Sensor fusion based bird tracker energy saving control method and apparatus

Abstract

This invention discloses an energy-saving control method and device for a bird tracker based on sensor fusion. The method uses an accelerometer to collect triaxial acceleration data to identify the bird's behavioral state, activating the heart rate and blood oxygenation sensors only when the bird is stationary and deactivating them during movement. It combines illumination data, location information, and the device's real-time time to determine signal obstruction environments and dynamically adjusts the operating strategies of the positioning and communication modules. Before positioning calculation, it detects the number of satellites in a low-power mode, performing full positioning only when the target is met. Power management independently controls the on / off operation of the positioning, communication, and storage modules, with standby power supplied only to the real-time clock module. The corresponding device includes a main control unit, a sensor data acquisition and tracking unit, and an energy-saving control unit, integrating multiple types of sensors and a power management module. This invention significantly reduces the tracker's ineffective power consumption and extends its battery life in the field through multi-sensor fusion and dynamic energy-saving scheduling, adapting to the long-term monitoring needs of birds.

Description

Technical Field

[0001] This invention relates to the field of satellite tracking technology, and in particular to an energy-saving control method and device for a bird tracker based on sensor fusion. Background Technology

[0002] Long-term wildlife monitoring is a fundamental technical means for ecological research and biodiversity conservation. Among them, birds, as indicator species of environmental change, play an important supporting role in habitat assessment and ecosystem analysis by collecting data on their migration trajectories, behavioral patterns and physiological states.

[0003] In existing publicly available wildlife tracking technologies, early solutions relied on single satellite positioning modules, which could only acquire spatiotemporal trajectory data and were insufficient to meet the needs of multi-dimensional and refined research. To address this issue, multi-sensor fusion monitoring technology has become a well-known development trend in the field. By integrating multiple types of sensors, such as positioning modules, heart rate and blood oxygen sensors, accelerometers, and barometers, it achieves comprehensive acquisition of location, behavioral, physiological, and environmental data, significantly enhancing the scientific research value of the monitoring data.

[0004] However, for free-roaming wild animals, tracking devices must meet the rigid constraints of miniaturization and lightweight design to avoid interfering with their natural behavior, resulting in strict limitations on the battery capacity of the devices. In applications where rechargeable batteries cannot be recycled in the wild, the power consumption of the devices directly determines their service life, thus restricting the continuity and integrity of data collection. Therefore, the contradiction between the functional improvements brought about by multi-sensor fusion and the increased energy consumption has become a recognized core technical bottleneck in this field. How to achieve long-term stable monitoring under energy constraints is a technical problem that urgently needs to be solved. Although multi-sensor fusion technology has been applied in wildlife monitoring, existing publicly available solutions still have the following technical shortcomings in terms of energy management: (1) Energy redundancy caused by lack of sensor collaboration: In the existing equipment, each sensor uses a fixed period or an independent preset strategy to collect data. There is a lack of dynamic collaboration mechanism based on the monitoring scenario and animal behavior status, resulting in ineffective energy consumption during non-critical periods and failing to achieve optimal system-level energy efficiency. (2) Problem of incompatibility between static power consumption strategy and dynamic environment: Traditional energy-saving schemes (such as fixed interval hibernation) are pre-set static management, which cannot respond to the differences in animal day and night behavior, habitat type (forest / open area) and weather changes, resulting in fluctuations in collection and communication conditions, either wasting the opportunity to save energy during hibernation in low activity periods, or losing key data due to insufficient sampling in high activity periods. (3) Lack of context-aware predictive energy saving capability: Existing solutions only respond passively to the current state and do not utilize multi-sensor fusion data, such as behavior recognition results and environmental trends, to predict the future workload of the equipment and cannot achieve forward-looking power consumption optimization. For example, they fail to predictively extend the positioning interval based on the long-term resting behavior of animals. Summary of the Invention

[0005] The purpose of this invention is to provide an energy-saving control method and device for a multi-sensor fusion wildlife tracker. It can dynamically optimize energy consumption allocation under the harsh conditions of limited battery capacity by intelligently sensing animal behavior and environmental context, coordinating sensor scheduling and predictive control, and maximizing the working life of the device while ensuring the quality of key data acquisition.

[0006] To achieve the above objectives, the present invention provides an energy-saving control method for a bird tracker based on sensor fusion, comprising: Step 1: Collect three-axis acceleration data of birds using an accelerometer, identify the behavior of birds based on the three-axis acceleration data, and turn on the heart rate and blood oxygenation physiological sensor only when the birds are stationary, and turn off the heart rate and blood oxygenation physiological sensor when the birds are in slow activity or fast flight. Step 2: Collect ambient light data through the light sensor, combine it with the bird's current location information stored in the positioning module and the device's real-time time, determine whether the tracker is in a signal-obstructed environment, and dynamically adjust the working strategies of the positioning module and the communication module based on the judgment result. Step 3: Before performing the positioning calculation, the control positioning module obtains the number of visible satellites in a low-power satellite search mode. The control positioning module performs a complete positioning calculation only when the number of visible satellites is not less than the preset satellite number threshold. If the number of visible satellites is less than the preset satellite number threshold, the positioning operation is skipped. Step 4: The positioning module, communication module, and storage module are independently controlled by the power management circuit. When the device is in standby mode, only the real-time clock module is powered, achieving microampere-level standby power consumption.

[0007] Furthermore, step 1 specifically includes: Step 1.1: Collect the bird's three-axis acceleration data (x-axis, y-axis, and z-axis) using an accelerometer and transmit the three-axis acceleration data to the main control unit; Step 1.2: The main control unit calculates the composite vector of triaxial acceleration based on the triaxial acceleration data; Step 1.3: Based on the triaxial acceleration composite vector and the preset continuous sampling window judgment rules, identify the bird's behavioral state: If the combined triaxial acceleration vectors corresponding to a preset number of consecutive sampling windows are all no greater than the threshold between being stationary and slowly moving, then the state is determined to be stationary. If the composite vector of triaxial acceleration remains between the critical values ​​of stationary and slow activity and slow activity and fast flight, and simultaneously meets the following two conditions: ① the composite vector of triaxial acceleration corresponding to a preset number of consecutive sampling windows falls within the threshold range; ② the difference between the composite vector of triaxial acceleration of any two adjacent sampling windows is not greater than the preset fluctuation threshold, then it is determined to be a slow activity state. If the composite vector of triaxial acceleration is not less than the critical value between slow activity and fast flight, and the composite vector of triaxial acceleration corresponding to a preset number of consecutive sampling windows satisfies that the difference between any two adjacent sampling windows is not greater than a preset fluctuation threshold, then it is determined to be a fast flight state. The duration of a single sampling window, the preset number of consecutive sampling windows corresponding to each behavior state, the threshold values ​​for stationary and slow-moving activities, the threshold values ​​for slow-moving and fast-flying activities, and the preset fluctuation threshold are all pre-configured.

[0008] Furthermore, step 2 specifically includes: Step 2.1: Collect ambient light data in real time using a light sensor according to a preset sampling interval; Step 2.2, based on the latitude φ and real-time time of the bird's current location information stored in the positioning module, calculate the solar altitude angle h using equation (1): (1) Where δ is the solar declination and ω is the hour angle. This is a multiplication operation; Step 2.3: Based on the solar altitude angle calculated in Step 2.2 and the current daytime state, set the basic illumination threshold; Step 2.4: Compare the ambient light data collected in Step 2.1 with the actual light threshold. The actual light threshold is the product of the base light threshold and the preset seasonal correction coefficient. If the ambient light data is lower than the actual light threshold for the corresponding time period and the duration is not less than the preset duration, it is determined that the tracker is currently in a signal-obscured environment; otherwise, it is determined to be an open environment. Step 2.5: Based on the environmental assessment results from Step 2.4, the main control unit dynamically adjusts the operating strategies of the positioning module and the communication module. If it is an open environment, the control positioning module performs positioning operations at preset positioning intervals, and the control communication module performs data transmission operations at preset communication intervals. If the environment is signal-blocked, the positioning interval of the positioning module will be extended to the first preset extension interval, and the communication interval of the communication module will be extended to the second preset extension interval.

[0009] Furthermore, the light threshold is dynamically adjusted by preset seasonal correction coefficients: the value for summer is in the range of 1.0 to 1.1, the value for spring and autumn is set to 1.0, and the value for winter is in the range of 0.8 to 0.9.

[0010] Furthermore, step 3 specifically includes: Step 3.1: Before the positioning module performs positioning calculation, the main control unit first controls the positioning module to enter the satellite search function only, without performing complete positioning calculation. Step 3.2: The positioning module completes the detection of the number of visible satellites within the preset satellite search time and feeds back the detection results to the main control unit; Step 3.3: The main control unit compares the number of visible satellites with the preset satellite number threshold. If the number of visible satellites is not less than the preset satellite number threshold, the control positioning module will start a complete positioning calculation, complete the positioning within the preset positioning time, and output the location data. If the number of visible satellites is less than the preset satellite number threshold, the positioning module will stop searching for satellites and skip the current positioning operation.

[0011] This invention also provides an energy-saving control device for a bird tracker based on sensor fusion, comprising a main control unit, a sensor data acquisition and tracking unit, and an energy-saving control unit; the sensor data acquisition and tracking unit includes an accelerometer, a heart rate and blood oxygenation physiological sensor, a light sensor, a positioning module, a communication module, and a storage module; the energy-saving control unit includes a power management module; The main control unit collects triaxial acceleration data of birds through an accelerometer, identifies the behavioral state of birds based on the triaxial acceleration data, and turns on the heart rate and blood oxygenation physiological sensor only when the bird is stationary, and turns off the heart rate and blood oxygenation physiological sensor when the bird is in a slow activity state or in a fast flight state. The main control unit collects ambient light data through a light sensor, combines the bird's current location information stored in the positioning module with the device's real-time time, determines whether the device is in a signal-obstructed environment, and dynamically adjusts the working strategies of the positioning module and the communication module based on the determination result. Before controlling the positioning module to perform positioning calculation, the main control unit first controls the positioning module to obtain the number of visible satellites in a low-power satellite search mode. Only when the number of visible satellites is not less than a preset satellite number threshold, the main control unit controls the positioning module to perform a complete positioning calculation; if the number of visible satellites is less than the preset satellite number threshold, the positioning operation is skipped. The power management module independently controls the power supply to the positioning module, communication module, and storage module. When the device is in standby mode, only the real-time clock module is powered, achieving microampere-level standby power consumption.

[0012] Furthermore, the specific process by which the main control unit identifies the bird's behavioral state based on triaxial acceleration data includes: Step 1.1: Collect the bird's three-axis acceleration data (x-axis, y-axis, and z-axis) using an accelerometer and transmit the three-axis acceleration data to the main control unit; Step 1.2: The main control unit calculates the composite vector of triaxial acceleration based on the triaxial acceleration data; Step 1.3: The main control unit identifies the bird's behavioral state based on the triaxial acceleration composite vector and the preset continuous sampling window judgment rules. If the combined triaxial acceleration vectors corresponding to a preset number of consecutive sampling windows are all no greater than the threshold between being stationary and slowly moving, then the state is determined to be stationary. If the composite vector of triaxial acceleration remains between the critical values ​​of stationary and slow activity and slow activity and fast flight, and simultaneously meets the following two conditions: ① the composite vector of triaxial acceleration corresponding to a preset number of consecutive sampling windows falls within the threshold range; ② the difference between the composite vector of triaxial acceleration of any two adjacent sampling windows is not greater than the preset fluctuation threshold, then it is determined to be a slow activity state. If the composite vector of triaxial acceleration is not less than the critical value between slow activity and fast flight, and the composite vector of triaxial acceleration corresponding to a preset number of consecutive sampling windows satisfies that the difference between any two adjacent sampling windows is not greater than the preset fluctuation threshold, then it is determined to be a fast flight state. The duration of a single sampling window, the preset number of consecutive sampling windows corresponding to each behavior state, the threshold values ​​for stationary and slow-moving activities, the threshold values ​​for slow-moving and fast-flying activities, and the preset fluctuation threshold are all pre-configured.

[0013] Furthermore, the working process of the sensor data acquisition and tracking unit includes: Step 2.1: The light sensor collects ambient light data in real time according to a preset sampling interval and transmits it to the main control unit; Step 2.2, the main control unit calculates the solar altitude angle h based on the latitude φ and the device real-time time in the bird's current location information stored in the positioning module, using equation (1): (1) Where δ is the solar declination and ω is the hour angle. This is a multiplication operation; Step 2.3: The main control unit sets the basic illumination threshold based on the solar altitude angle calculated in Step 2.2 and the current daytime state; Step 2.4: The main control unit compares the ambient light data collected in step 2.1 with the actual light threshold. The actual light threshold is the product of the basic light threshold and the preset seasonal correction coefficient. If the ambient light data is lower than the actual light threshold for the corresponding time period and the duration is not less than the preset duration, the device is determined to be in a signal-blocked environment. Otherwise, it is determined to be in an open environment. Step 2.5: Based on the environmental assessment results from Step 2.4, the main control unit dynamically adjusts the operating strategies of the positioning module and the communication module. If it is an open environment, the control positioning module performs positioning operations at preset positioning intervals, and the control communication module performs data transmission operations at preset communication intervals. If the environment is signal-blocked, the positioning interval of the positioning module will be extended to the first preset extension interval, and the communication interval of the communication module will be extended to the second preset extension interval.

[0014] Furthermore, the light threshold is dynamically adjusted by preset seasonal correction coefficients: the value for summer is in the range of 1.0 to 1.1, the value for spring and autumn is set to 1.0, and the value for winter is in the range of 0.8 to 0.9.

[0015] Furthermore, the working process of the energy-saving control unit includes: Step 3.1: Before the positioning module performs positioning calculation, the main control unit controls the positioning module to only start the satellite search function and does not perform the complete positioning calculation. Step 3.2: The positioning module completes the detection of the number of visible satellites within the preset satellite search time and feeds back the detection results to the main control unit; Step 3.3: The main control unit compares the number of visible satellites with the preset satellite number threshold. If the number of visible satellites is not less than the preset satellite number threshold, the control positioning module will start a complete positioning calculation, complete the positioning within the preset positioning time, and output the location data. If the number of visible satellites is less than the preset satellite number threshold, the positioning module will stop searching for satellites and skip the current positioning operation.

[0016] This invention employs a dynamic energy-saving control strategy based on multi-source intelligent sensing, completely changing the traditional fixed-time energy consumption mode of trackers and upgrading from "simple timing" to "contextual awareness." It overcomes the limitations of purely hardware-based energy saving by intelligently determining when to collect effective data at high-value moments and executing high-energy-consuming tasks under high-probability-of-success conditions, significantly reducing overall power consumption and extending fieldwork time, while simultaneously improving the quality of scientific data and the efficiency of storage and communication resource utilization. It constructs a closed-loop intelligent system of "perception-decision-execution," integrating multi-dimensional information such as animal behavior, ambient light, and satellite signals to achieve global adaptive optimization, proactively adapting to complex dynamic environments such as day / night cycles, seasons, and habitat differences. Furthermore, it enhances system robustness and task reliability in extreme environments through proactive energy management, avoiding unnecessary power consumption that could lead to equipment failure and ensuring stable execution of long-term monitoring tasks and complete data acquisition. Attached Figure Description

[0017] Figure 1 This is a structural schematic diagram of the energy-saving control device for a bird tracker based on sensor fusion provided in this embodiment. Detailed Implementation

[0018] In the accompanying drawings, the same or similar reference numerals are used to denote the same or similar elements or elements having the same or similar functions. The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0019] In the description of this invention, the terms "center," "longitudinal," "lateral," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention 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 limiting the scope of protection of this invention.

[0020] like Figure 1 As shown, the sensor fusion-based bird tracker energy-saving control device provided in this embodiment includes a main control unit 1, a sensor data acquisition and tracking unit 2, and an energy-saving control unit 3. Through multi-source sensor data fusion and adaptive control logic, it achieves precise control of device energy consumption, balancing monitoring accuracy and battery life. The composition of each unit will be explained in detail below. The main control unit 1 possesses efficient edge computing capabilities, supporting multi-sensor data collaborative processing and module control logic execution. Its core performance must meet the edge computing requirements of an algorithm response time of no more than 50 milliseconds and a behavior state recognition accuracy of no less than 90%. Specifically, this includes: The MCU (Micro Controller Unit) circuit 11 uses a low-power microcontroller, preferably an STM32L476RG, but MSP430F5529, ESP32-C3, or other existing models with similar functions can also be used. It supports single-cycle multiplication and hardware division, and integrates multiple communication interfaces such as I2C, UART, and USB, enabling direct data interaction with the sensors and functional modules of the sensor data acquisition and tracking unit 2. In sleep mode, the power consumption is no more than 1 microamp, meeting the energy-saving requirements of this invention. The RTC (Real-Time Clock) circuit 12 is a real-time clock module with a timed wake-up function. As the core trigger unit in the low-power standby state of the device, it supports a configurable wake-up cycle from 1 minute to 24 hours, and the wake-up current is no more than 0.5 microamps. The reset circuit 13 is used for abnormal reset of the MCU circuit 11 to ensure the long-term stable operation of the device of the present invention.

[0021] The interface circuit 14 includes UART interface, I2C interface, USB interface, SW interface, SDIO interface, etc., so it can adapt to the communication needs of different sensors and functional modules. The storage circuit 15 includes an FRAM storage unit and an SD card storage unit. The FRAM storage unit is used to store important parameters such as acceleration threshold, satellite number threshold, and seasonal correction coefficient. The SD card storage unit is used to store multi-dimensional data collected by the sensor data acquisition and tracking unit 2. The sensor data acquisition and tracking unit 2 includes multiple sensors and functional modules connected to the main control unit 1. All components are selected for low power consumption, as detailed below: An accelerometer is used to collect bird motion data and transmit the collected data to the main control unit 1, thus accurately outputting triaxial acceleration data for vector synthesis calculations. The preferred accelerometer model is the MPU6050, but the ADXL345, LIS3DH, or other existing models with the same functions can also be used. It integrates a triaxial accelerometer and a triaxial gyroscope, with a configurable acceleration measurement range, a sampling frequency supporting continuous acquisition at 50 Hz, an operating power consumption of no more than 200 microamps, and a sleep power consumption of no more than 1 microamp. The heart rate and blood oxygenation physiological sensor is used to collect physiological parameters of birds. The preferred model is MAX30102, but MAX30205, Si1147, or other existing models with the same function can also be used. It integrates red or infrared LEDs, photodetectors, and signal processing circuits. The operating power consumption is no more than 1 mA, and the sleep power consumption is no more than 0.1 μA. It only starts when the main control unit 1 determines that the bird is in a stationary state, and turns off during slow activity or rapid flight to avoid invalid data collection and energy waste during movement. The light sensor is used to collect ambient light data. The preferred model is BH1750, but TSL2561, OPT3001, or other existing models with the same function can also be used. The measurement range is 0 to 65535 lux, the operating power consumption is no more than 10 microamps, the sleep power consumption is no more than 0.5 microamps, the sampling interval can be dynamically adjusted by the main control unit 1, and the data is transmitted through the I2C interface. The temperature sensor is used to collect ambient temperature data to help determine environmental adaptability. It communicates with the main control unit 1 via an I2C interface and its power consumption is no more than 5 microamps. The positioning module includes a positioning module and a crystal oscillator. The preferred module is the Ublox NEO-6M dual-mode GPS / BeiDou positioning module, but the Ublox NEO-7M, BeiDou BDS-M8M, or other existing models with similar functions can also be used. It supports low-power satellite search mode, with a warm-start time of no more than 25 seconds, operating power consumption of no more than 30 mA, and sleep power consumption of no more than 1 μA. The positioning module transmits location and real-time device time information to the main control unit 1 via a UART interface, with a positioning accuracy of no more than 10 meters. The communication module includes a communication module and a SIM card. The preferred communication module is the BC95-B5 NB-IoT communication module, but other models with the same functionality can also be used. The operating power consumption is no more than 20 mA, and the sleep power consumption is no more than 1 μA. The communication module communicates with the main control unit 1 via a USB interface, supports batch data transfer, and can upload data from the storage module to the server. The antenna switching module includes an antenna and an RF switching switch, enabling time-division multiplexing of the positioning antenna of the positioning module and the communication antenna of the communication module. To reduce the size and weight of the device and make it suitable for small birds to carry, the overall weight of the device must not exceed 10 grams. The response time of the RF switching switch is no more than 1 millisecond, and the insertion loss is no more than 0.5 dB. The energy-saving control unit 3 is the core guarantee for the low-power operation of the device of this invention, and includes a power management module and a storage module, wherein: The power management module, under the control of the main control unit 1, dynamically manages and distributes the power supply, enabling independent power on / off control for the positioning module, communication module, and storage module. Specifically, the power management module includes a power supply, battery protection circuit, DC-DC (Direct Current to Direct Current) step-down circuit, load switching circuit, and auxiliary circuits, among which: The power supply uses a 100mAh miniature rechargeable lithium battery with an operating voltage range of 3.4V to 4.2V. It can be equipped with a miniature solar panel suitable for outdoor lighting conditions. The solar panel is connected to the lithium battery through a charging management chip TP4056 to achieve solar charging. In this way, the device of this invention can last for no less than 360 days in the absence of light. The battery protection circuit uses DW01 and 8205A chips, which have overcharge, over-discharge and overcurrent protection functions to prevent damage to the lithium battery. The first DC-DC step-down circuit can use the TPS62130 chip to provide a separate power supply for the heart rate and blood oxygenation physiological sensor; the second DC-DC step-down circuit can use the AMS1117-3.3 to provide power for the low-power module cluster; both chips support enable signal control, and the power consumption is no more than 0.1 microamps when turned off. The load switch circuit uses the low-power load switch SiP32441 to control the power paths of the positioning module, communication module and SD card storage unit respectively. The main control unit 1 outputs high and low level signals through the GPIO port to control the conduction and de-conduction of the load switch. The auxiliary circuit includes a hardware watchdog circuit, a Hall switch circuit, and a battery voltage divider acquisition circuit.

[0022] The low-power control logic of the power management module includes: The device is in low-power standby mode most of the time. At this time, only the main power bus branch supplies power to the real-time clock module. The enable terminals of the DC-DC buck circuits of the other branches are turned off and the load switches are disconnected. All non-essential power supply units are configured to low power state. The standby current is only at the microamp level, and the basic power consumption is reduced to the nanoamp or microamp level. The storage module combines FRAM memory and SD card storage units. The FRAM memory offers fast read / write speeds and consumes no more than 1 microamp, storing pre-configured parameters such as acceleration thresholds, satellite count thresholds, seasonal correction coefficients, and algorithm parameters. The SD card storage unit supports SPI interface communication, consumes no more than 5 milliamps, and stores acquired acceleration data, heart rate and blood oxygenation data, positioning data, and environmental data, meeting long-term data storage requirements. The energy-saving control method for bird trackers based on sensor fusion provided in this embodiment includes: Step 1, Behavioral Perception and Physiological Data Acquisition and Control: Triaxial acceleration data of the bird is acquired via an accelerometer. Based on this data, the bird's behavioral state is identified. The heart rate and oxygenation sensor is activated only when the bird is stationary, and deactivated during slow activity or rapid flight. Specifically, this includes: Step 1.1: Collect the bird's x-axis, y-axis, and z-axis three-axis acceleration data using an accelerometer and transmit the three-axis acceleration data to the main control unit 1; The main control unit controls the accelerometer to collect the bird's x-axis, y-axis, and z-axis three-axis acceleration data at a frequency of 50Hz and transmits the collected raw data to the main control unit for processing.

[0023] Step 1.2: The main control unit 1 calculates the composite vector of the three-axis acceleration based on the aforementioned three-axis acceleration data. Specifically, when the main control unit 1 identifies the bird's behavioral state, it calculates the composite vector of the three-axis acceleration based on the instantaneous sampled values ​​of the bird's x-axis, y-axis, and z-axis accelerations collected by the accelerometer. The method for calculating the composite vector is as follows: the composite vector can be obtained by first squaring the instantaneous sampled values ​​of the x-axis, y-axis, and z-axis accelerations respectively, then summing them, and finally taking the square root of the sum.

[0024] Step 1.3: Based on the triaxial acceleration composite vector and a preset number of continuous sampling windows, identify the bird's behavioral state: Two acceleration thresholds were set: one for stationary and slow-moving conditions, and the other for slow-moving and fast-flying conditions. In this implementation, the threshold for stationary and slow-moving conditions was set to 0.1g, and the threshold for slow-moving and fast-flying conditions was set to 0.5g. These two specific values ​​were obtained by referencing existing technologies and combining them with data from our research team's previous work: 1.6 million raw triaxial acceleration data points from 160 bean geese and swan geese between October 2025 and January 2026.

[0025] Temporal correlation analysis is performed on the triaxial acceleration composite vector data over a continuous time period (e.g., 3 seconds) to avoid misjudgment based on data from a single moment. First, a single sampling window is used as the basic unit for motion state determination. The duration of a single sampling window, the preset number of consecutive sampling windows corresponding to each behavioral state, the thresholds for stationary and slow-moving activities, the thresholds for slow-moving and fast-flying activities, and the preset fluctuation thresholds are all pre-configured. The duration of a single window is comprehensively determined based on the sensor sampling frequency, the inertia of motion state changes, and real-time requirements. This ensures the timeliness and accuracy of motion state identification while filtering out instantaneous noise interference. In this embodiment, the duration of a single window is 0.1 seconds.

[0026] If the triaxial acceleration composite vectors corresponding to the preset number of continuous sampling windows are all not greater than the critical value between being stationary and slowly moving, then it is determined to be a stationary state; in this embodiment, the preset number is 10, which can be adjusted according to actual needs.

[0027] If the triaxial acceleration composite vector remains between the critical values ​​of stationary and slow-moving states and the critical values ​​of slow-moving and fast-flying states, and simultaneously meets the following two conditions: ① the triaxial acceleration composite vectors corresponding to a preset number of consecutive sampling windows all fall within the threshold range; ② the difference between the triaxial acceleration composite vectors of any two adjacent sampling windows is not greater than a preset fluctuation threshold, then the state is determined to be slow-moving. In this embodiment, the preset number is 3, which can be adjusted according to actual needs.

[0028] If the composite vector of triaxial acceleration is not less than the critical value between slow activity and fast flight, and the composite vector of triaxial acceleration corresponding to a preset number of continuous sampling windows satisfies that the difference between any two adjacent sampling windows is not greater than a preset fluctuation threshold, then it is determined to be a fast flight state; in this embodiment, the preset number is 2, which can be adjusted according to actual needs.

[0029] Step 1.4: The main control unit controls the start and stop of the heart rate and blood oxygenation physiological sensor based on the behavior state recognition results: only when the bird is determined to be in a stationary state, an enable signal is sent through the I2C interface to turn on the heart rate and blood oxygenation physiological sensor and collect the bird's heart rate and blood oxygen data; if the bird is determined to be in a slow activity state or a fast flight state, a shutdown signal is immediately sent to cut off the power supply to the heart rate and blood oxygenation physiological sensor and stop data collection.

[0030] Step 2, Environmental Perception and Module Strategy Adjustment: Ambient lighting data is collected using a light sensor. Combined with the bird's current location information stored in the positioning module and the device's real-time time, it is determined whether the tracker is in a signal-obstructed environment. Based on the determination result, the operating strategies of the positioning and communication modules are dynamically adjusted. Specifically, this includes: Step 2.1: Collect ambient light data in real time using a light sensor according to a preset sampling interval; the sampling interval is preset, for example, 1 minute, and the data acquisition accuracy is preset to 1 lux.

[0031] Step 2.2, based on the latitude φ and the device's real-time time (year, month, day, hour, minute) from the bird's current location information stored in the positioning module, calculate the solar altitude angle h using equation (1): (1) Where δ is the solar declination, determined by the date, such as the solar declination on the summer solstice being approximately 23.5° and the solar declination on the winter solstice being approximately -23.5°; ω is the hour angle, determined by longitude and time, with the hour angle at noon being 0°; This refers to multiplication operations.

[0032] Step 2.3: Based on the solar altitude angle calculated in Step 2.2 and the current daytime state, set the basic illumination threshold in lux. The specific setting rules are as follows: (1) The daytime state is daytime and the solar altitude angle is greater than 0°, and the corresponding basic illumination threshold is set to not less than 300; (2) The daytime state is dusk or dawn, and the solar altitude angle is not greater than 0° and greater than -6°, and the corresponding basic illumination threshold value is in the range of 50 to 300. (3) The daytime state is nighttime, and the solar altitude angle is not greater than -6°, and the corresponding basic illumination threshold is set to less than 50; Step 2.4 compares the ambient light data collected in Step 2.1 with the actual light threshold, which is the product of the base light threshold and a preset seasonal correction coefficient. To adapt to different seasons where "the same sun altitude angle results in different perceived illuminance," the light threshold is dynamically adjusted using a preset seasonal correction coefficient to improve the accuracy of environmental judgment. In this embodiment, the value for summer is between 1.0 and 1.1, the values ​​for spring and autumn are both set to 1.0, and the values ​​for winter are between 0.8 and 0.9. If the ambient light data is lower than the actual light threshold for the corresponding time period, and the duration is not less than a preset duration (e.g., 5 minutes), the tracker is determined to be in a signal-obstructed environment such as a dense forest or canyon; otherwise, it is determined to be in an open environment.

[0033] Step 2.5: Based on the environmental assessment results from Step 2.4, the main control unit 1 dynamically adjusts the operating strategies of the positioning module and the communication module. In an open environment, the positioning module performs positioning operations at preset positioning intervals, and the communication module performs data transmission operations at preset communication intervals. In this embodiment, the preset positioning interval is set to 30 minutes / time, and the preset communication interval is set to 1 hour / time, which can be adjusted according to actual needs.

[0034] If the signal is blocked, the positioning interval of the positioning module is extended to the first preset extension interval, and the communication interval of the communication module is extended to the second preset extension interval. In this embodiment, the first preset extension interval is set to 2 hours / time, and the second preset extension interval is set to 4 hours / time to avoid power consumption waste caused by frequent attempts to locate and communicate in poor signal environments.

[0035] Step 3, Satellite Signal Prediction and Optimal Positioning: Before performing the positioning calculation, the control positioning module acquires the number of visible satellites in a low-power satellite search mode. The control positioning module performs a complete positioning calculation only if the number of visible satellites is not less than a preset satellite number threshold; if the number of visible satellites is less than the preset satellite number threshold, the positioning operation is skipped. Specifically, this includes: Step 3.1: Before the positioning module performs positioning calculation, the main control unit 1 controls the positioning module to enter a low-power satellite search mode. In this mode, the positioning module only starts the satellite search function and does not perform a complete positioning calculation. The power consumption is no more than 15 mA.

[0036] Step 3.2: The positioning module completes the detection of the number of visible satellites within the preset satellite search time and feeds back the detection results to the main control unit 1; in this embodiment, the preset satellite search time is set to 5 seconds, and the positioning module completes the detection of the number of visible satellites within 5 seconds.

[0037] Step 3.3: Set a preset satellite number threshold, for example, 5 satellites can meet the requirement of a positioning accuracy of not less than 10 meters, which can be adjusted according to the positioning accuracy requirements; the main control unit 1 compares the number of detected visible satellites with the preset satellite number threshold: If the number of visible satellites is not less than the preset satellite number threshold, the control positioning module starts the complete positioning calculation, completes the positioning within the preset positioning time and outputs the location data; in this embodiment, the preset positioning time is set to 25 seconds.

[0038] If the number of visible satellites is less than the preset satellite number threshold, the positioning module will stop searching for satellites and skip the current positioning operation to avoid unnecessary power consumption.

[0039] Step 4, Hardware-based energy saving: The power management module independently controls the power supply to the positioning module, communication module, and storage module. When the device is in standby mode, only the real-time clock module is powered, achieving microampere-level standby power consumption. Through the above control methods, the operating time of the high-power modules in the equipment is precisely reduced. Combined with specific comparative experimental data, the energy-saving effect of this solution is significant. No-light battery life test: Powered by a 100mAh micro lithium battery, under simulated natural bird activity conditions, with an average of 2 hours of flight, 4 hours of slow activity, and 18 hours of stillness per day, at an ambient temperature of 25℃ and a light intensity of 100 lux, the test results for three groups of samples (10 devices of this invention per group) show: Existing traditional single GPS trackers have an average operating life of 18.6 days. Existing timed sleep trackers have a sleep interval of 30 minutes and an average working life of 42.3 days. The device of this invention has an average working life of 78.9 days, which is 4.24 times longer than that of a traditional single GPS tracker and 1.82 times longer than that of a timed sleep tracker. Solar-assisted endurance test: When equipped with a micro solar panel, the device of this invention can last for more than 360 days without sunlight, which is 3 times longer than existing technologies, that is, more than 240 days longer, far exceeding the level of similar products. It should be noted that the specific parameters involved in this embodiment (such as sampling frequency, threshold value, working interval, module model, etc.) are only preferred solutions and are not limitations on the technical solution. Those skilled in the art can make adjustments according to the actual application scenario. For example, the sampling frequency of the accelerometer can be adjusted to 20 to 100 Hz, the satellite number threshold can be adjusted to 3 to 8 satellites according to the positioning accuracy requirements, and the working interval of the positioning module can be adjusted to 15 minutes to 4 hours, etc. All these adjustments fall within the protection scope of this invention.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Those skilled in the art should understand that modifications can be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A sensor fusion-based energy-saving control method for bird trackers, characterized in that, include: Step 1: Collect three-axis acceleration data of birds using an accelerometer, identify the behavior of birds based on the three-axis acceleration data, and turn on the heart rate and blood oxygenation physiological sensor only when the birds are stationary, and turn off the heart rate and blood oxygenation physiological sensor when the birds are in slow activity or fast flight. Step 2: Collect ambient light data through the light sensor, combine it with the bird's current location information stored in the positioning module and the device's real-time time, determine whether the tracker is in a signal-obstructed environment, and dynamically adjust the working strategies of the positioning module and the communication module based on the judgment result. Step 3: Before performing the positioning calculation, the control positioning module obtains the number of visible satellites in a low-power satellite search mode. The control positioning module performs a complete positioning calculation only when the number of visible satellites is not less than the preset satellite number threshold. If the number of visible satellites is less than the preset satellite number threshold, the positioning operation is skipped. Step 4: The positioning module, communication module, and storage module are independently controlled by the power management circuit. When the device is in standby mode, only the real-time clock module is powered, achieving microampere-level standby power consumption.

2. The energy-saving control method for a bird tracker based on sensor fusion according to claim 1, characterized in that, Step 1 specifically includes: Step 1.1: Collect the bird's three-axis acceleration data (x-axis, y-axis, and z-axis) using an accelerometer and transmit the three-axis acceleration data to the main control unit; Step 1.2: The main control unit calculates the composite vector of triaxial acceleration based on the triaxial acceleration data; Step 1.3: Based on the triaxial acceleration composite vector and the preset continuous sampling window judgment rules, identify the bird's behavioral state: If the combined triaxial acceleration vectors corresponding to a preset number of consecutive sampling windows are all no greater than the threshold between being stationary and slowly moving, then the state is determined to be stationary. If the composite vector of triaxial acceleration remains between the critical values ​​of stationary and slow activity and slow activity and fast flight, and simultaneously meets the following two conditions: ① the composite vector of triaxial acceleration corresponding to a preset number of consecutive sampling windows falls within the threshold range; ② the difference between the composite vector of triaxial acceleration of any two adjacent sampling windows is not greater than the preset fluctuation threshold, then it is determined to be a slow activity state. If the composite vector of triaxial acceleration is not less than the critical value between slow activity and fast flight, and the composite vector of triaxial acceleration corresponding to a preset number of consecutive sampling windows satisfies that the difference between any two adjacent sampling windows is not greater than a preset fluctuation threshold, then it is determined to be a fast flight state. The duration of a single sampling window, the preset number of consecutive sampling windows corresponding to each behavior state, the threshold values ​​for stationary and slow-moving activities, the threshold values ​​for slow-moving and fast-flying activities, and the preset fluctuation threshold are all pre-configured.

3. The energy-saving control method for a bird tracker based on sensor fusion according to claim 1, characterized in that, Step 2 specifically includes: Step 2.1: Collect ambient light data in real time using a light sensor according to a preset sampling interval; Step 2.2, based on the latitude in the bird's current location information stored in the positioning module. Based on the real-time data of the equipment, the solar altitude angle h is calculated using equation (1): （1） in, The solar declination, For the hour angle, This is a multiplication operation; Step 2.3: Based on the solar altitude angle calculated in Step 2.2 and the current daytime state, set the basic illumination threshold; Step 2.4: Compare the ambient light data collected in Step 2.1 with the actual light threshold. The actual light threshold is the product of the base light threshold and the preset seasonal correction coefficient. If the ambient light data is lower than the actual light threshold for the corresponding time period and the duration is not less than the preset duration, it is determined that the tracker is currently in a signal-obscured environment; otherwise, it is determined to be an open environment. Step 2.5: Based on the environmental assessment results from Step 2.4, the main control unit dynamically adjusts the operating strategies of the positioning module and the communication module. If it is an open environment, the control positioning module performs positioning operations at preset positioning intervals, and the control communication module performs data transmission operations at preset communication intervals. If the environment is signal-blocked, the positioning interval of the positioning module will be extended to the first preset extension interval, and the communication interval of the communication module will be extended to the second preset extension interval.

4. The energy-saving control method for a bird tracker based on sensor fusion according to claim 3, characterized in that, The light threshold is dynamically adjusted by a preset seasonal correction factor: the value for summer is in the range of 1.0 to 1.1, the value for spring and autumn is set to 1.0, and the value for winter is in the range of 0.8 to 0.

9.

5. The energy-saving control method for a bird tracker based on sensor fusion according to any one of claims 1-4, characterized in that, Step 3 specifically includes: Step 3.1: Before the positioning module performs positioning calculation, the main control unit first controls the positioning module to enter the satellite search function only, without performing complete positioning calculation. Step 3.2: The positioning module completes the detection of the number of visible satellites within the preset satellite search time and feeds back the detection results to the main control unit; Step 3.3: The main control unit compares the number of visible satellites with the preset satellite number threshold. If the number of visible satellites is not less than the preset satellite number threshold, the control positioning module will start a complete positioning calculation, complete the positioning within the preset positioning time, and output the location data. If the number of visible satellites is less than the preset satellite number threshold, the positioning module will stop searching for satellites and skip the current positioning operation.

6. An energy-saving control device for a bird tracker based on sensor fusion, characterized in that, It includes a main control unit, a sensor data acquisition and tracking unit, and an energy-saving control unit; the sensor data acquisition and tracking unit includes an accelerometer, a heart rate and blood oxygenation physiological sensor, a light sensor, a positioning module, a communication module, and a storage module; the energy-saving control unit includes a power management module; The main control unit collects triaxial acceleration data of birds through an accelerometer, identifies the behavioral state of birds based on the triaxial acceleration data, and turns on the heart rate and blood oxygenation physiological sensor only when the bird is stationary, and turns off the heart rate and blood oxygenation physiological sensor when the bird is in a slow activity state or in a fast flight state. The main control unit collects ambient light data through a light sensor, combines the bird's current location information stored in the positioning module with the device's real-time time, determines whether the device is in a signal-obstructed environment, and dynamically adjusts the working strategies of the positioning module and the communication module based on the determination result. Before controlling the positioning module to perform positioning calculation, the main control unit first controls the positioning module to obtain the number of visible satellites in a low-power satellite search mode. Only when the number of visible satellites is not less than a preset satellite number threshold, the main control unit controls the positioning module to perform a complete positioning calculation; if the number of visible satellites is less than the preset satellite number threshold, the positioning operation is skipped. The power management module independently controls the power supply to the positioning module, communication module, and storage module. When the device is in standby mode, only the real-time clock module is powered, achieving microampere-level standby power consumption.

7. The energy-saving control device for a bird tracker based on sensor fusion according to claim 6, characterized in that, The specific process by which the main control unit identifies bird behavior based on triaxial acceleration data includes: Step 1.1: Collect the bird's three-axis acceleration data (x-axis, y-axis, and z-axis) using an accelerometer and transmit the three-axis acceleration data to the main control unit; Step 1.2: The main control unit calculates the composite vector of triaxial acceleration based on the triaxial acceleration data; Step 1.3: The main control unit identifies the bird's behavioral state based on the triaxial acceleration composite vector and the preset continuous sampling window judgment rules. If the combined triaxial acceleration vectors corresponding to a preset number of consecutive sampling windows are all no greater than the threshold between being stationary and slowly moving, then the state is determined to be stationary. If the composite vector of triaxial acceleration remains between the critical values ​​of stationary and slow activity and slow activity and fast flight, and simultaneously meets the following two conditions: ① the composite vector of triaxial acceleration corresponding to a preset number of consecutive sampling windows falls within the threshold range; ② the difference between the composite vector of triaxial acceleration of any two adjacent sampling windows is not greater than the preset fluctuation threshold, then it is determined to be a slow activity state. If the composite vector of triaxial acceleration is not less than the critical value between slow activity and fast flight, and the composite vector of triaxial acceleration corresponding to a preset number of consecutive sampling windows satisfies that the difference between any two adjacent sampling windows is not greater than the preset fluctuation threshold, then it is determined to be a fast flight state. The duration of a single sampling window, the preset number of consecutive sampling windows corresponding to each behavior state, the threshold values ​​for stationary and slow-moving activities, the threshold values ​​for slow-moving and fast-flying activities, and the preset fluctuation threshold are all pre-configured.

8. The energy-saving control device for a bird tracker based on sensor fusion according to claim 6, characterized in that, The working process of the sensor data acquisition and tracking unit includes: Step 2.1: The light sensor collects ambient light data in real time according to a preset sampling interval and transmits it to the main control unit; Step 2.2, the main control unit uses the latitude from the bird's current location information stored in the positioning module. Based on the real-time data of the equipment, the solar altitude angle h is calculated using equation (1): （1） in, The solar declination, For the hour angle, This is a multiplication operation; Step 2.3: The main control unit sets the basic illumination threshold based on the solar altitude angle calculated in Step 2.2 and the current daytime state; Step 2.4: The main control unit compares the ambient light data collected in step 2.1 with the actual light threshold. The actual light threshold is the product of the basic light threshold and the preset seasonal correction coefficient. If the ambient light data is lower than the actual light threshold for the corresponding time period and the duration is not less than the preset duration, the device is determined to be in a signal-blocked environment. Otherwise, it is determined to be in an open environment. Step 2.5: Based on the environmental assessment results from Step 2.4, the main control unit dynamically adjusts the operating strategies of the positioning module and the communication module. If it is an open environment, the control positioning module performs positioning operations at preset positioning intervals, and the control communication module performs data transmission operations at preset communication intervals. If the environment is signal-blocked, the positioning interval of the positioning module will be extended to the first preset extension interval, and the communication interval of the communication module will be extended to the second preset extension interval.

9. The energy-saving control device for a bird tracker based on sensor fusion according to claim 8, characterized in that, The light threshold is dynamically adjusted by a preset seasonal correction factor: the value for summer is in the range of 1.0 to 1.1, the value for spring and autumn is set to 1.0, and the value for winter is in the range of 0.8 to 0.

9.

10. The energy-saving control device for a bird tracker based on sensor fusion according to any one of claims 6-9, characterized in that, The working process of the energy-saving control unit includes: Step 3.1: Before the positioning module performs positioning calculation, the main control unit controls the positioning module to only start the satellite search function and does not perform the complete positioning calculation. Step 3.2: The positioning module completes the detection of the number of visible satellites within the preset satellite search time and feeds back the detection results to the main control unit; Step 3.3: The main control unit compares the number of visible satellites with the preset satellite number threshold. If the number of visible satellites is not less than the preset satellite number threshold, the control positioning module will start a complete positioning calculation, complete the positioning within the preset positioning time, and output the location data. If the number of visible satellites is less than the preset satellite number threshold, the positioning module will stop searching for satellites and skip the current positioning operation.

Images