Interview Questions for Integrated Navigation System (INS)

An Integrated Navigation System (INS) is a self-contained navigation system that uses inertial sensors, primarily accelerometers and gyroscopes, to measure the acceleration and rotation of a vehicle. It then uses these measurements, along with initial position and orientation data, to continuously estimate its current position, velocity, and attitude. Think of it like a sophisticated internal compass and odometer combined, that works even without external signals like GPS.

The fundamental principle is based on the laws of Newtonian mechanics. By integrating the measured acceleration twice, we get the change in velocity and then the change in position. Similarly, integrating the angular rate from gyroscopes gives us the change in orientation. However, this process is inherently prone to errors that accumulate over time, a phenomenon known as drift, which requires sophisticated error modelling and compensation techniques.

INS architectures can be broadly classified into several types based on the sensor configuration and processing:

• Strapdown INS: This is the most common type where the inertial sensors are rigidly mounted to the vehicle’s body. All calculations are made with respect to the body frame, requiring frequent coordinate transformations. Think of it like a fixed camera observing the motion.
• Gimballed INS: In this architecture, the inertial sensors are mounted on gimbals, which mechanically isolate them from the vehicle’s motion. While offering advantages in terms of reduced drift, they are more complex, bulky, and susceptible to mechanical wear and tear. It’s like having a camera stabilized on a tripod.
• Loosely Coupled INS/GNSS: This architecture integrates data from the INS and a Global Navigation Satellite System (GNSS) separately. The INS provides short-term, high-rate position, velocity and attitude, while the GNSS provides long-term, absolute position information. GNSS corrects for INS drift. It’s like having two independent sources corroborating each other.
• Tightly Coupled INS/GNSS: Here, the INS and GNSS data are fused together in a tightly coupled Kalman filter, using a more sophisticated algorithm. This often leads to better accuracy than loosely coupled integration, as it handles the sensor data synergistically. It’s like having two sources directly informing each other.

The key sensors in an INS are:

• Accelerometers: These measure the specific force acting on the vehicle, which is the vector sum of the gravitational acceleration and the linear acceleration. They are essential for determining velocity and position.
• Gyroscopes: These measure the angular rate of rotation of the vehicle, providing information about its orientation (attitude). Accurate orientation estimation is critical for correctly interpreting accelerometer measurements.
• Magnetometers (optional): While not strictly an inertial sensor, magnetometers measure the Earth’s magnetic field, which can be used to help determine heading and reduce heading errors.

High-quality, low-noise sensors are crucial for achieving high accuracy in an INS. Modern INS often use MEMS (Microelectromechanical Systems) based sensors, which are miniaturized, cost-effective, and relatively robust.

Kalman filtering is a powerful algorithm used in INS to optimally estimate the vehicle’s state (position, velocity, attitude, and sensor biases) by fusing noisy sensor data with a dynamic model of the vehicle’s motion. It leverages the statistical properties of the sensor noise and the system dynamics to produce a best-estimate of the state.

The Kalman filter predicts the state in the next time step based on the previous state and the input measurements. It then updates this prediction when new sensor measurements become available, weighing the predicted state and the new measurements based on their respective uncertainties. This process is repeated iteratively to obtain a continually refined estimate of the navigation solution. Kalman filtering is especially effective in handling the accumulation of errors inherent in INS.

Sensor biases and drifts are significant error sources in INS. Biases represent constant offsets in sensor readings, while drifts refer to slow, time-varying changes in these offsets. These errors are typically modeled as stochastic processes (random walks) within the Kalman filter.

The Kalman filter estimates these biases along with the other state variables. By estimating and compensating for these biases, the filter effectively minimizes their impact on the navigation solution. This is a key aspect of achieving acceptable accuracy over extended periods. Regular calibration also helps to minimize these errors.

Various error sources affect the accuracy of an INS:

• Sensor Noise: Random fluctuations in sensor readings.
• Sensor Biases and Drifts: As discussed above.
• Misalignment Errors: Imperfect alignment of sensors with respect to the vehicle’s coordinate frame.
• Scale Factor Errors: Inaccuracies in the scaling of sensor readings.
• Earth Rotation Effects: The Earth’s rotation affects the apparent acceleration and must be accounted for.
• Gravitational Effects: Variations in the Earth’s gravitational field can lead to errors in position estimation.
• Computational Errors: Numerical errors during integration and filtering.

The severity of these errors varies depending on the quality of the sensors and the sophistication of the algorithms used.

INS calibration is a crucial step to improve accuracy and reduce errors. It involves determining the sensor biases, scale factors, and misalignment errors. Calibration is typically done in a controlled environment, often a stationary test setup or during a controlled manoeuvre.

The process usually involves:

• Static Calibration: The INS is held stationary for a period, allowing the Kalman filter to estimate the sensor biases. This involves careful leveling to minimize the effects of gravity.
• Dynamic Calibration: The INS is subjected to known maneuvers (e.g., rotations, accelerations) which helps in estimating scale factors and misalignment errors. This step might involve specialized equipment and algorithms.
• In-flight Calibration: For systems which allow it, calibration parameters can be refined using GPS data or other aiding sources, often as part of the Kalman filter updates. This is a way to periodically adjust calibration over time.

The specific calibration procedures depend on the type of INS and the level of accuracy required.

Sensor fusion in an Inertial Navigation System (INS) is the process of combining data from multiple sensors to obtain a more accurate and reliable estimate of the system’s position, velocity, and attitude than could be achieved using any single sensor alone. Think of it like having multiple witnesses describing an event – each witness might have a slightly different perspective, but by combining their accounts, you get a much clearer picture.

INS typically uses inertial measurement units (IMUs) containing accelerometers and gyroscopes. However, these sensors suffer from drift – their measurements accumulate errors over time. Sensor fusion integrates IMU data with other sensor data, such as GPS, to correct for this drift and improve overall accuracy. This integration usually involves sophisticated algorithms like Kalman filters, which mathematically weigh the data from different sources based on their reliability and accuracy.

For example, a Kalman filter might give more weight to GPS data when the system is outdoors and GPS reception is strong, and more weight to IMU data when GPS is unavailable (e.g., indoors or in a canyon).

The choice of sensor combination in an INS significantly impacts performance. Different combinations offer various advantages and disadvantages:

• GPS + IMU: This is the most common combination. GPS provides absolute position, while IMU provides high-rate measurements of acceleration and rotation. Advantages include high accuracy and global coverage. Disadvantages include vulnerability to signal blockage (tunnels, urban canyons) and susceptibility to jamming or spoofing.
• IMU only: Simpler and less expensive, but highly susceptible to drift. Suitable only for short-duration applications where high accuracy isn’t critical, such as in some robotics applications.
• GPS + IMU + Barometer: Adding a barometer provides altitude information, improving vertical accuracy and reducing reliance on GPS alone for altitude. Useful for applications such as UAV navigation.
• GPS + IMU + Other sensors (e.g., magnetometer, odometer): Adding a magnetometer can aid heading determination, particularly in GPS-denied environments. An odometer provides additional information about speed and distance traveled, useful in land-based applications. These additions improve robustness and resilience.

The optimal sensor combination depends on factors such as the application’s requirements for accuracy, cost, size, weight, and power consumption, as well as the operating environment.

GNSS (Global Navigation Satellite System) integration dramatically improves INS performance by mitigating the inherent drift of inertial sensors. GNSS provides absolute position and velocity information, acting as a reliable reference to correct for errors accumulating in the IMU data. This correction is crucial because the errors in IMU readings grow over time, making long-term navigation impossible with IMUs alone.

The integration process often involves a Kalman filter or similar estimator. The filter combines the GNSS measurements (which are less precise at short time intervals) with the high-frequency IMU measurements to produce a smoothed and more accurate estimate of the position, velocity, and attitude. When GNSS signals are temporarily unavailable (e.g., during signal blockage), the INS continues to provide navigation based on the IMU data, but its accuracy will degrade over time. Upon regaining GNSS signal, the system quickly re-aligns and improves accuracy.

Imagine a ship navigating across a large ocean. The INS is like the ship’s compass and odometer, providing continuous measurements. However, the compass may drift slightly over time, and the odometer may miscount revolutions. GNSS acts like a regularly updated map showing the ship’s precise location, allowing the crew to correct for any drifts in the compass and odometer readings.

Inertial navigation is a navigation technique that uses inertial sensors (accelerometers and gyroscopes) to track the movement of an object without relying on external references. It works by measuring the object’s acceleration and angular velocity to calculate its position, velocity, and attitude.

Accelerometers measure linear acceleration along three orthogonal axes. By integrating acceleration over time, the system estimates velocity. Integrating velocity over time yields the position. Gyroscopes measure angular velocity around three axes and allow the system to determine its orientation or attitude. This combined information provides a complete understanding of the object’s movement.

However, because the integration process involves accumulating errors from the sensors, an INS’s position and velocity estimates drift over time. This drift needs to be mitigated through sensor fusion, calibration, and other techniques.

INS utilize several coordinate systems to represent the object’s position, velocity, and attitude:

• Body Frame (b-frame): A coordinate system fixed to the object itself. Its axes move with the object.
• Earth-Centered, Earth-Fixed (ECEF) Frame (e-frame): A coordinate system with its origin at the Earth’s center, rotating with the Earth. Used for representing absolute positions.
• Local-Level, North-East-Down (NED) Frame (n-frame): A local coordinate system with the x-axis pointing north, the y-axis pointing east, and the z-axis pointing down. Convenient for representing positions relative to a local point.
• Navigation Frame: A frame based on the local tangent plane and aligned with the earth’s rotational axis.

Transformations between these coordinate systems are essential for calculations within the INS, involving rotation matrices to accurately translate measurements between frames.

Sensor failures in an INS are a serious concern that can lead to navigation errors. Handling them requires a multi-pronged approach:

• Redundancy: Using multiple sensors of the same type (e.g., multiple IMUs) allows for fault detection and isolation. If one sensor fails, the others can continue operation, although accuracy may be slightly reduced.
• Sensor Health Monitoring: Implementing algorithms to constantly monitor the health of each sensor based on consistency checks and data plausibility. Detection of an anomaly can trigger an alarm.
• Fault Detection and Isolation (FDI): Employing sophisticated algorithms that identify the faulty sensor and remove its data from the fusion process.
• Sensor Data Validation: Checks the plausibility of data by comparing it against expectations or other sensor readings. Outliers or improbable values can be flagged or rejected.
• Fail-safe mechanisms: Having backup modes of operation which might involve switching to a less accurate navigation system or reducing the system’s functionality until the failure is addressed.

The specific strategies used depend on the criticality of the application and the available redundancy.

Despite their capabilities, INS have limitations:

• Drift: The accumulation of errors over time, due to sensor imperfections, is inherent and unavoidable. This drift limits the accuracy of long-term navigation.
• Sensitivity to initial conditions: The accuracy of the INS is sensitive to the initial alignment of the sensors and the initial position and velocity estimates. Any inaccuracy in these initial conditions will propagate through the calculations.
• Cost and Complexity: High-quality IMUs and associated processing units can be expensive, making INS systems costly.
• Environmental effects: Temperature, pressure, and other environmental factors can affect sensor performance, leading to inaccuracies.
• Vulnerability to harsh conditions: High shock and vibration can damage the sensors and affect their performance.

Understanding these limitations is crucial for designing and implementing a robust and effective navigation solution.

An Inertial Navigation System (INS) determines its position, heading, and attitude using internally measured accelerations and rotational rates. Think of it like a sophisticated version of a self-contained, high-precision compass and speedometer combined.

Attitude refers to the orientation of the INS in three-dimensional space. It’s essentially how the INS is tilted and rotated relative to a known reference, often represented by Euler angles (roll, pitch, yaw) or quaternions. Imagine a gimbaled platform; its orientation defines the attitude.

Heading is simply the direction of movement, typically expressed as an angle relative to true north. A ship’s heading, for example, dictates its course.

Position refers to the INS’s location in geographical coordinates (latitude, longitude, and altitude). This is the most crucial output for navigation and typically achieved by double integrating the measured accelerations.

An INS estimates velocity by integrating the measured accelerations. This sounds simple, but it’s inherently complex. Each accelerometer measures acceleration along a specific axis. These accelerations are integrated once to obtain velocity and again to obtain position. However, this process accumulates errors over time due to various sources like sensor noise and imperfect alignment. Consider a simple example:

Let’s say an accelerometer measures a constant acceleration ‘a’ along the x-axis. The velocity ‘v’ at time ‘t’ can be estimated by:

where v(0) is the initial velocity. This integration process is continuously performed by the INS computer, using sophisticated algorithms to correct for errors.

The key difference between strapdown and gimbaled INS lies in how the inertial measurement unit (IMU) is mounted.

In a gimbaled INS, the IMU is mounted on a set of gimbals that mechanically isolate it from the vehicle’s movement. This keeps the IMU’s sensitive axes precisely oriented, even if the vehicle rotates. Think of it like a camera on a tripod—the camera (IMU) stays steady even if the tripod (vehicle) moves. However, gimbaled systems are complex, mechanically challenging, prone to failure, and have limited dynamic range.

A strapdown INS, in contrast, directly mounts the IMU to the vehicle’s body. This simplifies the system significantly, eliminating the complex gimbal mechanisms. The vehicle’s motion is mathematically compensated for using sophisticated algorithms that precisely calculate the orientation and motion of the IMU and vehicle. Strapdown INS has become the dominant design due to its improved reliability, reduced cost, and better dynamic performance. It’s analogous to having a camera mounted directly to a moving vehicle and computationally correcting for the motion during image processing.

An Inertial Measurement Unit (IMU) is the heart of an INS. It’s a package of sensors that measure the specific force (acceleration less gravity) and angular rates (rotation) of the vehicle. Think of it as the sensory system of the INS. A typical IMU includes three accelerometers measuring specific force along three orthogonal axes (x, y, and z) and three gyroscopes measuring angular rates about those same axes. The accuracy and precision of the IMU are critical for the performance of the entire INS.

There are several types of IMUs, categorized primarily by the type of sensors used:

• MEMS (Microelectromechanical Systems) IMUs: These use micromachined sensors fabricated on silicon wafers. They are small, lightweight, and relatively inexpensive, but offer lower accuracy than other types. They are commonly found in consumer devices such as smartphones.
• Fiber Optic Gyroscopes (FOG) IMUs: FOGs utilize the Sagnac effect to measure angular rates with high accuracy and stability. They are typically more expensive than MEMS but offer significantly improved performance. They find use in high-precision navigation systems.
• Ring Laser Gyroscopes (RLG) IMUs: RLGs are similar to FOGs, but use laser beams instead of fiber optics. They offer extremely high accuracy but are larger, more expensive, and more susceptible to lock-in issues (a phenomenon that can cause measurement errors).
• Dynamically Tuned Gyroscopes (DTG) IMUs: DTGs are highly accurate and robust but also fairly expensive. They combine aspects of both RLG and FOG technology.

The choice of IMU type depends on the specific application requirements, balancing cost, size, weight, power consumption, and desired accuracy.

INS alignment is the crucial initial process of establishing the INS’s initial attitude, heading, and position. It’s like setting up a compass and map before starting a journey. There are several methods, but they all aim to align the INS’s coordinate frame with the Earth’s coordinate frame. Common methods include:

• Static Alignment: This involves keeping the INS stationary for a certain period, allowing it to estimate its orientation based on the Earth’s gravity vector and rotation rate. This approach is simpler but takes more time.
• Dynamic Alignment: This uses motion information from the vehicle to perform alignment faster than static alignment. It’s particularly useful for platforms that don’t allow prolonged stationary periods.
• Aided Alignment: This utilizes external sensors, such as GPS, to aid in the alignment process. This significantly reduces the alignment time and improves accuracy.

The chosen method depends on the application’s requirements and constraints. For example, a submarine might use a longer static alignment to achieve high accuracy, while a rapidly maneuvering aircraft might require a faster dynamic or aided alignment.

Kalman filters are essential for INS to reduce errors and improve accuracy. Several types are used, but the most common are:

• Extended Kalman Filter (EKF): This is a nonlinear filter that linearizes the system around its current estimate, and hence is frequently applied to navigation system error modeling. It’s widely used due to its relative simplicity and effectiveness.
• Unscented Kalman Filter (UKF): The UKF is another nonlinear filter that addresses some limitations of the EKF by using a deterministic sampling technique to approximate the probability distribution. The UKF usually provides better accuracy than the EKF, especially in highly nonlinear systems.
• Adaptive Kalman Filter: This type of filter adjusts its parameters (like the process noise and measurement noise covariances) during operation based on observed data. This improves performance by adapting to changing system dynamics and environmental conditions.

The selection of a specific Kalman filter depends on the complexity of the system model, the nature of the noise, and the computational resources available. The performance of a navigation system critically depends on the choice of Kalman filter implementation.

Data pre-processing in an Inertial Navigation System (INS) is crucial for ensuring the accuracy and reliability of the navigation solution. Raw sensor data is often noisy, contains biases, and may be inconsistent. Pre-processing aims to clean and prepare this data for use in the INS algorithms. Think of it like preparing ingredients before cooking – you wouldn’t start baking a cake with unwashed flour, would you?

• Noise Filtering: Techniques like Kalman filtering or moving averages smooth out random fluctuations in the accelerometer and gyroscope readings, improving the signal-to-noise ratio. This is like removing small pebbles from your flour before baking.
• Bias Estimation and Compensation: Gyroscopes and accelerometers are susceptible to biases – systematic errors that lead to drift in the navigation solution. Pre-processing algorithms estimate and remove these biases, often using calibration procedures or advanced algorithms. This is analogous to ensuring your baking scale is properly zeroed before weighing ingredients.
• Data Alignment: Sensor data from different sources needs to be aligned in a common coordinate frame. This ensures that data from, for example, different gyroscopes on a platform, are correctly combined. This is like making sure all your measuring cups are aligned before pouring ingredients into a bowl.
• Outlier Detection and Removal: Sudden spikes or improbable values in the data (outliers) can significantly affect the INS’s performance. Pre-processing algorithms detect and remove these outliers to maintain solution integrity. This is similar to removing a spoiled ingredient to prevent ruining the entire recipe.

Effective data pre-processing is essential for achieving a high-quality navigation solution and is often customized to the specific sensors and application requirements.

Evaluating the accuracy of an INS involves comparing its estimated position, velocity, and attitude with a known reference. This reference might be a high-precision GPS system, a surveyed trajectory, or other independently verified data. Several methods are employed:

• Root Mean Square Error (RMSE): This statistical measure quantifies the average difference between the INS-estimated values and the reference values over a period. A lower RMSE indicates better accuracy. For instance, a lower RMSE in position indicates that the INS estimates the location more closely to the actual location.
• Bias Analysis: This evaluates the systematic errors in the INS solution, particularly the drift over time. A well-calibrated INS will have a small bias, indicating minimal long-term error accumulation.
• Comparison with Ground Truth: When possible, comparing the INS results with ground truth data (e.g., surveyed coordinates) provides a direct measure of accuracy. This is particularly useful for evaluating position accuracy.
• Residual Analysis: Examining the difference (residuals) between the INS measurements and the integrated navigation solution can highlight inconsistencies or errors in the system. Large residuals could signal sensor malfunction or problems with the INS algorithms.

The chosen evaluation method depends on the application and the available reference data. For example, in a land vehicle application, using a high-precision GPS receiver for comparison is common; whereas in aerospace, celestial navigation data may serve as a reference. The goal is to quantify and understand the error sources in order to improve the INS performance.

Common performance metrics for INS include:

• Position Accuracy (CEP, RMSE): Circular Error Probable (CEP) and RMSE represent the accuracy of position estimation. CEP denotes the radius of a circle containing 50% of the position errors, while RMSE is the square root of the mean of the squared position errors. Both metrics are often expressed in meters.
• Velocity Accuracy (RMSE): Similar to position accuracy, this metric measures the accuracy of velocity estimation, typically in meters per second.
• Attitude Accuracy (RMSE): This metric measures the accuracy of attitude (orientation) estimation in degrees or radians for roll, pitch, and yaw angles.
• Drift Rate: This indicates the rate at which position or velocity errors accumulate over time, often expressed in meters per hour or degrees per hour. A lower drift rate reflects better long-term accuracy.
• Alignment Time: This metric is relevant for systems requiring initial alignment and represents the time taken to accurately determine the initial orientation of the INS.

The importance of each metric depends heavily on the application. For instance, high-precision surveying might demand very low position error, whereas a simpler navigation system for a small drone might tolerate a slightly higher error. In all cases, the metrics provide quantitative measures for assessing and comparing INS performance.

Various algorithms are employed for INS, each with strengths and weaknesses:

• Mechanization Equations: These fundamental equations describe the propagation of position, velocity, and attitude based on the sensor measurements (accelerometer and gyroscope data). Different mechanization schemes exist (e.g., local-level, body-frame) depending on the coordinate system used. These equations are the cornerstone of any INS algorithm. They are often implemented using numerical integration techniques like the Runge-Kutta method.
• Kalman Filtering: This powerful algorithm optimally combines INS data with data from other sensors (e.g., GPS, barometers) to estimate the state of the system (position, velocity, attitude, and sensor biases). It handles noisy sensor measurements and uncertainties effectively, leading to improved accuracy. The Kalman filter continually updates its estimate based on new data, resulting in a continuously refined solution.
• Extended Kalman Filter (EKF): An adaptation of the Kalman filter for non-linear systems, commonly used when dealing with more complex dynamic models or sensor measurements. This is needed because INS models often exhibit nonlinear behavior.
• Unscented Kalman Filter (UKF): Another algorithm for non-linear systems offering advantages in certain situations compared to the EKF. The UKF provides a more accurate representation of the probability distribution of the state, leading to better performance in highly nonlinear scenarios.

The choice of algorithm depends on the application’s complexity, the desired accuracy, the computational resources available, and the types of sensors integrated with the INS.

Handling data from multiple sensors with varying update rates requires careful consideration and often involves data fusion techniques. The different update rates necessitate a strategy to combine the information effectively without losing valuable data or introducing inconsistencies.

• Data Interpolation/Extrapolation: Data from the sensor with a lower update rate can be interpolated (estimating values between measurements) or extrapolated (estimating future values) to match the update rate of the higher-frequency sensor. For example, if one sensor updates at 100 Hz and another at 10 Hz, interpolation or extrapolation would be used to match the data rates before fusion.
• Sensor Fusion Algorithms: Methods like Kalman filtering, as described above, excel at fusing data from sensors with different update rates. These algorithms handle the uncertainties associated with interpolation/extrapolation and optimally combine the information to improve overall accuracy and reliability. For example, the Kalman filter weights the sensor measurements according to their respective uncertainties.
• Time Synchronization: Precise time synchronization across all sensors is critical. Any timing mismatches will lead to errors in the fused data. Using a high-precision clock is essential for data synchronization.

The specific approach depends on the sensor characteristics, the accuracy requirements, and the computational capabilities of the system. Carefully chosen methods maintain the integrity of the data and provide a robust navigation solution.

Dead reckoning in an INS is the process of estimating the current position, velocity, and attitude based solely on previous position and sensor measurements. It’s like navigating a ship using only its compass and log (speed measurement) – without any external references. It’s essentially the integration of accelerometer and gyroscope data to estimate the change in position and attitude over time.

The process starts with an initial known position, velocity, and attitude. The accelerometers measure acceleration, which is integrated once to get velocity and twice to get position. The gyroscopes measure angular rate, which is integrated to get the orientation. However, errors accumulate rapidly in dead reckoning due to sensor noise, biases, and inaccuracies in the integration process. This accumulation is often described as ‘drift’. This drift means that a dead reckoning-only system will quickly become inaccurate over time.

Because of the accumulating errors, dead reckoning is rarely used on its own in navigation applications. It’s typically used in conjunction with other sensors and algorithms (like Kalman filtering) to improve its accuracy and correct for drift. Think of it as a short-term position estimate that constantly needs external corrections to stay accurate.

INS finds applications in a vast array of fields:

• Aerospace: In aircraft, missiles, and spacecraft for navigation, guidance, and control. For instance, inertial navigation is critical for aircraft navigation during periods of GPS outage.
• Automotive: In advanced driver-assistance systems (ADAS) and autonomous vehicles, providing accurate position, velocity, and heading information. For example, anti-lock braking systems can use the vehicle’s inertial measurements.
• Robotics: In mobile robots, allowing them to navigate autonomously and perform precise movements, even in GPS-denied environments (like indoors). The movement of an industrial robot arm can be precisely controlled using INS.
• Marine: In ships and submarines, for navigation and stability control, especially in challenging conditions where GPS signals may be unreliable. Submarines, for example, use inertial navigation for underwater navigation.
• Surveying and Mapping: For accurate geospatial positioning in areas with limited or no GPS coverage. Detailed topographic surveys can benefit from the high accuracy of inertial measurements.

The specific application often dictates the level of accuracy, size, power consumption, and cost requirements for the INS. Innovations in MEMS (Microelectromechanical Systems) technology have made smaller, lighter, and lower-cost INS units possible, expanding their reach into diverse applications.