spatial region is located within the valid detection region.

[0012] According to some embodiments of this application, the step of fusing the determination based at least on the trigger signal, the result of the target activation determination, and the region verification result to obtain the existence determination result includes: The basic existence state is determined to be valid only if the trigger signal is valid, the target response is determined to exist, and the dominant spatial region is located within the valid detection region; The existence determination result is determined based on the preset existence preservation mode and the basic existence state; wherein, the existence preservation mode includes strict mode and extended preservation mode; In the strict mode, the existence determination result is maintained as valid only if the basic existence state is valid; In the extended maintenance mode, if the basic existence state is valid, or if the existence determination result of the historical detection period is valid and a target response exists in the current detection period, the existence determination result is maintained as valid.

[0013] According to some embodiments of this application, the generation of a control signal based on the existence determination result includes a time continuity confirmation mechanism: The presence determination results within a historical detection period of a continuously preset number of data points are statistically analyzed. If the existence determination result is valid within the consecutive preset number of historical detection cycles, the control signal with the state "on" is generated.

[0014] According to some embodiments of this application, generating a control signal based on the existence determination result further includes: When the existence determination result changes from valid to invalid, start the shutdown timer; If the countdown value of the timer does not reach the delay shutdown threshold, the control signal remains in the on state. When the countdown value of the shutdown timer reaches the delay shutdown threshold, a control signal is generated indicating that the status is off.

[0015] According to some embodiments of this application, the delay shutdown threshold is obtained in the following manner: Based on the target activation determination results, the region verification results, the motion-related signal features, and the static-related signal features within the historical detection period, the target existence confidence is calculated using a weighted average. The delay-off threshold is positively adjusted based on the confidence level of the target's existence.

[0016] According to some embodiments of this application, the step of determining target activation based on the signal feature set to determine whether a target response exists within the current detection period further includes: The static correlation signal features are subjected to time smoothing to obtain smoothed static correlation signal features. Attached Figure Description

[0017] The present application will be further described below with reference to the accompanying drawings and embodiments, wherein: Figure 1 This is a schematic diagram of the overall process of a target presence detection method provided in an embodiment; Figure 2 This is a schematic diagram illustrating the process of determining the dominant spatial region based on spatial location information in the embodiment. Figure 3 This is a schematic diagram of the process for determining the existence of a determination result based on fusion judgment in the embodiment; Figure 4 This is a schematic diagram of the process for generating an activation control signal based on a continuous detection cycle in the embodiment. Figure 5 This is a schematic diagram of the process for generating control signals based on a time-delay shutdown mechanism in the embodiment. Detailed Implementation

[0018] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0019] In the description of this application, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0020] In the description of this application, "several" means one or more, "multiple" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0021] In the description of this application, unless otherwise expressly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0022] In the description of this application, the terms "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0023] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0024] The target presence detection method of this embodiment is applied to the presence detection system of this embodiment. The presence detection system of this embodiment includes at least a first sensor, a radio frequency (RF) detection sensor, a control unit, and an external output device. The first sensor is used to detect environmental changes and output trigger information; for example, it may be a passive infrared (PIR) sensor. The RF detection sensor is used to output a detection signal related to the presence of the target; for example, it may be a millimeter-wave detection sensor, or other RF detectors capable of actively emitting detection signals and acquiring target response information. The control unit is electrically connected to the first sensor, the RF detection sensor, and the external output device, respectively, and is used to execute presence detection logic and output control signals. The external output device may be, for example, a lighting device, a display device, an alarm device, or other controlled load; for example, in a more specific application scenario, it may be an indoor lighting load, such as a cabinet light, a work light, or a table lamp.

[0025] The target presence detection method in this embodiment can be executed cyclically according to a detection cycle. In each detection cycle, the control unit determines whether a valid target exists in the current space based on the user-defined detection range parameters, the output of the first sensor, and the output of the radio frequency detection sensor, and controls the working state of the external output device according to the determination result.

[0026] like Figure 1 As shown, the target presence detection method in the embodiment includes, but is not limited to, steps S100 to S700: S100: Construct an effective detection area based on user-defined detection range parameters; S200: The environment is detected by the first sensor, and a trigger signal with a valid state is output when the trigger condition is met; S300. When the trigger signal is valid, acquire the signal feature set output by the radio frequency detection sensor in the current detection cycle, wherein the signal feature set includes at least motion-related signal features, static-related signal features, and spatial position-related information. S400: Target activation determination is performed based on the signal feature set to determine whether a target response exists within the current detection period; S500: Determine the dominant spatial region corresponding to the target response based on spatial location-related information, and determine whether the dominant spatial region is located within the effective detection area to obtain the region verification result; S600: At least based on the trigger signal, the result of the target activation determination, and the regional verification result, a fusion determination is performed to obtain the existence determination result; S700: Based on the existence determination result, a control signal is generated to control the external output device.

[0027] In step S100, the control unit first reads the detection range parameter set by the user and constructs an effective detection area accordingly. The detection range parameter can be input by the user through a DIP switch, button, knob, communication interface, host computer software, mobile terminal application, or other configuration interface. This parameter can be expressed as the maximum effective detection distance, or as an area number corresponding to the distance, gear information, or other parameters that can characterize the target spatial range.

[0028] The control unit logically establishes an effective detection area based on the detection range parameters to define the target's activity space as recognized by the system. It should be understood that an effective detection area does not necessarily require physically altering the sensor's inherent detection range. Rather, it preferably involves the control unit filtering and verifying the target's spatial location during subsequent decision-making, thereby achieving logical constraints on the detection space. Thus, even if the radio frequency detection sensor itself has a large detection capability, the system can still respond only to targets within the nearby or designated areas that the user wishes to focus on.

[0029] In some implementations, when the RF detection sensor outputs discrete range partition information, the effective detection area can be represented as a set of several effective range partitions; when the RF detection sensor outputs continuous target range information, the effective detection area can be represented as a continuous spatial range not exceeding the maximum effective detection distance. Both expressions are logically equivalent and are used for subsequent area verification processing.

[0030] In step S200, the first sensor continuously monitors the environment to acquire information about environmental changes. In some embodiments, the first sensor is a PIR sensor, which detects changes in thermal radiation or infrared radiation; when an environmental change that meets preset conditions is detected, the output is a valid trigger signal. The trigger conditions can be preset according to the application scenario, for example, corresponding to human movement, a preset level of heat source change, or other conditions suitable for triggering subsequent detection processes. In this embodiment, the first sensor is preferably used as a front-end trigger layer to continuously monitor environmental changes with low power consumption and to initiate or enhance subsequent radio frequency detection processing when necessary. This reduces invalid detections and improves overall power consumption performance; it also helps to reduce the probability of misjudgments caused by weak noise or occasional disturbances in a stable environment.

[0031] In step S300, when the trigger signal is valid, the control unit acquires the set of signal features output by the radio frequency detection sensor during the current detection cycle. The set of signal features includes at least motion-related signal features, static-related signal features, and spatial position-related information.

[0032] Among them, motion-related signal features are used to characterize the dynamic behavior of the target, such as signal changes caused by walking, waving, and body displacement; static-related signal features are used to characterize signal changes in the target's static or slightly moving state, such as lingering, breathing, and slight shaking; spatial location-related information is used to characterize the target's spatial location relative to the sensor, which can be represented as one or more candidate distance partitions, target distance values, area identifiers, or location identifiers obtained by further mapping from distance.

[0033] In some implementations, after acquiring the signal feature set, the control unit may also preprocess the original detection data or feature data, such as filtering, normalization, background suppression, outlier suppression, or other processing that helps improve detection stability, but this application does not limit this.

[0034] In step S400, the control unit performs target activation determination based on the signal feature set to determine whether a target response exists within the current detection cycle. Target response characterizes a valid response caused by the presence, proximity, activity, stillness, or slight movement of a human body, rather than an invalid response caused by environmental noise, airflow disturbance, thermal fluctuations, reflection interference, or the movement of non-target objects. In this step, the control unit can make a judgment by comprehensively considering motion-related signal features and static-related signal features. For example, when motion-related signal features meet the corresponding determination conditions, a dynamic target response can be considered to exist; when static-related signal features meet the corresponding determination conditions, a static or slightly moving target response can be considered to exist. The control unit can also combine historical cycle data, background noise level, sensor sensitivity settings, or other auxiliary information for comprehensive judgment to improve the ability to identify real human targets. It should be noted that the core of this step is to confirm "whether a valid response that can be considered a target is detected within the current detection cycle," and its output can be a binary activation result or a determination result containing activation state and related intensity information, for use in subsequent steps.

[0035] In step S500, the control unit determines the dominant spatial region corresponding to the current target response based on spatial location-related information, and further determines whether the dominant spatial region is located within the effective detection area to obtain the region verification result.

[0036] The dominant spatial region refers to the area that best represents the main spatial location of the target within the current detection cycle. When the radio frequency detection sensor outputs multiple candidate locations, the control unit can select the most representative one as the dominant spatial region, such as the candidate region with the strongest response, the most significant energy, the most stable position, or that meets a preset priority rule. When the radio frequency detection sensor only outputs a single target distance or a single area identifier, the area corresponding to that location can be directly used as the dominant spatial region.

[0037] After determining the dominant spatial region, the control unit compares it with the effective detection region established in step S100 to determine whether the target is within the user-defined effective detection range. If the dominant spatial region is within the effective detection range, the region verification result is valid; if the dominant spatial region is outside the effective detection range, the region verification result is invalid. Through step S500, the system not only determines "whether a target is detected," but also further determines "whether the target is within the spatial range that the user is truly concerned with," thereby achieving logical compression of the detection space and suppression of false triggers.

[0038] In step S600, the control unit performs a fusion judgment based at least on the trigger signal, the target activation determination result, and the region verification result to obtain an existence determination result. In one embodiment, the control unit can use a rule-based fusion method to comprehensively process the above information. For example, a valid target is determined to exist in the current detection period only when the first sensor outputs a valid trigger signal, the radio frequency detection sensor detects a target response, and the dominant spatial region of the target is located within the valid detection region. In addition, the control unit can also use weighted fusion, hierarchical decision-making, state machine determination, or other suitable fusion methods according to application requirements, as long as it can simultaneously utilize the aforementioned multiple information dimensions to comprehensively determine the existence of the target.

[0039] Through the S600 fusion step, the target presence detection method of the embodiment combines the environmental change information provided by the first sensor, the presence feature information provided by the radio frequency detection sensor, and the spatial constraint information provided by the effective detection area, so that the final presence determination result no longer depends on a single sensor or a single threshold, but is based on multi-dimensional verification, which can more effectively distinguish between real targets and non-target interference.

[0040] In step S700, the control unit generates a control signal based on the presence determination result obtained in step S600, and outputs the control signal to an external output device to control its operating state. Taking a lighting application as an example, when the presence determination result is valid, the control unit outputs an on control signal to drive the lighting load to light up or maintain its illumination; when the presence determination result is invalid, the control unit outputs an off control signal, or causes the lighting load to enter a standby state. It should be understood that the specific form of the control signal can be a high or low level signal, a PWM signal, a serial control command, a bus control command, or other control forms adapted to a specific output drive circuit, and this application does not limit this.

[0041] Through steps S100 to S700, the target presence detection method of this embodiment achieves the following: first, a first sensor provides a pre-stage trigger; then, a radio frequency detection sensor provides target presence characteristics; and finally, spatial verification is performed by combining the target with a user-defined effective detection area. After fusion and judgment, a control result is output. Compared to schemes relying solely on a single PIR sensor or a single millimeter-wave sensor, this target presence detection method can more effectively suppress false triggering at long distances, triggering in non-target areas, and erroneous responses caused by environmental interference while maintaining detection sensitivity. It is suitable for scenarios with high requirements for detection range and response accuracy.

[0042] In some embodiments, the aforementioned radio frequency (RF) detection sensor may employ different data output architectures. It should be noted that the target presence detection method in the embodiments does not depend on the specific distance resolution, partitioning method, or internal implementation structure of the RF detection sensor. Considering the differences in signal processing links, target extraction methods, and data interface forms among different models of millimeter-wave sensors, radar modules, or other active RF detectors, the output modes of the RF detection sensor are abstracted and unified, supporting at least the following two output modes: Discrete range partitioning mode. In discrete range partitioning mode, the RF detection sensor divides its detection space along the range direction into multiple discrete range partitions (also called range gates, range intervals, or detection segments), and outputs the detection results for each partition. The range partitions can be numbered in ascending order, for example, numbered as follows: i =0,1,2,…,N, where smaller numbers generally indicate spatial regions closer to the sensor, and larger numbers indicate spatial regions farther from the sensor. In this mode, each distance partition can correspond to a preset distance interval. If the sensor has a basic distance resolution Δd, then the... i Each distance partition can correspond to a distance of approximately [distance from the sensor]. i The position of Δd, or a distance segment represented by that position. It should be understood that the correspondence between the partition and the actual distance can be linear or non-linear; it can be directly given by the sensor's internal algorithm, or it can be mapped by the control unit based on the sensor output using a lookup table; this application does not impose any restrictions on this. In this mode, the spatial location-related information output by the RF detection sensor is represented as several candidate distance partitions and their corresponding signal characteristics. In other words, within a detection cycle, the system does not necessarily obtain only a single spatial position, but may simultaneously obtain multiple presence response candidates, each containing its own distance partition and the motion-related signal characteristics and static-related signal characteristics associated with that partition. For example, in the first... k Within each detection period, a candidate set in the following form can be obtained:

[0043] in, i Indicates the candidate distance partition number. E move,i,k Indicates the first k Within the first detection cycle i Motion-related signal features corresponding to each distance partition E static,i,k Indicates the first k Within the first detection cycle i The static correlation signal characteristics corresponding to each distance partition.

[0044] In the discrete distance partitioning mode, the effective detection region is preferably represented as a set of distance partitions whose numbers do not exceed the effective judgment threshold. That is, the user-defined maximum effective detection distance is first converted into the corresponding maximum effective partition number, and then the logically effective region is constructed using this number as the boundary. For example, if the user-defined maximum effective detection distance is... d user The system converts the sensor's distance resolution Δd or the partition mapping relationship into an effective judgment threshold. i user In one implementation, the threshold can be obtained by calculation, rounding, or table lookup, for example:

[0045] Alternatively, the control unit can directly obtain the valid partition number corresponding to the user-defined distance based on a preset mapping table. In this case, the valid detection area can be represented as:

[0046] This means that any distance partition with a number less than or equal to the effective judgment threshold is considered a valid detection area; any distance partition with a number greater than the effective judgment threshold is considered space outside the valid detection area.

[0047] Global distance mode. In global distance mode, the RF detection sensor does not output the detection results of multiple discrete distance zones, but instead outputs the overall signal characteristics and single target distance information for the current detection cycle, or outputs a region identifier that can directly characterize the current main target position. This type of mode is common in some highly integrated millimeter-wave presence detection modules, which have already completed target extraction or distance estimation internally. The external interface no longer exposes the zone-by-zone energy distribution, but directly provides data such as global motion energy, global static energy, and target distance. In this mode, spatial location-related information is preferably represented as a single target distance or its mapped region identifier. Wherein: the single target distance can be understood as the distance value that best represents the target's spatial position within the current detection cycle; the mapped region identifier can be understood as the logical region number obtained by the control unit based on this distance value. In one embodiment, the RF detection sensor can provide the target distance related to motion and the target distance related to statics respectively, and the control unit then selects the more representative distance as the target distance for the current detection cycle based on the corresponding signal characteristic strength. Alternatively, the sensor can directly output a unique target distance internally, which is not limited in this application.

[0048] In global distance mode, the effective detection area can be characterized as a continuous spatial range that does not exceed the maximum effective detection distance. If the user-defined maximum effective detection distance is... d user The effective detection area can then be represented as:

[0049] In other words, as long as the distance to a single detected target is not greater than the maximum effective detection distance, the target is in the effective detection area in space; conversely, if the distance to the target is greater than the maximum effective detection distance, the target is outside the effective detection area.

[0050] To ensure consistency with the discrete distance partitioning pattern within the system, in some implementations, the control unit can also convert the distance to a single target into a region identifier using a mapping function, for example:

[0051] in, d k For the first k Distance of a single target within a detection cycle R k The region identifier obtained by mapping, f (*) represents the mapping function. The mapping function can be linear partitioning, piecewise functions, lookup tables, or sensor-built-in region mapping methods. For example, in one implementation, linear quantization can be performed at a fixed distance resolution, so that the distance of a single target in the global distance mode can logically be converted into a region identifier consistent with the partition mode, so that subsequent processing modules can use a unified region verification process.

[0052] In some embodiments, the aforementioned step S400, "determining target activation based on signal feature set to determine whether there is a target response in the current detection period," may further include, but is not limited to, step S410: S410. If the motion-related signal characteristics are not less than the preset motion determination threshold, or the static-related signal characteristics are not less than the preset static determination threshold, determine that there is a target response within the current detection period.

[0053] In step S410, after acquiring the signal feature set output by the radio frequency detection sensor within the current detection cycle, the control unit analyzes the motion-related signal features and static-related signal features, and determines whether a target response exists within the current detection cycle based on a preset judgment threshold. The target response preferably represents a valid signal change caused by a human body entering, moving, staying, remaining, or micro-moving, rather than invalid fluctuations caused by environmental disturbances, non-target object activity, or sensor noise.

[0054] The purpose of step S410 is to further answer the question of "whether the currently detected signal is sufficient to characterize the existence of a real target," given that the radio frequency detection sensor has already output relevant detection features. In other words, it is not sufficient for the sensor to determine the existence of a target simply by detecting any energy change; rather, a target activation determination mechanism is needed to screen the detection results for validity.

[0055] In some implementations, the signal feature set includes at least: motion-related signal features, used to characterize the signal intensity corresponding to the target's dynamic behavior, such as changes caused by human walking, approaching, swinging arms, turning around, etc.; static-related signal features, used to characterize the signal intensity corresponding to the target's static or slightly moving state, such as changes caused by human standing, pausing, breathing, slight posture changes, etc.; and spatial location-related information, used to characterize the target position or candidate position corresponding to the current response.

[0056] It should be understood that motion-related signal characteristics and static-related signal characteristics can be provided by different types of radio frequency detection sensors. Their specific forms can be energy values, amplitude values, intensity values, power values, spectral characteristic values, statistical characteristic values, or characterization quantities obtained by internal algorithm processing. As long as they can respectively reflect the existence characteristics of dynamic targets and static / micro-moving targets, this application does not limit them.

[0057] In some implementations, the control unit compares motion-related signal features with preset motion determination thresholds and static correlation signal features with preset static determination thresholds. When the motion-related signal features are not less than the preset motion determination thresholds, or the static correlation signal features are not less than the preset static determination thresholds, a target response is determined to exist within the current detection period. The above determination rule can be expressed as:

[0058] in: A k Indicates the first k Target activation determination results within each detection cycle; A k =1 indicates that a target response exists within the current detection period; A k =0 indicates that no valid target response was detected in the current detection period; E move,k Indicates the first k Motion-related signal characteristics within each detection cycle; E static,k Indicates the first k Static correlation signal characteristics within a detection period; T move Indicates the motion determination threshold; T staticThis indicates the static judgment threshold.

[0059] Understandably, the purpose of setting motion and static detection thresholds is to distinguish between effective responses caused by the human body and non-target interference responses. Specifically, in practical applications, although radio frequency detection sensors have high sensitivity, they may still detect the following interference: Slight dynamic disturbances such as swaying curtains or shaking door curtains; Background changes caused by air conditioning air, hot air flow, steam flow, etc. Pet activity, object swaying, or small hanging objects moving; Multipath reflections or false echoes caused by metal utensils, countertop edges, cabinet structures, etc. Electromagnetic noise, transient fluctuations, or occasional abnormal signals.

[0060] Without threshold constraints on motion-related and static-related signal features, the system may easily mistake the aforementioned interference for the presence of a target. By using preset thresholds for filtering, only responses meeting certain amplitude or intensity conditions are allowed to proceed to the next stage, thereby improving the system's reliability in identifying real human targets. It should be understood that these thresholds are not merely for simply suppressing low-amplitude noise; more importantly, they establish the "minimum feature conditions required for a human target response." That is, the system only determines the presence of a target response when the detected features are sufficient to characterize human behavior or subtle movements.

[0061] In one embodiment, to improve the stability of static target detection, the control unit may further perform time smoothing on the static correlation signal features before comparing them with the static determination threshold, obtaining smoothed static correlation signal features, and then performing the threshold comparison in step S410. For example, exponential smoothing can be used for this process.

[0062] in: Indicates the first k Smoothed static correlation signal characteristics for each detection cycle; Indicates the first k The original static correlation signal characteristics of each detection period; α is a smoothing coefficient, satisfying 0 < α < 1. In some implementations, the smoothing coefficient α can be taken as 0.1. 0.6. A larger α results in a faster response, while a smaller α provides a stronger smoothing effect and higher stability. In practical applications, the specific value of the smoothing coefficient α can be set and adjusted based on the background noise fluctuation of the system deployment environment and prior experimental test data. For example, in scenarios with drastic changes in the environmental background signal and strong multipath interference, the smoothing filtering effect can be enhanced by increasing the proportion of historical data (i.e., correspondingly decreasing the α value); while in scenarios with relatively quiet environments where the system is required to respond quickly to minute target movements, the α value can be appropriately increased based on the expected response time in actual testing. The smoothing coefficient α can be adjusted manually on-site or adaptively determined by the control unit based on the noise variance during the initial environmental scan period.

[0063] The reason for smoothing the static correlation signal characteristics over time is that static correlation signals usually correspond to micro-motions or weak responses, with small amplitudes and are more susceptible to environmental noise, reflection fluctuations, and transient disturbances. Therefore, time smoothing can reduce the impact of short-term jumps on the judgment results. In contrast, motion correlation signals tend to change rapidly and have larger amplitudes, emphasizing fast response. Therefore, in the preferred embodiment, motion correlation signals may not be smoothed.

[0064] Understandably, in some implementations, target activation determination can be adapted to different radio frequency detection sensor output modes.

[0065] When the radio frequency detection sensor outputs in discrete range partition mode, multiple candidate range partitions and their corresponding motion-related and static-related signal features can be obtained within the current detection cycle. At this time, the control unit can perform target activation determination for each candidate range partition separately. For example, for the first... i The distance partition, at the nth distance partition, k Within each detection period, the target activation result can be expressed as:

[0066] in: A i,k Indicates the first k Within the first detection cycle i Activation results of each distance partition; E move,i,k Indicates the first i Motion-related signal characteristics of each distance partition; E static,i,k Indicates the first i Static correlation signal characteristics of each distance partition; T move,i , T static,iThese represent the motion detection threshold and static detection threshold corresponding to the distance partition, respectively. In one embodiment, all distance partitions can share the same set of thresholds; in another embodiment, thresholds can be set separately for different distance partitions. For example, if the echo attenuation of a more distant distance partition is greater or its noise characteristics are different, a different threshold parameter can be used compared to that of a closer distance partition to improve overall detection performance. Furthermore, the control unit can construct an activation region set from all distance partitions that meet the activation conditions for subsequent steps to select the dominant spatial region and perform region verification.

[0067] When the radio frequency detection sensor outputs in global distance mode, the control unit can directly perform target activation determination based on the overall motion-related signal characteristics and overall static-related signal characteristics within the current detection cycle. That is, it determines whether a target response exists within the current detection cycle according to the aforementioned unified formula. In this case, the target activation determination result is usually a single result, directly representing whether a valid target response exists throughout the current detection cycle. If a target response is determined to exist, the corresponding single target distance or its mapped region can be used as input for subsequent spatial verification.

[0068] like Figure 2 As shown, in some embodiments, the aforementioned step S500, "determining the dominant spatial region corresponding to the target response based on spatial location-related information, and determining whether the dominant spatial region is located within the effective detection area to obtain the region verification result," may further include, but is not limited to, steps S510 to S520: S510. In the discrete distance partitioning mode, determine the representative energy corresponding to each candidate distance partition, and select the candidate distance partition with the largest representative energy as the dominant spatial region; wherein, the representative energy is determined by the larger value between the motion-related signal feature and the smoothed static-related signal feature of the corresponding candidate distance partition. S520. In global distance mode, the spatial region mapped by the distance to a single target is directly used as the dominant spatial region.

[0069] Understandably, within the current detection cycle, the control unit first determines the dominant spatial region corresponding to the current target response based on the spatial location information output by the radio frequency detection sensor. Subsequently, it compares the dominant spatial region with the effective detection region constructed in step S100 to output the region verification result. The dominant spatial region uniquely characterizes the most representative target spatial location within the current detection cycle, thus providing a unified and clear comparison object for subsequent spatial validity verification. The reason for determining the dominant spatial region is that in actual detection, the radio frequency detection sensor may simultaneously output multiple candidate location responses, or although it may only output a single target distance, this distance still needs to be categorized into a processable region object within the system. Without this step, the system may not be able to uniquely determine the spatial attribution of the current target during subsequent judgments, thereby affecting the comparison results of the effective detection regions and the stability of the overall existence determination.

[0070] In step S510, when the radio frequency detection sensor adopts a discrete range partitioning mode, the system can obtain multiple candidate range partitions and their corresponding signal characteristics within the current detection period. These multiple candidate range partitions typically originate from the aforementioned target activation determination step; that is, during the target activation determination process, the range partitions determined to have target responses form a candidate set.

[0071] Let the first k The candidate distance partition set within each detection period is:

[0072] Where: i represents the candidate distance partition number; A i,k Indicates the first k Within the first detection cycle i Target activation determination results for each distance partition; A i,k =1 indicates that the distance partition is determined to have a target response. For the set For each candidate distance partition, the control unit determines the representative energy of that candidate distance partition and selects the candidate distance partition with the largest representative energy as the dominant spatial region of the current detection cycle.

[0073] In some implementations, the first i The candidate distance partitions in the th are at the . k Representative energy within each detection cycle The representative energy is determined by the larger value between the motion-related signal features and the smoothed static-related signal features of the distance partition. Using the "take the larger value" method to determine the representative energy has good engineering adaptability. This is because: during some detection cycles, the dynamic behavior of the human body is more obvious, and the motion-related signal features can more accurately represent the target; while during other detection cycles, the human body may be in a static or slightly moving state, and the smoothed static-related signal features are more representative of the actual state. By taking the larger value of the two within each candidate distance partition, the representative energy of that candidate distance partition is always dominated by the more representative signal, thus avoiding the weakening of the target representation ability due to simple averaging or fixed dependence on a single feature.

[0074] After obtaining the representative energy of each candidate distance partition, the control unit selects from the candidate set. The candidate distance partition with the highest energy is selected as the dominant spatial region for the current detection cycle. This can be represented as:

[0075] in: R dom,k Indicates the first k The dominant spatial region within each detection cycle; Indicates the first i The system selects the distance partition with the strongest signal strength from among the multiple activated candidate distance partitions as the region that best represents the target's main spatial location within the current detection period.

[0076] In step S520, when the radio frequency detection sensor adopts global range mode, the sensor typically does not output the zone-by-zone energy information of multiple candidate range zones. Instead, it outputs the overall motion-related signal characteristics, the overall static-related signal characteristics, and the single target range or its mapped region identifier. In this mode, since only single target range information exists in the current detection cycle, there is no need to compare multiple candidate range zones. The control unit can directly use the spatial region mapped by the single target range as the dominant spatial region. That is:

[0077] in: R k Indicates the first k Spatial region identifiers obtained by mapping the distance of a single target within a detection cycle; R dom,k Indicates the first k The dominant spatial region within each detection cycle.

[0078] Understandably, for the source of a single target distance, in one embodiment, the radio frequency detection sensor can directly output the single target distance. d k The control unit determines the corresponding area based on this. In another embodiment, the radio frequency detection sensor can output the target distance related to motion. d move,k and target distance related to statics d static,k At this point, the control unit can select a more representative distance as the single target distance for the current detection cycle based on the magnitude relationship between the overall motion-related signal characteristics and the overall static-related signal characteristics. In one embodiment, the control unit can select the single target distance... d k Through mapping function f (*) Mapped to spatial region identifiers: The mapping function can be linear partitioning, piecewise mapping, lookup table mapping, or a sensor-built-in mapping method. For example, when using equal-width partitioning, the following can be used: , where Δd represents the logical distance resolution.

[0079] In some embodiments, the aforementioned step S600, "performing a fusion determination based at least on the trigger signal, the result of the target activation determination, and the region verification result to obtain an existence determination result," further includes, but is not limited to, step S611: S611. The determination result is valid only if the trigger signal is valid, the target response is determined to exist, and the dominant spatial region is located within the valid detection region.

[0080] Step S611 can serve as a basic valid existence determination mechanism.

[0081] In step S611, the control unit acquires the following within the current detection cycle: the trigger signal output by the first sensor; the target activation determination result output by the target activation determination step; and the region verification result output by the spatial region verification step. Based on the above three results, the control unit performs a basic fusion determination to output the existence determination result within the current detection cycle.

[0082] Let the trigger signal of the first sensor be denoted as T k It is used to characterize whether the first sensor detects an environmental change that meets a preset trigger condition within the current detection period. For example, in the case where the first sensor is a PIR sensor: T k =1 indicates that a valid infrared change or human movement trigger was detected within the current detection period; T k=0 indicates that no valid trigger was detected within the current detection cycle. The first sensor trigger signal can be used as the output of the front-end trigger layer in the overall judgment logic to characterize "whether there are environmental changes worthy of further attention".

[0083] Let the target activation determination result be denoted as A k It is used to characterize whether the motion-dependent signal characteristics and static-dependent signal characteristics output by the radio frequency detection sensor within the current detection period meet the effective target response conditions. Wherein: A k =1 indicates that a target response exists within the current detection period; A k =0 indicates that there is no valid target response in the current detection period. This result is usually obtained from the aforementioned step S400, and it essentially answers the question of "whether a valid radio frequency response that can be considered to be caused by the human body has been detected".

[0084] Let the regional verification result be denoted as Z k It is used to characterize whether the dominant spatial region determined within the current detection period is located within the user-defined valid detection area. Wherein: Z k =1 indicates that the dominant spatial region is located within the effective detection area; Z k =0 indicates that the dominant spatial region is outside the effective detection region. This result can be obtained from the aforementioned step S500, which essentially answers the question of "whether the currently detected target is within the spatial range that the user expects the system to respond to".

[0085] For example, the control unit performs basic fusion determination according to the following rules: the existence determination result is valid only if the trigger signal is valid, a target response is determined to exist, and the dominant spatial region is located within the valid detection region. The existence determination result for the current detection cycle can be recorded as... P k Then its judgment logic can be expressed as:

[0086] Or, equivalently, it can be represented as:

[0087] in: P k =1 indicates the first k The presence determination result within each detection cycle is valid; P k =0 indicates the first k The presence determination result within a detection cycle is invalid.

[0088] Understandably, in the fusion determination step S611, the first sensor trigger signal is used to provide initial trigger constraints. Its function is not to independently confirm the presence of a human body, but rather to provide a pre-gating condition for subsequent presence determinations. For example, in kitchen, workbench, or indoor lighting applications, PIR sensors can continuously monitor environmental changes with low power consumption. Only when a human body approaches, thermal radiation changes, or other preset trigger phenomena are detected is the current detection cycle considered valuable for further confirmation. This reduces over-response to weak radio frequency fluctuations when the environment is stable over a long period. The target activation determination result is used to distinguish between valid target responses and interference responses. Even if the first sensor triggers, it is still necessary to confirm whether the response detected by the radio frequency detection sensor is sufficient to characterize the presence of a human body. For example, some environmental changes may cause false PIR triggers, but do not correspond to a real human body; or some radio frequency responses may exist, but are insufficient to exceed the target activation determination threshold and should not be considered valid targets. The area verification result is used to constrain the spatial location of the target. Even if a valid target response is detected, if the target is located in a distant area, background area, adjacent area, or non-target activity area that the user is not concerned with, it should not be considered a valid presence. For example, in scenarios like under-cabinet lighting or desk lamps, users typically want the light fixture to respond only to people in the vicinity of the work area, and not to activities in other areas of the room. Area validation results are used to implement this logical spatial constraint.

[0089] In other embodiments, the aforementioned step S600, which involves "fusing the results of the trigger signal, the target activation determination, and the area verification to obtain an existence determination result," introduces an existence-maintaining mode in addition to the basic valid existence determination mechanism described in step S611. This reduces frequent light-off or state transitions caused by short-term positional shifts, boundary jitter, or instantaneous signal fluctuations. Therefore, as... Figure 3 As shown, in some other embodiments, step S600 may include, but is not limited to, steps S621 to S622: S621. The basic existence state is determined to be valid only if the trigger signal is valid, the target response is determined to exist, and the dominant spatial region is located within the valid detection region. S622. Based on the preset existence retention mode and basic existence state, determine the existence determination result; wherein, the existence retention mode includes strict mode and extended retention mode; in strict mode, the existence determination result is maintained as valid only if the basic existence state is valid; in extended retention mode, the existence determination result is maintained as valid if the basic existence state is valid, or if the existence determination result of the historical detection period is valid and there is a target response in the current detection period.

[0090] Within the current detection cycle, the control unit first determines the basic existence state based on the triggering result of the first sensor, the target activation determination result, and the area verification result; then, based on the preset existence retention mode and the basic existence state, it determines the existence determination result for the current detection cycle. The existence retention mode includes at least two types: strict mode and extended retention mode.

[0091] Similar to step S611 above, in step S621, the control unit acquires, within the k-th detection cycle, the trigger signal output by the first sensor, denoted as... T k The target activation determination result is denoted as... A k The regional verification result is denoted as... Z k .in: T k =1 indicates that the first sensor outputs a valid trigger signal within the current detection cycle. T k =0 indicates that it was not triggered; A k =1 indicates that a target response exists within the current detection period. A k =0 indicates that there is no valid target response; Z k =1 indicates that the dominant spatial region is located within the effective detection area. Z k =0 indicates that the dominant spatial region is outside the effective detection region. Based on the above three results, the control unit determines the basic existence state. Let the basic existence state be... B k Then it can be expressed as:

[0092] Or, equivalently, it can be represented as: B k = T k ∧ A k ∧ Z k The above logic indicates that the basic existence state of the current detection cycle is determined to be valid only when the following three conditions are met simultaneously: the first sensor outputs a valid trigger signal; a target response exists within the current detection cycle; and the dominant spatial region corresponding to the target is located within the valid detection region set by the user.

[0093] In step S622, a state preservation mechanism is further introduced on the basis of the basic existing state, so that the system can meet the user's spatial constraint requirements while taking into account the continuity of lighting control and user experience.

[0094] In one embodiment, the system pre-sets a presence-holding mode, or the user selects the presence-holding mode through a configuration interface. The configuration interface can be, for example, a DIP switch; a button; a software configuration interface; a communication interface; or a host computer or mobile terminal configuration method. In one embodiment, the presence-holding mode includes at least two types: strict mode and extended hold mode. The introduction of these modes is to adapt to different application requirements. For example, in some application scenarios, the user wants the lighting to remain on only when the target is strictly within the effective detection area; in this case, strict mode can be used to obtain a stronger spatial constraint effect. In other application scenarios, the user wants the lighting to remain on even after briefly retreating or slightly leaving the effective detection area once they have entered the work area, as long as the target can still be detected; in this case, extended hold mode can be used to improve the naturalness and continuity of the user experience.

[0095] In strict mode, the control unit determines the existence determination result solely based on the basic existence state of the current detection cycle. Let the existence determination result of the current detection cycle be... P k In strict mode, we have: ,Right now

[0096] This means that the existence determination result of the current detection period is only valid if the basic existence state of the current detection period is valid; once the basic existence state of the current detection period is invalid, the existence determination result of the current detection period immediately becomes invalid.

[0097] Due to the existence of the basic state B k Since the target is required to be within the effective detection area, in strict mode, existence must remain consistent with the VEDZ spatial constraints. That is, a target is considered to exist only when it is within the effective detection area; once the target leaves the effective detection area, or although there is a target response but the triggering and area conditions are no longer met, the system exits the effective existence state.

[0098] In extended hold mode, the control unit considers not only the basic existence state of the current detection cycle, but also the existence determination results of historical detection cycles and whether a target response still exists in the current detection cycle. In one embodiment, if the basic existence state of the current detection cycle is valid, then the existence determination result of the current detection cycle is valid; if the basic existence state of the current detection cycle is invalid, but the existence determination result of historical detection cycles is valid, and a target response still exists in the current detection cycle, then the existence determination result of the current detection cycle remains valid. For example, the above logic can be expressed as:

[0099] Or, equivalently, it can be represented as:

[0100] in: P k 1 indicates the existence determination result of the previous detection cycle; A k =1 indicates that the target response still exists within the current detection period.

[0101] In extended maintenance mode, the system distinguishes between "entering a valid existence state" and "maintaining a valid existence state": entering a valid existence state still requires meeting strict basic existence conditions, namely... B k =1, meaning the target must be effectively triggered and confirmed within the VEDZ; provided that the target was already in a valid existence state in the previous detection cycle, as long as a target response is detected in the current detection cycle, the system can still maintain the existence determination result as valid even if the target is temporarily not located within the VEDZ.

[0102] Extended hold mode introduces a combination of "historical valid state + current target response" to give the system a certain degree of state continuity capability, thereby avoiding erroneous shutdown due to short-term boundary crossings or boundary jitter. For example, in practical applications, the following situations may occur: the user slightly moves backward; the user turns to pick up an object, causing a short-term deviation of the dominant area; changes in user posture cause boundary jitter in distance estimation; multipath reflections or slight positional changes cause the dominant area to switch near the critical boundary. If strict mode is still used in the above situations, the device may immediately lose its valid presence state when it briefly leaves the VEDZ, resulting in frequent switching and affecting the user experience.

[0103] It should be noted that the above implementation method uses the presence determination result immediately adjacent to the previous detection cycle. P k 1 represents the historical detection cycle state. This implementation is simple, highly real-time, and sufficient to meet the needs of most lighting control scenarios. However, in other embodiments, the existence determination results of the most recent historical detection cycles can be logically summarized. For example, as long as at least one of the most recent consecutive cycles is valid, it can be considered that the condition of "the existence determination result of the historical detection cycle is valid" is met. Such an implementation can further enhance stability in environments with more obvious boundary fluctuations.

[0104] In some embodiments, step S700 includes a time continuity confirmation mechanism. Based on the existence determination result of the current detection cycle, an activation control signal is output only when the detection cycle is determined to be valid for a consecutive preset number of detection cycles, in order to further suppress erroneous activation caused by instantaneous noise, short-term boundary jitter, and occasional misjudgments. Therefore, as Figure 4 As shown, in some embodiments, step S700 includes, but is not limited to, steps S711 to S712: S711. Statistical analysis of the existence determination results within the historical detection period of a continuously preset quantity; S712. If the judgment result is valid within a consecutive preset number of historical detection cycles, generate a control signal with the status "on".

[0105] In steps S711 and S712, the control unit presets a continuous confirmation window length N, where N is an integer greater than or equal to 1, representing the number of continuous confirmation detection cycles required before generating the start control signal. The continuous confirmation window length N can be a fixed preset value or can be configured by the user. For example, it can be set through one of the following methods: DIP switch; button selection; software configuration; communication interface parameter distribution; product factory pre-configuration. When N is small, the system starts up faster, but the ability to suppress momentary false alarms is relatively weak; when N is large, the system has a stronger ability to suppress false alarms, but the start response time is relatively longer. Therefore, the specific value of N can be selected according to the actual application scenario. For example, in close-range task lighting scenarios with high requirements for response speed, a smaller N can be selected; in scenarios with strong environmental interference and sensitivity to false lighting, a larger N can be selected.

[0106] The control unit outputs an "on" control signal to activate the external output device only when the presence determination results for the most recent N consecutive detection cycles are all valid. Understandably, the presence determination results here can be the direct result obtained from the aforementioned fusion determination, or the result processed by strict mode or extended hold mode.

[0107] In some embodiments, to avoid frequent switching of the control signal due to factors such as short-term target occlusion, attitude changes, reduced micro-motion, instantaneous sensor noise, or boundary jitter, such as...Figure 5 As shown, in some embodiments, step S700 may include, but is not limited to, steps S721 to S723: S721. When the determination result changes from valid to invalid, start the shutdown timer; S722. If the timing value of the timer is not reached, the control signal is kept in the on state. S723. When the timing value of the timer reaches the delay shutdown threshold, a control signal with the status "off" is generated.

[0108] In steps S721 to S723, during the t-th detection cycle, if a determination result changes from a valid state in the previous detection cycle to an invalid state in the current detection cycle, the control unit starts a shutdown timer. The shutdown timer records the duration for which the system remains in the "no confirmed target" state. If, in a subsequent detection cycle, the determination result becomes valid again, the shutdown timer is cleared or reset, and the control signal remains on. If the determination result remains invalid, the shutdown timer accumulates according to the detection cycle until a preset delay shutdown threshold is reached.

[0109] For example, turning off the timer can be represented as:

[0110] in, P ( t ) indicates the first t The existence determination result of each detection cycle P ( t )=1 indicates that a valid judgment result exists. P ( t )=0 indicates that the judgment result is invalid; T off(t) Indicates the first t The shutdown timer value for each detection cycle; Δ t This represents the time increment corresponding to one detection cycle.

[0111] If the countdown timer value has not reached the delayed shutdown threshold, the control unit maintains the control signal in the ON state. That is, even if the target is not confirmed to exist in the current detection cycle, the system does not immediately output a shutdown signal, but instead provides a certain delay to maintain control stability and user experience. When the countdown timer value reaches or exceeds the delayed shutdown threshold, the control unit generates a OFF control signal to shut down the external output device. Therefore, the control signal can be expressed as:

[0112] Among them, O(t ) indicates the first t Control signals for each detection cycle This indicates the threshold for delayed shutdown.

[0113] It should be understood that the delayed shutdown threshold can be a fixed value, such as 3 seconds, 5 seconds, 10 seconds or other preset durations, or it can be set by the user through a DIP switch, button, software interface or communication interface. Different thresholds can be selected for different application scenarios: for scenarios where a quick response to shutdown is desired, a shorter delay can be set; for scenarios where there are many instances of short-term stillness, turning around, obstruction, or leaving and then quickly returning, a longer delay can be set.

[0114] In engineering implementation, the control unit performs the following processing at the end of each detection cycle: First, it reads the current judgment result; if the result is valid, it immediately outputs an on control signal and resets the off timer; if the result is invalid, it checks whether the off timer has been started. If it has not been started, it starts counting from the current cycle; if it has been started, it continues to accumulate. Then, it compares the accumulated time with the delayed off threshold. If the threshold is not reached, the output remains on; if the threshold is reached, the output is off. Therefore, even if the target detection is interrupted within a short period, the system will not immediately shut down the external output device, thus reducing the probability of false shutdown.

[0115] In some embodiments, the delayed shutdown threshold is not fixed but adaptively adjusted based on the target presence confidence. That is, the control unit calculates the target presence confidence based on the target activation determination results, region verification results, motion-related signal features, and static-related signal features within the historical detection period, and positively adjusts the delayed shutdown threshold according to the target presence confidence, so that the time kept on is extended when the probability of the target presence is high, and the time kept on is shortened when the probability of the target presence is low.

[0116] The control unit can perform statistical or weighted processing on relevant data from the most recent M consecutive detection cycles to obtain the confidence level of the target's presence. C ( t The confidence level of target presence can comprehensively reflect the following factors: first, whether the target activation determination result remains valid; second, whether the dominant spatial region remains within the valid detection region; third, the strength of motion-related signal features; and fourth, the strength of static correlation signal features or smoothed static correlation signal features. Since the target activation determination result and the region verification result directly characterize "whether a target was detected" and "whether the target is located within the valid region," in one implementation, these two items can be given higher weights, while motion-related and static correlation signal features participate in the weighting as supplementary quantitative indicators.

[0117] For example, the confidence level of the target can be expressed as:

[0118] in, This represents the statistical value or average value of target activation determination results within the historical detection period. This represents the statistical value or average value of regional verification results within a historical testing period. This represents the normalized statistical values ​​of the motion-related signal characteristics. This represents the normalized or smoothed statistical characteristics of the static correlation signal. w 1. w 2. w 3. w 4 is the weighting coefficient, and it satisfies... w 1+ w 2+ w 3+ w 4=1. In one implementation, C ( t The above key weight coefficients are normalized to the interval [0,1]. w 1. w 2. w 3. w The specific numerical settings and adjustments for 4 can be derived by training and calculating historical sample data (such as target activity recordings or raw data trajectories recorded by sensors under different environments) in the preset application scenario using machine learning classification algorithms (such as logistic regression) or multi-attribute decision evaluation methods (such as entropy weighting or analytic hierarchy process AHP). At the engineering debugging level, target activation determination results and area verification results (i.e., ...) are typically assigned according to actual needs. w 1 and w 2) Larger base weights ensure that the system has a decisive constraint on the existence of the target both logically and spatially; simultaneously, based on feedback from the false negative and false positive rates in specific scenarios (such as office areas or corridors), the weight ratios corresponding to motion-related and statically related signal features can be dynamically fine-tuned (i.e., w 3 and w 4), thereby adapting to the sensitivity requirements of different application environments.

[0119] Based on the confidence level of the target's existence, the delayed shutdown threshold can be determined as follows:

[0120] in, Th base Based on the base delay shutdown threshold, Lambdaλ is an adjustment coefficient used to characterize the degree of influence of confidence level on the delayed shutdown threshold. In practical applications, the adjustment coefficient λ is mainly set based on the maximum allowable delay shutdown time increment of the scenario. During system deployment, the value of λ can be directly determined by the difference between the user-defined maximum tolerable delay duration (e.g., 20 seconds) and the basic delayed shutdown threshold (e.g., 5 seconds) (set to 15 seconds in this example). Furthermore, λ can also be manually set by looking up a table or adaptively calling different preset values ​​based on the functional attributes of the application space (e.g., different requirements such as automatic shutdown when people leave a passageway area, and continuous illumination while on in an office area). Therefore, when C ( t When the value is large, it indicates that a strong and reliable target response has been present for several recent detection cycles, and the system accordingly increases the delay shutdown threshold; when C ( t When the value is small, it indicates a lower confidence level in the target's existence. The system then uses a shorter hold time to improve the shutdown response speed. For example, in one application scenario, if a target has just briefly disappeared from the boundary of the effective detection area, but has consistently met the requirements of target activation and successful area verification over previous detection cycles, while maintaining a high level of static correlation signal characteristics, a high confidence level in the target's existence can be obtained. In this case, even if a short-term invalidation occurs in the current cycle, the system can still appropriately extend the shutdown waiting time to avoid prematurely shutting down the control output for targets still nearby. Conversely, if there are only scattered weak motion signals, unstable area verification results, and low static correlation signal characteristics over previous detection cycles, the confidence level in the target's existence is low, and the system can reach the shutdown condition more quickly, thereby reducing unnecessary holding caused by occasional interference.

[0121] In another alternative implementation, the target existence confidence level is not limited to a linear weighted form; it can also be implemented using piecewise functions, lookup tables, fuzzy logic methods, or simple rating scales. For example, the target existence confidence level can be divided into high, medium, and low levels based on the number of historical target activations and the number of successful regional verifications, and then corresponding to different delay-off threshold ranges for each level. The key is to achieve a positive adjustment relationship where "the higher the confidence level, the larger the delay-off threshold." Furthermore, in practical implementations, to avoid control anomalies caused by excessively large or small delay-off thresholds, additional measures can be taken... Th off(t) Set an upper and lower limit value. For example, limit it to... Th min and Th Th max The calculation result is taken between the lower and upper limits. When the result is less than the lower limit, the lower limit is used; when the result is greater than the upper limit, the upper limit is used. This method ensures both adaptive adjustment capability and predictability of system behavior and engineering stability.

[0122] The embodiments of this application have been described in detail above with reference to the accompanying drawings. However, this application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of this application. Furthermore, unless otherwise specified, the embodiments and features described in the embodiments of this application can be combined with each other.

Claims

1. A target presence detection method, characterized in that, The target presence detection method is applied to a presence detection system including a first sensor and a radio frequency detection sensor, and includes: Based on the user-defined detection range parameters, an effective detection area is constructed; The first sensor detects the environment and outputs a valid trigger signal when the triggering conditions are met. When the trigger signal is valid, the signal feature set output by the radio frequency detection sensor in the current detection period is obtained, wherein the signal feature set includes at least motion-related signal features, static-related signal features, and spatial position-related information. Target activation is determined based on the signal feature set to ascertain whether a target response exists within the current detection period. Based on the spatial location-related information, the dominant spatial region corresponding to the target response is determined, and it is determined whether the dominant spatial region is located within the effective detection region to obtain the region verification result; A fusion determination is performed based at least on the trigger signal, the result of the target activation determination, and the region verification result to obtain an existence determination result; Based on the existence determination result, a control signal is generated to control the external output device.

2. The target presence detection method according to claim 1, characterized in that, The step of determining target activation based on the signal feature set to ascertain whether a target response exists within the current detection period includes: If the motion-related signal feature is not less than a preset motion determination threshold, or if the static-related signal feature is not less than a preset static determination threshold, then the target response is determined to exist within the current detection period.

3. The target presence detection method according to claim 1, characterized in that, The output modes of the radio frequency detection sensor include discrete distance partition mode and global distance mode; In the discrete distance partitioning mode, the effective detection area is characterized as a set of distance partitions whose numbers do not exceed the effective judgment threshold; The spatial location-related information is represented by several candidate distance partitions and their corresponding signal features; In the global distance mode, the effective detection area is characterized as a continuous spatial range that does not exceed the maximum effective detection distance; the spatial location-related information is represented as the distance to a single target or the area identifier mapped thereto.

4. The target presence detection method according to claim 3, characterized in that, The process of determining the dominant spatial region corresponding to the target response based on the spatial location-related information, and determining whether the dominant spatial region is located within the effective detection region to obtain a region verification result, includes: In the discrete distance partitioning mode, the representative energy corresponding to each candidate distance partition is determined, and the candidate distance partition with the largest representative energy is selected as the dominant spatial region; wherein, the representative energy is determined by the larger value between the motion-related signal feature and the smoothed static-related signal feature of the corresponding candidate distance partition; In the global distance mode, the spatial region mapped by the single target distance is directly used as the dominant spatial region.

5. The target presence detection method according to claim 1, characterized in that, The fusion determination based at least on the trigger signal, the result of the target activation determination, and the region verification result to obtain an existence determination result includes: The presence determination result is valid only if the trigger signal is valid, the target response is determined to exist, and the dominant spatial region is located within the valid detection region.

6. The target presence detection method according to claim 1, characterized in that, The fusion determination based at least on the trigger signal, the result of the target activation determination, and the region verification result to obtain an existence determination result includes: The basic existence state is determined to be valid only if the trigger signal is valid, the target response is determined to exist, and the dominant spatial region is located within the valid detection region; The existence determination result is determined based on the preset existence preservation mode and the basic existence state; wherein, the existence preservation mode includes strict mode and extended preservation mode; In the strict mode, the existence determination result is maintained as valid only if the basic existence state is valid; In the extended maintenance mode, if the basic existence state is valid, or if the existence determination result of the historical detection period is valid and a target response exists in the current detection period, the existence determination result is maintained as valid.

7. The target presence detection method according to any one of claims 1 to 6, characterized in that, The step of generating a control signal based on the existence determination result includes a time continuity confirmation mechanism: The presence determination results within a historical detection period of a continuously preset number of data points are statistically analyzed. If the existence determination result is valid within the consecutive preset number of historical detection cycles, the control signal with the state "on" is generated.

8. The target presence detection method according to any one of claims 1 to 6, characterized in that, The step of generating a control signal based on the existence determination result further includes: When the existence determination result changes from valid to invalid, start the shutdown timer; If the countdown value of the timer does not reach the delay shutdown threshold, the control signal remains in the on state. When the countdown value of the shutdown timer reaches the delay shutdown threshold, a control signal is generated indicating that the status is off.

9. The target presence detection method according to claim 8, characterized in that, The delayed shutdown threshold is obtained in the following way: Based on the target activation determination results, the region verification results, the motion-related signal features, and the static-related signal features within the historical detection period, the target existence confidence is calculated using a weighted average. The delay-off threshold is positively adjusted based on the confidence level of the target's existence.

10. The target presence detection method according to any one of claims 1 to 6, characterized in that, The step of determining target activation based on the signal feature set to ascertain whether a target response exists within the current detection period further includes: The static correlation signal features are subjected to time smoothing to obtain smoothed static correlation signal features.

Images