Driven, among other factors, by affordable GPS, powerful processors and low-cost sensors, Unmanned Aerial Vehicles (UAVs) have proliferated in recent years. Amidst this boom, the industry has focused on concerns like flight-time, payload/size, and precision. Flight-time and payload are significantly impacted by the power requirements, not just to drive the motors to keep a UAV aloft, but also for the processing required to calculate and correct errors in attitude, heading and tracking the flight plan. In most cases, navigation is primarily carried out by calculations based on GPS coordinates. However, in order to avoid losing guidance during a GPS drop-out, a UAV must rely on its on-board inertial sensors for short-term navigation while dead reckoning. But, there’s the rub – existing commercial MEMS gyroscope solutions cannot always be relied upon to provide the precision required for proper attitude, heading or navigation.
Figure 1. Orientation Angles of a Common Quadcopter
Imagine that you are trying to autonomously guide a quadcopter drone through a maneuver that involves applying a pitch rotation of +30deg and -30deg, This can be achieved with a rotation rate of approximately 5 deg/s clockwise and counter-clockwise – for about six seconds in each direction – as shown in Figure 2. Now let us imagine this maneuver was carried out under two different environmental conditions: (1) in the absence of any vibration, and (2) in the presence of vibration stimulus not so dissimilar to quadcopter running motors.
Figure 2. Input Stimulus and Desired 30-deg Pitch Maneuver
The problem with consumer grade MEMS gyros is that they are easily overloaded by shock and vibration. Based on actual results from a consumer-grade device, Figure 3 illustrates the response of one of these 3-axis gyroscopes to external shock and vibration. In this figure the first row is the typical measured output data under vibration from a DC motor similar to that found in a quadcopter drone. The second row shows the pitch, roll and yaw angles calculated from the measured rotation data.
Even with oversampling and filtering, the signal cannot be properly reconstructed using electronics or software. This is reflected in the calculated results from the gyroscope. Even in absence of an actual maneuver by the gyrocopter in the roll or yaw axis (the actual rotation rate is 0), vibration induced errors register an angular shift of approximately 5-degrees and ±3-degrees in each of those axes, respectively. When you factor in an average speed of 10-20 mph, this represents a deviation several feet from a programmed destination during only a simple 20 second maneuver! Imagine how this could play out during a full mission.
Figure 3. Simulated Orientation Error of Tuning-Fork/Shell-Based MEMS Gyroscope
Given these limitations of traditional gyroscope architectures, the engineers at Qualtré have developed a new class of MEMS vibratory gyroscope based on degenerate bulk-acoustic modes of circular disks. A Bulk Acoustic Wave (BAW) gyroscope relies on the transfer of energy between two degenerate BAW modes typically operating in MHz regime. This is an inherently rigid construction and is far less susceptible to vibrations as demonstrated by its response to the same input stimulus. In Figure 4, the first row shows the actual response of Qualtré’s 3-axis industrial-grade gyroscope to the same DC motor vibration. The X, Y and Z-axis signals are clean – and this results in the highly accurate angle calculations in the second row. Nothing beats a clean signal!
Figure 4. Simulated Orientation Error of Bulk-Acoustic Wave MEMS Gyroscope
To understand why existing tuning-fork and shell-type architectures perform poorly against BAW technology, it is important to understand the fundamental difference between the two technologies. Although all vibratory MEMS gyroscopes rely on the rotation-induced Coriolis force, the operating modes of the two technologies are different. In a tuning fork device, there is a transfer of energy between two flexural modes of a proof-mass held in place via suspension springs. This design is low-frequency, typically between 10-50 kHz, and is inherently compliant. In fact the susceptibility of these devices to shock and vibration has been exposed in widely publicized examples such as the “gyrophone” and recent publications of academics in KAIST who have downed drones by application of sound-waves!
Utilizing bulk-acoustic wave resonant modes, a BAW gyroscope is able to operate at frequencies that are orders-of-magnitude higher than most environmental and/or acoustic frequencies. This high stiffness of the BAW disk resonator gives Qualtre gyroscopes considerable inherent immunity to shock and vibration and access to a clean “rate” signal without need for complex processing – and without risking the UAV to maneuver errors or even crashes. This frees up processor bandwidth for other important functions like image stabilization, better optimized algorithms, or longer flight time. Additionally, a higher resonator frequency also reduces the noise floor, significantly enhancing the signal-to-noise ratio without increasing size and power requirements of the gyro.
To learn more about BAW MEMS, go to: