Unmanned Systems Technology 006 | ECA Inspector Mk2 USV | Antenna systems | Northwest UAV NW-44 | Unmanned ground vehicles | Navigation systems | Lunar X challenge
70 new class of sensors described below.) Driven electrostatically by electrodes, they vibrate at their natural resonant frequency in their designed mode until the vehicle or component in which they are installed rotates about the axis being measured, when the Coriolis effect causes simultaneous movement along another axis. This is picked up electrostatically by sense electrodes, generating output voltages proportional to the angular velocity. Sensors using flex and twist modes are the most common but are susceptible to vibration, shock and stiction (a combination of sticking and friction), all of which can cause errors or failures. Furthermore, in high-performance applications they need packaging that maintains a vacuum to minimise the damping effects of air, which causes the resonance to decay. Much of the cost of high-performance MEMS gyros is incurred in giving them resistance to shock and vibration, and in vacuum packaging. Bulk acoustic wave robustness Gyros that vibrate in bulk mode, known as bulk acoustic wave (BAW) devices, are a new technology created to address the shortcomings of the other types of MEMS gyro, and promise greater accuracy and bias stability along with greater reliability, higher performance and lower cost for high-end consumer applications and low-end tactical/industrial applications including small UAVs. MEMS gyros used in some small UAVs have proved vulnerable to vibration from the vehicle’s electric motors and, according to a South Korean study published in August 2015, the craft can be brought down by attack with sound waves. The problem is that frequencies at which conventional tuning fork MEMS gyros resonate are in the 10-50 kHz range, low enough for mechanical vibration and acoustic tones to set them off and overwhelm the angle rate signals the vehicle’s control system needs. BAW sensors, in contrast, use silicon discs that resonate in the 1-10 MHz range, far higher than any likely acoustic interference. This means BAW MEMS gyros can be used without much in the way of correction software to clean up the signal, it is claimed, freeing up processor capacity for other functions such as image stabilisation. An additional claim is that the higher resonant frequency lowers the noise floor, which improves the February/March 2016 | Unmanned Systems Technology Focus | Navigation systems MEMS gyros are good enough now for some consumer automotive and low-end tactical/ industrial-grade applications GPS- and GLONASS-capable, and ready for BeiDou and QZSS, this multi-GNSS module also features an embedded wideband omnidirectional antenna (Courtesy of u-blox) signal-to-noise ratio without increasing the gyro’s size or power requirements. The accuracy of a gyro is usually expressed as an error proportional to the true value of the angle or angle rate being measured, in terms of a percentage, parts per million (ppm) or even parts per billion (ppb). Bias is the long-term average output of the gyro when it is not being rotated, while bias stability is a measure of how quickly this changes. With that in mind, MEMS gyros are now good enough for some consumer, automotive and low-end tactical/ industrial-grade applications, and are beginning to be competitive in higher- end tactical applications. Consumer applications such as motion interfaces for sports and fitness, for example, need an accuracy of about 3% and a bias stability of 10°/s, while automotive applications such as electronic stability programs it is 0.3% and 1°/s respectively. Meanwhile, the lower-end tactical/industrial applications such as ammunition and rocket guidance need an accuracy to 10 ppm and a bias stability of 10°/h. As tactical applications such as platform stabilisation in UAVs and other aircraft become more demanding though, the required standard jumps to 1 ppm and 1°/h. There is a further jump for short-term navigation: in missiles
Made with FlippingBook
RkJQdWJsaXNoZXIy MjI2Mzk4