Inertial sensing boosts precision and productivity in smart farming

The ADIS16576 from ADI is a MEMS IMU that represents a significant breakthrough in both functional integration and core sensor performance, with particularly notable improvements in Vibration Rectification Error (VRE) - a 10x improvement for the gyroscope and a 50x improvement for the accelerometer. The fundamental function of a MEMS IMU is to provide triaxial angular rate sensing around three mutually orthogonal axes (roll, pitch, yaw) and triaxial linear acceleration sensing along the same three axes.
 
The accelerometer provides mean (or static) angle estimation, while integrated gyroscope measurements provide real-time angular displacement. The system processor combine these two angle estimation sources to generate reliable feedback control information for GNC or pointing control systems. Operating in this manner, an accelerometer VRE of 1.3 mg under 4 g of vibration means the GNC platform can maintain an attitude angle better than 0.1° without assistance from any other sensing function. This is particularly beneficial for UAVs, especially those whose vibration levels vary significantly with thrust.
 
In gyroscopes, VRE can cause rapid and persistent changes in bias, leading to erroneous motion correction and, in the worst case, platform instability. Previous-generation devices could exhibit VRE responses exceeding 300°/h under 8 g rms, whereas the ADIS16576 offers a response of 12°/h, significantly reducing the estimation/correction burden on other system sensing modes. One of the most important functional improvements of this MEMS IMU is its scalable external synchronization. By incorporating a user-programmable clock scaling function, system developers can now drive 4000 Hz IMU data sampling using slower system-level references, such as GPS or video sync. This achieves tight coupling with Pulse Per Second (PPS) or perception sensing references while preserving the various digital processing options afforded by higher data sampling rates.
 
Take an autonomous vehicle platform as an example; it can use a 20 Hz GPS reference and a 200x scale factor to generate an internal sampling rate of 4000 Hz. Furthermore, the system can employ an on-board decimation filter to reduce the output data rate by a factor of 20x (to 200 Hz). In more dynamic scenarios, such as a crop inspection drone operating in windy conditions, the system processor may need to read and process data at the maximum sampling rate to ensure stability and maneuverability.

Inertial sensors also play a pivotal role in IoT systems, which can be used for continuous monitoring of animal location and physiological conditions. Typical embodiments include tags affixed to the ear, tail, or body, or smart collars worn around the neck. These tags not only help manage herd location but, more importantly, continuously provide animal health data such as activity levels, feeding times, and respiration rates. Furthermore, newer tags possess the capability to track heart rate and other vital signs. Neck-worn collars have become invaluable tools for detecting estrus, rumination, lameness, and other conditions in cattle. One of the core requirements for such IoT systems is low power consumption, as maintaining batteries (rechargeable or primary) for large herds is a tedious and challenging task.
 
ADI's ADXL366 offers exceptional low-power advantages. This triaxial accelerometer incorporates internal voltage regulation, allowing it to be connected directly to a battery. Its minimum operating voltage is 1.1 V, and it can provide motion data at 100 Hz with a power consumption of only about 1 µW. This energy consumption level is lower than the self-discharge rate of a coin cell battery. When used in a neck-worn collar, this accelerometer can adjust between low-power and low-noise modes, providing a minimum signal between 3 mg and 8 mg rms, sufficient to distinguish chewing, rumination, and respiration rate (RR). The ADXL380 offers enhanced vital sign monitoring capability, with operational noise levels nearly two orders of magnitude lower across a 4 kHz bandwidth. For an equivalent bandwidth comparison at 200 Hz, the equivalent noise for this accelerometer is 0.4 mg rms. The high Signal-to-Noise Ratio (SNR) combined with wide bandwidth allows this triaxial accelerometer to function as a "stethoscope," collecting heart rate information via ballistocardiograms or capturing various noises associated with breathing, digestion, and other physiological functions.
 
Another core capability of ultra-low-power inertial sensors is supporting system-level power management for IoT nodes. The ADXL366 provides a dedicated wake-up mode that can issue interrupts based on detected motion profiles to manage the power cycle of electronic systems. The accelerometer offers a rich set of programmable parameters to configure the desired motion profile and, most importantly, to wake and sample at full bandwidth. This capability is crucial for avoiding aliasing and false detection. In wake-up mode, the ADXL366 consumes an astonishingly low 180 nA. Leveraging this feature, high-power sensors, radios, and other components can be powered down when not needed, thereby extending the useful lifetime of the sensor node.

Finally, the discussion turns to the integration of inertial sensing and AI analytics, aimed at achieving predictive maintenance in smart farming. As modern farms expand in scale, they increasingly rely on high capital expense machinery for production. Such equipment must not only operate with precision but also withstand the harsh environments and intensive demands of seasonal farming tasks. Equipment failure during the short planting or harvesting season can lead to significant economic losses.
 
For example, precision equipment like seeders or harvesters often must operate in adverse conditions such as rain, wind, dust, mud, and rock fragments . In these environments, changes in key vibration signatures can be used to predict equipment issues in advance, allowing for maintenance during periods that minimally impact peak-demand production times. Mechanical vibration analysis (similar to vital sign monitoring in livestock) can precisely identify the failure modes and timing of various problems in mechanical elements, such as bearing faults, axle misalignment, imbalance, looseness, gear faults, and other issues. Consider a bearing defect, such as spalling or deviation from an ideal spherical shape. Each time this defect contacts the machine surface, it creates an impulse on the platform, triggering a complex vibration response containing both fundamental and broadband content.
 
Therefore, to achieve predictive maintenance, accelerometers need to meet three important performance criteria: low noise (for early prediction), high bandwidth (to detect all spectral contents and support fault classification), and a sufficiently high measurement range. However, the last requirement is often overlooked because the magnitude of acceleration is proportional to the square of the frequency (ω²). If not considered, high-frequency spectral content can saturate the sensor. ADI's new ADXL382 triaxial digital accelerometer in a compact package meets all three requirements. This product supports measurement ranges of ±15 g, ±30 g, and ±60 g, offers an 8 kHz bandwidth, and features ultra-low noise of less than 55 μg/√Hz.
 
The ADXL382 boasts industry-leading noise performance, enabling high-precision applications with minimal calibration. Its low noise and low power characteristics allow for accurate measurement of audio signals or heart sounds even in high-vibration environments. Furthermore, the ADXL382 enables true system-level performance, including a built-in micropower temperature sensor, single-, double-, or triple-tap detection, and a state machine to prevent false triggering.

Automation and technological advancement in agriculture are key forces in addressing critical global challenges such as food security, labor shortages, and environmental sustainability. By adopting innovations like AI, robotics, and precision farming, agricultural production can enhance efficiency, reduce costs, and ensure a more sustainable future for food production. Within the agricultural technology ecosystem, inertial sensors provide core sensing capabilities and play a pivotal role. However, careful consideration must be given when selecting sensors to ensure their performance and functionality align with the specific application. The relevant solutions offered by ADI will accelerate the development of smart farming and stand as one of the optimal choices for related applications.