Air conditioner control method, device and equipment based on millimeter radar waves and storage medium

Abstract

The invention relates to the technical field of air conditioner control, in particular to an air conditioner control method, device and equipment based on millimeter radar waves and a storage medium. Interference sources such as fans and shaking objects can be effectively filtered, the robustness of the system is remarkably improved, a real person target and a non-biological target are accurately distinguished, and the method has the advantage of being low in misjudgment rate; compared with a traditional infrared sensor, the target type is confirmed based on continuous radar data of millimeter radar waves, and environment temperature interference and direction limitation are successfully overcome; on the basis, the working state of the air conditioner is dynamically adjusted according to the target type, the track characteristics and the static coordinates, the problems that in the prior art, micro-motion target detection is inaccurate, and non-biological targets are prone to being misjudged are effectively solved, finally, dynamic and accurate control over the air conditioner is achieved, the working energy consumption of the air conditioner is reduced, and the user experience is improved.

Description

Technical Field

[0001] This invention relates to the field of air conditioning control technology, and in particular to an air conditioning control method, device, equipment and storage medium based on millimeter radar waves. Background Technology

[0002] Smart furniture needs to accurately sense the presence, location, and movement of the human body in order to achieve automated control and personalized services. Millimeter-wave radar can penetrate non-metallic obstacles such as clothing and curtains, and is not affected by environmental factors such as light and temperature. It can accurately detect the human body in all weather and all scenarios, so it is widely used in various smart furniture such as smart sofas, smart mattresses, smart air conditioners, and smart lamps.

[0003] Taking smart air conditioners as an example, traditional air conditioners typically use infrared sensors for human detection. However, infrared sensors are easily affected by ambient temperature and can only detect the presence of people in a specific direction, making it impossible to accurately track and dynamically sense the position of people. On the other hand, millimeter-wave radar can accurately track the position of people in the room and adjust the airflow direction according to the position of people, solving the problems of uneven airflow and direct airflow onto people, thus improving the user experience. At the same time, millimeter-wave radar can also eliminate false triggers from interference objects such as fans, curtains, and green plants, ensuring that the air conditioner turns off in time or enters energy-saving mode when no one is around, achieving the goal of energy saving.

[0004] However, existing millimeter-wave radar technology still has significant shortcomings in smart furniture applications.

[0005] On the one hand, when the human body transitions from motion to stillness, its behavioral patterns and physical characteristics change significantly. During motion, the human body's position and velocity parameters change greatly, making it easy for millimeter-wave radar to track and collect data. However, after transitioning from motion to stillness, the human body only experiences micro-movements such as breathing and slight limb movements, generating weak and complex signals that pose a significant challenge to the data acquisition and processing of millimeter-wave radar. In particular, acquiring the X-axis coordinates of a stationary human body is difficult because the characteristics of micro-movement signals in the X-axis direction are not obvious and are difficult to distinguish from interference signals such as environmental noise. This results in millimeter-wave radar being unable to accurately obtain coordinate information, making it impossible for smart home systems to accurately determine the human body's position. For example, smart air conditioners rely on the human body's position to adjust their airflow direction. If accurate X-axis coordinates cannot be obtained, the airflow direction cannot be precisely adjusted, affecting control accuracy and user experience.

[0006] On the other hand, in complex home environments, millimeter-wave radar can easily misidentify non-biological targets, such as curtains swaying in the wind or green plants, as human beings, causing smart furniture to malfunction. For example, the swaying of curtains may be mistaken by radar as human activity, causing smart air conditioners that should be turned off to remain on, resulting in energy waste.

[0007] It is evident that existing technologies still need improvement and enhancement. Summary of the Invention

[0008] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide an air conditioning control method based on millimeter radar waves, which solves the problems of inaccurate detection of micro-moving targets and easy misjudgment of non-biological targets in the prior art, and realizes dynamic and precise control of the air conditioning working status.

[0009] The first aspect of this invention provides an air conditioning control method based on millimeter radar waves, comprising: acquiring continuous radar data and preprocessing it to obtain preprocessed continuous data; extracting trajectory features based on the preprocessed continuous data; confirming the target type based on the extracted trajectory features, wherein the target type includes human targets and non-biological targets; when the target type is a human target, if the trajectory features indicate that the human target changes from a moving state to a stationary state, then acquiring the stationary coordinates of the human target; and adjusting the operating state of the air conditioner based on the target type, stationary coordinates, or trajectory features.

[0010] Optionally, in a first implementation of the first aspect of the present invention, the step of acquiring continuous radar data and preprocessing it to obtain preprocessed continuous data includes: acquiring continuous radar data, wherein the continuous radar data includes multiple frames of raw radar information collected based on a preset time interval, each frame of raw radar information including X coordinate, Y coordinate, target energy value and its corresponding timestamp; performing outlier filtering on the continuous radar data to obtain filtered continuous data; and using a moving average filtering algorithm to perform noise smoothing on the X coordinate and Y coordinate of the continuous radar data respectively to obtain preprocessed continuous data.

[0011] Optionally, in a second implementation of the first aspect of the present invention, the step of extracting trajectory features based on preprocessed continuous data includes: extracting coordinate data from each frame of original radar information and constructing a continuous coordinate sequence based on the extracted coordinate data; calculating the ratio of displacement difference to time difference between two adjacent frames of original radar information to obtain multiple displacement change rates; calculating the standard deviation of target energy values ​​in multiple frames of original radar information to obtain energy fluctuation features; performing time-series analysis on the Y coordinate in multiple frames of original radar information, counting the number of reciprocating movements within a preset unit time to obtain motion periodicity features; and integrating the continuous coordinate sequence, multiple displacement change rates, energy fluctuation features, and motion periodicity features to obtain trajectory features.

[0012] Optionally, in a third implementation of the first aspect of the present invention, the step of confirming the target type based on the extracted trajectory features, wherein the target type includes real-life targets and non-biological targets, includes: acquiring a pre-constructed human motion template, wherein the human motion template includes trajectory continuity features and direction consistency features; when the continuous coordinate sequence and multiple displacement change rates all conform to the human motion template, and the motion periodicity feature is without a fixed period, the target type is confirmed as a real-life target; when the continuous coordinate sequence and multiple displacement change rates do not conform to the human motion template, and the motion periodicity feature is with a fixed period, and the energy fluctuation feature is less than a preset energy value with no obvious fluctuation, the target type is confirmed as a non-biological target.

[0013] Optionally, in the fourth implementation of the first aspect of the present invention, when the target type is a real person target, if the trajectory features indicate that the real person target has changed from a moving state to a stationary state, then obtaining the stationary coordinates of the real person target includes: when the target type is a real person target, if the trajectory features indicate that the real person target has changed from a moving state to a stationary state, then obtaining the X-coordinate and its corresponding timestamp of the last valid original radar information before the stationary state, and obtaining the X-coordinates and corresponding timestamps of the three original radar information frames before the last valid original radar information frame; calculating the average velocity before the stationary state based on the X-coordinates and corresponding timestamps of the three original radar information frames before the stationary state; and determining the stationary coordinates of the real person target using a linear prediction algorithm based on the calculated average velocity before the stationary state, combined with the X-coordinate and corresponding timestamp of the last valid original radar information frame.

[0014] Optionally, in a fifth implementation of the first aspect of the present invention, adjusting the working state of the air conditioner based on the target type, stationary coordinates, or trajectory features includes: when the target type is a non-biological target, controlling the air conditioner to enter an energy-saving mode or a standby state; when the target type is a real person target, if the trajectory features indicate that the real person target is in motion, then adjusting the working state of the air conditioner based on the trajectory features; if the trajectory features indicate that the real person target is in a stationary state, then adjusting the working state of the air conditioner based on the stationary coordinates.

[0015] Optionally, in the sixth implementation of the first aspect of the present invention, the step of adjusting the working state of the air conditioner based on the target type, stationary coordinates or trajectory features further includes: recording the correction scenario in which the air conditioner's action is corrected, and obtaining trajectory feature data under the correction scenario; storing the obtained trajectory feature data in a pre-built sample library; and when a preset iteration time is reached, optimizing the preset energy value without significant fluctuations and the human motion template based on the trajectory feature data stored in the sample library.

[0016] A second aspect of the present invention provides an air conditioning control device based on millimeter radar waves, comprising: a processing module for acquiring continuous radar data and preprocessing it to obtain preprocessed continuous data; an extraction module for extracting trajectory features based on the preprocessed continuous data; a confirmation module for confirming the target type based on the extracted trajectory features, wherein the target type includes human targets and non-biological targets; an acquisition module for acquiring the stationary coordinates of the human target if the trajectory features indicate that the human target has changed from a moving state to a stationary state when the target type is a human target; and an adjustment module for adjusting the working state of the air conditioner based on the target type, stationary coordinates, or trajectory features.

[0017] A third aspect of the present invention provides an air conditioning control device based on millimeter radar waves, the air conditioning control device based on millimeter radar waves comprising: a memory and at least one processor, the memory storing instructions; at least one processor calling the instructions in the memory to cause the air conditioning control device based on millimeter radar waves to execute the various steps of the air conditioning control method based on millimeter radar waves described in any of the preceding claims.

[0018] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the air conditioning control method based on millimeter radar waves described in any of the preceding claims.

[0019] The technical solution of this invention leverages the penetrating power and environmental interference resistance of millimeter-wave radar. By collecting continuous radar data of the target and performing preprocessing and trajectory feature analysis, it can effectively filter interference sources such as fans and shaking objects, significantly improving system robustness. This allows for accurate differentiation between human targets and non-biological targets, characterized by a low false positive rate. Compared to traditional infrared sensors, the continuous radar data based on millimeter-wave radar successfully overcomes environmental temperature interference and directional limitations in target type confirmation. Furthermore, based on the target type, trajectory characteristics, and stationary coordinates, the air conditioner's operating status is dynamically adjusted, effectively solving the problems of inaccurate detection of micro-moving targets and easy misjudgment of non-biological targets in existing technologies. Ultimately, this achieves dynamic and precise control of the air conditioner, reducing energy consumption and improving user experience without requiring additional hardware costs. Attached Figure Description

[0020] Figure 1 A logic flowchart of an air conditioning control method based on millimeter radar waves provided in an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of the structure of an air conditioning control device based on millimeter radar waves provided in an embodiment of the present invention;

[0022] Figure 3This is a schematic diagram of the structure of an air conditioning control device based on millimeter radar waves provided in an embodiment of the present invention. Detailed Implementation

[0023] This invention provides an air conditioning control method, apparatus, device, and storage medium based on millimeter radar waves. In this invention, the terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0024] This application discloses an air conditioning control method based on millimeter radar waves. For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the air conditioning control method based on millimeter radar waves in this invention includes:

[0025] 101. Acquire continuous radar data and preprocess it to obtain preprocessed continuous data;

[0026] In this embodiment, a 24GHz millimeter-wave radar module is used, which is installed on the air conditioner body, such as near the air outlet or inside the panel, to ensure that the radar detection range covers the main activity areas of the user in the room, such as the core activity spaces of the living room and bedroom, and to avoid incomplete data collection due to obstruction. The millimeter-wave radar module collects data at a preset time interval of 50ms / frame. Each frame of data includes the target's X coordinate (unit: m), Y coordinate (unit: m), target energy value reflecting the intensity of micro-motion within the range of 0-100, and a timestamp accurate to the millisecond level.

[0027] 102. Extracting trajectory features based on preprocessed continuous data;

[0028] In this embodiment, the extracted trajectory features include continuous coordinate sequences, displacement change rate, energy fluctuation features, and motion periodicity features.

[0029] 103. Based on the extracted trajectory features, the target type is identified, including human targets and non-biological targets;

[0030] 104. When the target type is a real person, if the trajectory features indicate that the real person has changed from a moving state to a stationary state, then obtain the stationary coordinates of the real person.

[0031] 105. Adjust the working status of the air conditioner based on the target type, stationary coordinates, or trajectory characteristics.

[0032] This application discloses an air conditioning control method based on millimeter-wave radar. Leveraging the penetrating power and environmental interference resistance of millimeter-wave radar, by collecting continuous radar data of the target and performing preprocessing and trajectory feature analysis, it can effectively filter interference sources such as fans and moving objects, significantly improving system robustness. This allows for accurate differentiation between human and non-biological targets with a low false positive rate. Compared to traditional infrared sensors, continuous radar data based on millimeter-wave radar successfully overcomes environmental temperature interference and directional limitations in target type identification. Furthermore, based on the target type, trajectory characteristics, and stationary coordinates, the air conditioning operating state is dynamically adjusted, effectively solving the problems of inaccurate detection of micro-moving targets and easy misjudgment of non-biological targets in existing technologies. Ultimately, this achieves dynamic and precise control of the air conditioning, reduces energy consumption, improves user experience, and eliminates the need for additional hardware costs.

[0033] Furthermore, in this embodiment of the invention, the step of acquiring continuous radar data and preprocessing it to obtain preprocessed continuous data includes:

[0034] 201. Acquire continuous radar data, which includes multiple frames of raw radar information collected based on a preset time interval. Each frame of raw radar information includes X coordinate, Y coordinate, target energy value and its corresponding timestamp.

[0035] In this embodiment, the preset time interval is set to 50 milliseconds, which is 20 frames per second; the original radar information format of each frame is as follows: X = 2.5 meters, Y = 4.0 meters, energy value is 18, and timestamp is 1620000000000 milliseconds; continuous radar data is formed by continuously collecting 100 frames of data.

[0036] 202. Perform outlier filtering on continuous radar data to obtain filtered continuous data;

[0037] In this embodiment, the effective detection range of the radar is set as: X∈[0,10m], Y∈[0,10m], and the target energy value∈[1,100]. By performing outlier filtering on continuous radar data, data that exceeds the effective detection range, such as the original radar information with X=11m and energy value of 0, is removed, thereby obtaining filtered continuous data.

[0038] 203. Using a moving average filtering algorithm, noise smoothing is applied to the X and Y coordinates of continuous radar data to obtain preprocessed continuous data.

[0039] In this embodiment, a 5-frame sliding window is used to smooth the noise of the X and Y coordinates of the original radar information. For example, for the X coordinate sequence [2.5, 2.6, 2.4, 2.7, 2.5, 2.6], after smoothing the original radar information in the 5th frame, the X coordinate value is (2.5+2.6+2.4+2.7+2.5) / 5 = 2.54 meters, and after smoothing the original radar information in the 6th frame, the X coordinate value is (2.6+2.4+2.7+2.5+2.6) / 5 = 2.56 meters, thereby effectively reducing the interference of instantaneous noise.

[0040] In this embodiment, a standardized data preprocessing procedure is used to accurately filter out various types of invalid noise data, improve the spatiotemporal stability and measurement accuracy of radar data, provide reliable input for subsequent trajectory feature extraction and target recognition, and effectively avoid the risk of target misjudgment caused by abnormal raw data; for example, by removing extreme coordinate values, the probability of non-biological targets being misidentified as real targets is reduced.

[0041] Furthermore, in this embodiment of the invention, the extraction of trajectory features based on preprocessed continuous data includes:

[0042] 301. Extract the coordinate data of each frame of raw radar information and construct a continuous coordinate sequence based on the extracted coordinate data;

[0043] In this embodiment, the X and Y coordinate data of the 20 frames of preprocessed original radar information are extracted to construct a continuous coordinate sequence, such as [(1.0, 2.0), (1.2, 2.1), (1.3, 2.2), ..., (3.0, 2.5)], which clearly and intuitively shows the target's motion trajectory.

[0044] 302. Calculate the ratio of displacement difference to time difference between two adjacent frames of original radar information to obtain multiple displacement change rates;

[0045] In this embodiment, the time difference between adjacent frames is 50 milliseconds. For example, if the X-coordinate of the original radar information in frame t is 1.2 meters and the X-coordinate of the original radar information in frame (t-1) is 1.0 meters, then the displacement difference is 0.2 meters, and the displacement change rate is 0.2m / 0.05s, or 4 meters / second. Since the normal walking speed range of the human body is 1-1.5m / s, this speed exceeds the normal speed range of the human body and may be considered an abnormal situation. In another case, if the X-coordinate of the original radar information in a certain frame is 1.3 meters and the X-coordinate of the original radar information in the previous frame (frame t-1) is 1.2 meters, then the displacement change rate is 0.1m / 0.05s, or 2 meters / second. This displacement change rate is within the normal speed range of the human body.

[0046] 303. Calculate the standard deviation of the target energy value in multiple frames of raw radar information to obtain the energy fluctuation characteristics;

[0047] In this embodiment, 10 frames of target energy values ​​[8, 9, 8, 7, 9, 8, 7, 9, 8, 7] are taken, and the standard deviation is calculated: First, the mean is calculated, which is 8; then, the sum of squares of the differences between the target energy value of each frame and the mean is calculated, which is (0+1+0+1+1+0+1+1+0+1)=6; finally, the standard deviation is calculated, which is approximately equal to 0.77, which is less than 5, so it is determined that the energy is stable.

[0048] 304. Perform time-series analysis on the Y-coordinate in multiple frames of raw radar information, count the number of reciprocating motions within a preset unit time, and obtain the motion periodicity characteristics.

[0049] In this embodiment, the preset unit time is 1 second, corresponding to 20 frames of original radar information; the Y coordinate of the original radar information is analyzed in time sequence. For example, the Y coordinate of the curtain moves back and forth within the range of [1.5, 2.5] meters: it rises from 1.5 meters to 2.5 meters in 1 second, then falls back to 1.5 meters, then rises back to 2.5 meters, and finally returns to 1.5 meters, repeating this cycle twice. Its periodicity is 2 times / second, that is, the period is 0.5Hz.

[0050] 305. By integrating continuous coordinate sequences, multiple displacement change rates, energy fluctuation characteristics, and motion periodicity characteristics, trajectory characteristics are obtained;

[0051] In this embodiment, the trajectory features are visualized with precise geometric shapes. By constructing a multi-dimensional trajectory feature extraction system, the motion patterns of the target are comprehensively and deeply depicted from multiple dimensions such as the spatial distribution, time series, and dynamic changes of the motion trajectory. This process can effectively capture the essential differences exhibited by different types of targets during motion, thus providing objective and quantitative scientific basis for distinguishing between human beings and non-biological targets. For example, when a human being moves, its energy consumption exhibits significant fluctuations, and these fluctuations do not have fixed periodic characteristics, reflecting the flexibility and randomness of the movement of living organisms. In contrast, the energy state of non-biological targets is relatively stable, and their movement patterns often follow fixed periodic patterns, exhibiting regularity under mechanical or preset program control. By integrating and analyzing multi-dimensional trajectory features, the problems of insufficient recognition accuracy and susceptibility to interference caused by relying on only a single feature in the prior art are effectively solved, significantly improving the accuracy and reliability of target classification in complex scenarios, and providing more solid data support for subsequent intelligent decision-making and response.

[0052] Furthermore, in this embodiment of the invention, the step of confirming the target type based on the extracted trajectory features includes: real human targets and non-biological targets, comprising:

[0053] 401. Obtain a pre-constructed human motion template, wherein the human motion template includes trajectory continuity features and direction consistency features;

[0054] In this embodiment, the human motion template is a feature set based on the natural movement patterns of real people, used to distinguish between real human targets and non-biological targets. The template includes two features: trajectory continuity and direction consistency. Trajectory continuity is reflected in the displacement difference between adjacent frames being ≤0.8m, which conforms to the human stride length, and the displacement difference between consecutive 5 frames shows a gradual change, such as 0.3→0.5→0.6→0.4→0.2m, that is, without abrupt changes. Direction consistency is reflected in the number of times the trajectory direction reverses within 1 second being ≤1 time. For example, when moving from left to right, there is only one brief turning back within 1 second, and the reversal is not frequent.

[0055] 402. When the continuous coordinate sequence and multiple displacement change rates all conform to the human motion template, and the motion periodicity characteristic is that there is no fixed period, the target type is confirmed to be a real human target.

[0056] For example, in a continuous coordinate sequence, the displacement differences between adjacent frames are 0.3 meters, 0.5 meters, 0.4 meters, and 0.6 meters, respectively, all not exceeding 0.8 meters, which meets the requirements for trajectory continuity; the direction reverses once within 1 second, meeting the standard for directional consistency; the displacement change rate is mostly between 0.5 and 1.5 meters per second, which matches the normal human movement speed; the periodicity of the movement is manifested in the random number of reciprocations within 1 second, without a fixed period; based on the above indicators, the target is determined to be a real person.

[0057] 403. When the continuous coordinate sequence and multiple displacement change rates do not conform to the human motion template, and the motion periodicity is a fixed period, and the energy fluctuation characteristics are less than the preset no obvious fluctuation energy value, the target type is confirmed as a non-biological target.

[0058] For example, in a continuous coordinate sequence, the displacement difference between adjacent frames is 1.2 meters, exceeding the threshold of 0.8 meters, which does not meet the requirement of trajectory continuity; the direction reverses 5 times within 1 second, which does not meet the standard of directional consistency; the motion periodicity feature is that it repeats twice within 1 second with a fixed period of 0.5 Hz, which has a fixed period; the energy fluctuation feature value is 2, which is lower than the preset value of 5, and does not show obvious fluctuation energy; based on the above analysis, the target is determined to be a non-biological target, such as a curtain.

[0059] In this embodiment, by using a pre-constructed human motion feature template combined with multi-dimensional trajectory features, targets within the monitoring range can be accurately classified, enabling precise differentiation between real human activity and non-biological interference targets. This effectively solves the common problem of misjudging micro-moving targets and non-biological interference in traditional detection technologies. For example, it can accurately avoid misidentification problems caused by indoor airflow disturbances such as swaying curtains and moving clothes, significantly improving the accuracy and robustness of target detection. Through reliable human activity status judgment criteria, the air conditioner can dynamically adjust its operating parameters based on the actual human presence, thereby minimizing ineffective energy consumption and unnecessary manual user intervention, ensuring a comfortable experience while achieving energy saving and efficiency improvement.

[0060] Furthermore, in this embodiment of the invention, when the target type is a real person, if the trajectory features indicate that the real person has changed from a moving state to a stationary state, then obtaining the stationary coordinates of the real person includes:

[0061] 501. When the target type is a real person, if the trajectory features indicate that the real person changes from a moving state to a stationary state, then obtain the X coordinate of the last valid original radar information before it stops and its corresponding timestamp, and obtain the X coordinate of the three original radar information frames before the last valid original radar information and their corresponding timestamps.

[0062] In this embodiment, when the change in X and Y coordinates of a real target is less than 0.1 meters in five consecutive frames of raw radar information (e.g., X coordinate is 3.0 meters for five consecutive frames, and the change does not exceed 0.05 meters), it can be determined that the real target has changed from a moving state to a stationary state. Based on the aforementioned example, the last frame of valid raw radar information is confirmed to show: X = 3.0 meters, timestamp t0 = 1620000001000 milliseconds; the first three frames of raw information are as follows: t-3 frame: X = 2.8 meters, timestamp = 1620000000950 milliseconds; t-2 frame: X = 2.9 meters, timestamp = 1620000000900 milliseconds; t-1 frame: X = 3.0 meters, timestamp = 1620000000900 milliseconds;

[0063] = 1620000000850 milliseconds.

[0064] 502. Calculate the average velocity before coming to a stop based on the X coordinates and corresponding timestamps of the first three frames of raw radar information;

[0065] In this embodiment, based on the above example, the time interval is t0-t-3=150 milliseconds, and the displacement change is 3.0 meters-2.8 meters=0.2 meters. Therefore, the average speed is 0.2 meters / 0.15 seconds≈1.33 meters / second.

[0066] 503. Based on the calculated average velocity before stationary motion, combined with the X-coordinate of the last frame of valid original radar information and its corresponding timestamp, a linear prediction algorithm is used to determine the stationary coordinates of the real target.

[0067] In this embodiment, the linear prediction formula is: X 静止 =X 最后 +average speed × (t) 当前 -t0); Since the velocity approaches 0 when stationary, X is taken in actual operation. 静止 =3.0 meters, which means maintaining the last valid coordinates to ensure the accuracy of the target coordinates in a stationary state.

[0068] In this embodiment, to address the issue of X-axis coordinate acquisition failure in existing millimeter-wave radar when detecting slightly moving or stationary human targets, a linear prediction algorithm is introduced. By analyzing historical data of the target's motion trajectory, it achieves accurate completion and dynamic tracking of the X-axis coordinates of a stationary human body. This overcomes the limitations of traditional radar in capturing signals from slightly moving targets, ensuring that the air conditioning control system can continuously acquire stable spatial position parameters of the human body even in low-dynamic scenarios such as when the user is sitting or resting. This effectively avoids equipment misjudgment and shutdown or air supply angle deviation caused by the loss of coordinate information. By constructing a closed-loop control mechanism of real-time detection, trajectory prediction, coordinate completion, and dynamic adjustment, the air conditioner can optimize the air supply strategy in real time based on the completed stationary coordinates, achieving seamless and accurate positioning throughout the entire process from dynamic movement to static rest, significantly improving user comfort.

[0069] Furthermore, in this embodiment of the invention, adjusting the working state of the air conditioner based on the target type, stationary coordinates, or trajectory features includes:

[0070] 601. When the target type is a non-biological target, control the air conditioner to enter energy-saving mode or standby mode;

[0071] In this embodiment, when a non-biological target is detected, such as a swaying curtain, the air conditioner is controlled to enter the energy-saving mode: in the cooling state, the cooling temperature is increased from 26°C to 28°C, and the fan speed is reduced from high to low; if only non-biological targets are detected for 5 consecutive minutes, that is, no real human targets are detected for 5 consecutive minutes, the air conditioner will automatically switch to standby mode, with only the indicator light remaining on.

[0072] 602. When the target type is a real person, if the trajectory features indicate that the real person is in motion, then the working state of the air conditioner is adjusted based on the trajectory features; if the trajectory features indicate that the real person is stationary, then the working state of the air conditioner is adjusted based on the stationary coordinates.

[0073] In this embodiment, the air conditioning is controlled for a real person in motion. For example, when the trajectory features of the real person show that they are moving from the left side of the living room (X=1m) to the right side (X=5m), the air sweeping angle of the air conditioner is gradually adjusted from 20° to the left side to 70° to the right side based on the trajectory features to ensure that the air supply follows the position of the human body. If the real person moves faster, for example, more than 1m / s, the current temperature is maintained.

[0074] In this embodiment, the air conditioning control is designed for a real person who changes from a moving state to a stationary state. That is, when the real person is stationary at the moment of control, for example, the stationary coordinates of the real person are X=3m, Y=2m, and the distance from the air conditioner is 1.5m. Based on the stationary coordinates, the fan speed is adjusted from high to medium to avoid discomfort from direct airflow. If the stationary position is directly in front of the air conditioner, for example, Y=1m, the anti-direct airflow mode is activated, that is, the air supply angle is adjusted upward by 30°.

[0075] In this embodiment, in the field of modern smart home appliances, the traditional one-size-fits-all control method of air conditioners is gradually failing to meet diverse practical needs. To effectively solve this technical problem, a strategy is proposed to dynamically adjust the air conditioner's operating parameters according to different target types and states. Specifically, when a non-biological target is detected, the air conditioner will automatically switch to energy-saving mode or standby mode to minimize unnecessary energy consumption and practice the concept of green environmental protection. When a person is detected to be in motion, the air conditioner will intelligently follow the airflow to ensure that the user can enjoy cool and comfortable airflow no matter where they are. If a person is detected to be stationary, the air conditioner will optimize the airflow speed, ensuring indoor air circulation while adjusting the airflow speed to the most suitable state to avoid the discomfort caused by strong direct airflow. Through refined and intelligent control strategies, not only is the user's comfort experience significantly improved, but energy waste is also greatly reduced, realizing intelligent and personalized control of the air conditioner.

[0076] Furthermore, in this embodiment of the invention, the adjustment of the air conditioner's operating state based on target type, stationary coordinates, or trajectory features further includes:

[0077] 701. Record the correction scenario where the air conditioner's actions are corrected, and obtain the trajectory feature data under the correction scenario;

[0078] 702. Store the acquired trajectory feature data in a pre-built sample library;

[0079] For example, when a user notices that the air conditioner has entered standby mode due to misjudging a non-biological target (such as swaying green plants), the user manually turns on the air conditioner. At this time, the corrected scenario is recorded: the trajectory features are periodic motion with a frequency of 1 time / second (i.e., a period of 1Hz), and the energy fluctuation feature value is 4 (the preset no-fluctuation energy value is 5). The system judges it as a non-biological target, but in reality, the user needs the air conditioner to run. At this time, the trajectory feature data of the corrected scenario, i.e., the periodicity of 1Hz, the energy standard deviation of 4, the coordinate sequence, etc., are stored in a pre-built sample library and marked as "misjudged as non-biological - need to be retained" sample.

[0080] 703. When the preset iteration time is reached, optimize the preset energy value without obvious fluctuations and the human motion template based on the trajectory feature data stored in the sample library.

[0081] In this embodiment, the preset iteration time is set to 2:00 AM every day. When 2:00 AM arrives each day, the energy value without significant fluctuations is optimized based on the sample library data. For example, the original preset energy value without fluctuations is 5, but because the target type with an energy standard deviation of 4 is misjudged in multiple correction scenarios, the energy value without fluctuations is adjusted to 6. In addition, the human motion template can also be optimized. For example, a slow movement feature is added, and its displacement change rate is set to 0.3-0.5 m / s to better adapt to the movement patterns of the elderly and children.

[0082] In this embodiment, by recording the user's corrective actions during actual use and analyzing the specific needs and potential problems reflected behind these actions, and based on sample data stored in a pre-built sample library, the templates and thresholds used in the control process are iteratively optimized in a targeted manner. This allows the system to adapt to different home environments, solving the problem of poor adaptability due to fixed parameters in complex scenarios. It also achieves a steady improvement in target recognition accuracy and a continuous enhancement in the level of intelligent air conditioning control. Ultimately, this reduces the frequency of manual intervention by users, bringing them a more convenient, intelligent, and comfortable user experience.

[0083] The air conditioning control method based on millimeter radar waves in the embodiments of the present invention has been described above. The air conditioning control device based on millimeter radar waves in the embodiments of the present invention will be described below. Please refer to [link / reference]. Figure 2 One embodiment of the air conditioning control device based on millimeter radar waves in this invention includes:

[0084] Processing module 801 is used to acquire continuous radar data and perform preprocessing to obtain preprocessed continuous data;

[0085] Extraction module 802 is used to extract trajectory features based on preprocessed continuous data;

[0086] The confirmation module 803 is used to confirm the target type based on the extracted trajectory features, wherein the target type includes human targets and non-biological targets;

[0087] The acquisition module 804 is used to acquire the stationary coordinates of the real person when the target type is a real person target, if the trajectory features indicate that the real person target has changed from a moving state to a stationary state.

[0088] The adjustment module 805 is used to adjust the working status of the air conditioner based on the target type, stationary coordinates, or trajectory characteristics.

[0089] Based on the same ideas as the methods in the above embodiments, the apparatus provided in this application can implement the methods in the above embodiments.

[0090] above Figure 2The air conditioning control device based on millimeter radar waves in the embodiments of the present invention will be described in detail from the perspective of modular functional entities. The air conditioning control device based on millimeter radar waves in the embodiments of the present invention will be described in detail from the perspective of hardware processing.

[0091] Figure 3 This is a schematic diagram of an air conditioning control device 900 based on millimeter-wave radar provided in an embodiment of the present invention. The air conditioning control device 900 based on millimeter-wave radar can vary considerably due to different configurations or performance characteristics. It may include one or more central processing units (CPUs) 910 and a memory 920, and one or more storage media 930 (e.g., one or more mass storage devices) storing application programs 933 or data 932. The memory 920 and storage media 930 may be temporary or persistent storage. The program stored in the storage media 930 may include one or more modules (not shown in the diagram), each module may include a series of instruction operations on the air conditioning control device 900 based on millimeter-wave radar. Furthermore, the processor 910 may be configured to communicate with the storage media 930 and execute the series of instruction operations in the storage media 930 on the air conditioning control device 900 based on millimeter-wave radar to implement the steps of the air conditioning control method based on millimeter-wave radar provided in the above-described method embodiments.

[0092] The millimeter-radar wave-based air conditioning control device 900 may also include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input / output interfaces 960, and / or one or more operating systems 931, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 3 The structure of the air conditioning control device based on millimeter radar waves shown does not constitute a limitation on air conditioning control devices based on millimeter radar waves. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0093] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of an air conditioning control method based on millimeter radar waves.

[0094] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0095] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0096] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An air conditioning control method based on millimeter radar waves, characterized in that, include: Acquire continuous radar data and preprocess it to obtain preprocessed continuous data; Trajectory features are extracted from preprocessed continuous data; The target type is determined based on the extracted trajectory features, and the target type includes human targets and non-biological targets; When the target type is a real person, if the trajectory features indicate that the real person has changed from a moving state to a stationary state, then obtain the stationary coordinates of the real person. Adjust the air conditioner's operating status based on target type, stationary coordinates, or trajectory characteristics.

2. The air conditioning control method based on millimeter radar waves according to claim 1, characterized in that, The process of acquiring continuous radar data and preprocessing it to obtain preprocessed continuous data includes: Acquire continuous radar data, which includes multiple frames of raw radar information collected based on a preset time interval. Each frame of raw radar information includes the X coordinate, Y coordinate, target energy value and its corresponding timestamp. Outlier filtering is performed on continuous radar data to obtain filtered continuous data. A moving average filtering algorithm is used to smooth the noise in the X and Y coordinates of continuous radar data to obtain preprocessed continuous data.

3. The air conditioning control method based on millimeter radar waves according to claim 2, characterized in that, The extraction of trajectory features based on preprocessed continuous data includes: Extract the coordinate data of each frame of raw radar information, and construct a continuous coordinate sequence based on the extracted coordinate data; Calculate the ratio of displacement difference to time difference between two adjacent frames of raw radar information to obtain multiple displacement change rates; Calculate the standard deviation of the target energy value in multiple frames of raw radar information to obtain energy fluctuation characteristics; A time-series analysis of the Y-coordinate in multiple frames of raw radar information is performed to count the number of reciprocating movements within a preset unit time, thereby obtaining the periodicity characteristics of the motion. By integrating continuous coordinate sequences, multiple displacement change rates, energy fluctuation characteristics, and motion periodic characteristics, trajectory features are obtained.

4. The air conditioning control method based on millimeter radar waves according to claim 3, characterized in that, The target type is determined based on the extracted trajectory features. The target type includes human targets and non-biological targets, including: Obtain a pre-constructed human motion template, which includes trajectory continuity features and direction consistency features; When the continuous coordinate sequence and multiple displacement change rates all conform to the human motion template, and the motion periodicity characteristic is that there is no fixed period, the target type is confirmed to be a real human target. When the continuous coordinate sequence and multiple displacement change rates do not conform to the human motion template, and the motion periodicity is a fixed period, and the energy fluctuation characteristics are less than the preset energy value with no obvious fluctuation, the target type is confirmed as a non-biological target.

5. The air conditioning control method based on millimeter radar waves according to claim 2, characterized in that, When the target type is a real person, if the trajectory features indicate that the real person has changed from a moving state to a stationary state, then the stationary coordinates of the real person are obtained, including: When the target type is a real person, if the trajectory features indicate that the real person has changed from a moving state to a stationary state, then obtain the X coordinate of the last valid original radar information before it stops and its corresponding timestamp, and obtain the X coordinates and corresponding timestamps of the three original radar information frames before the last valid original radar information. The average velocity before coming to a stop is calculated based on the X coordinates and corresponding timestamps of the first three frames of raw radar information. Based on the calculated average velocity before coming to a standstill, combined with the X-coordinate of the last frame of valid raw radar information and its corresponding timestamp, a linear prediction algorithm is used to determine the stationary coordinates of the real target.

6. The air conditioning control method based on millimeter radar waves according to claim 1, characterized in that, The adjustment of the air conditioner's operating status based on target type, stationary coordinates, or trajectory characteristics includes: When the target type is a non-biological target, control the air conditioner to enter energy-saving mode or standby mode; When the target type is a real person, if the trajectory features indicate that the real person is in motion, the air conditioner's operating status is adjusted based on the trajectory features; if the trajectory features indicate that the real person is stationary, the air conditioner's operating status is adjusted based on the stationary coordinates.

7. The air conditioning control method based on millimeter radar waves according to claim 4, characterized in that, The process of adjusting the air conditioner's operating state based on target type, stationary coordinates, or trajectory features further includes: Record the correction scenarios where the air conditioner's actions are corrected, and obtain trajectory feature data under the correction scenarios; The acquired trajectory feature data is stored in a pre-built sample library; When the preset iteration time is reached, the preset energy value without significant fluctuations and the human motion template are optimized based on the trajectory feature data stored in the sample library.

8. An air conditioning control device based on millimeter radar waves, characterized in that, include: The processing module is used to acquire continuous radar data and perform preprocessing to obtain preprocessed continuous data. The extraction module is used to extract trajectory features based on preprocessed continuous data; The confirmation module is used to confirm the target type based on the extracted trajectory features, wherein the target type includes human targets and non-biological targets; The acquisition module is used to acquire the stationary coordinates of a real person when the target type is a real person target, if the trajectory features indicate that the real person target has changed from a moving state to a stationary state. The adjustment module is used to adjust the working status of the air conditioner based on the target type, stationary coordinates, or trajectory characteristics.

9. An air conditioning control device based on millimeter radar waves, characterized in that, The air conditioning control device based on millimeter radar waves includes: a memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the millimeter-radar wave-based air conditioning control device to perform the steps of the millimeter-radar wave-based air conditioning control method as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the various steps of the air conditioning control method based on millimeter radar waves as described in any one of claims 1-7.

Images