by Marcel van Hak
Industrial Applications Product Manager, Xsens
In the past decade, MEMS inertial sensor technology has taken huge steps in terms of accuracy, reliability and cost. The most innovative manufacturers of inertial measurement units (IMUs), attitude and heading reference systems (AHRSs) and inertial navigation systems (INSs) have taken this opportunity to bring products to market that are good alternatives to costly and bulky fiber optic gyros (FOGs). With MEMS IMUs comprising almost 40% of the commercial market in value in 2010, compared to 13% for FOGs (Yole Developpement), it is clear that MEMS IMUs are integrated in an increasingly larger number of applications. This article explores some of the applications that greatly benefit from this new class of motion trackers in the $1,000 to $5,000 cost range. For an introduction on MEMS based AHRSs, please read our earlier interview with Per Slycke, CTO of Xsens.
One of the main applications for motion tracking is control and stabilization of unmanned vehicles, platforms and gimbals. In these applications, a low latency of under a few milliseconds is essential. The latest MEMS AHRS products meet these needs, with latencies that are less than 2 ms. Due to the dynamics involved in these applications, vibrations and long-lasting accelerations appear frequently, leading to temporary errors and to errors that accumulate over time (drift). Signal processing is extremely important in order to cope with the negative effects of vibrations. A new class of vibration rejecting gyroscopes, featuring high bandwidths, combined with powerful signal processing algorithms allows vibrations with frequencies of up to 200 Hz to be rejected. Negative effects of long-lasting accelerations are addressed with in sensor fusion, discussed further on in this article.
Figure 1. An unmanned helicopter used for aerial filming.
One application of control and stabilization is low-cost camera gimbals. Miniature helicopters, with rotors measuring only a few meters in diameter, and with a maximum payload of 5-10 kg, are now commonly used for the filming of commercials, movies and corporate films. In this application, all challenges come together: a heavily vibrating helicopter -- small sized, yet so fast that long-lasting accelerations are present -- and a very small space available for payload (including the IMU or AHRS). The response of the gimbaled camera has to have low latency to get smooth images and to follow fast moving objects. Combining all these requirements previously necessitated the use of a highly accurate (and thus bulky and heavy) IMU which would not even make such a product technically feasible. With industrial grade MEMS and intelligent sensor fusion algorithms, the performance requirements are met at just a fraction of the price of a FOG.
Sensor fusion algorithms
Traditional limitations of MEMS based AHRSs, such as magnetic distortions by steel, permanent magnets, or electric currents, have been resolved in the past few years. The solution is based on high-performance sensor fusion algorithms that compensate fully for magnetic distortions by detecting the distortion and adapting the heading estimation mechanism accordingly.
Figure 2. Range of MEMS IMU, AHRS and GPS/INS products from Xsens.
Another example of the importance of sensor fusion algorithms is when the application experiences long-lasting accelerations. For accelerometers in AHRSs, there is no difference between gravity and lateral accelerations. If these lateral accelerations are not properly compensated the roll and pitch values will be incorrect. New sensor fusion algorithms can detect these accelerations and again adapt the sensor fusion according to the measured dynamics. While magnetic distortions can appear in many industrial applications, long-lasting accelerations are mostly found in airborne applications and ground based vehicles.
The importance of sensor fusion algorithms is underscored by the fact that manufacturers of mobile devices, processors and motion sensors now work with specialized software vendors for their motion tracking needs in consumer electronics applications.
Whereas IMUs and AHRSs are often used to actively control and stabilize a gimbal or an unmanned vehicle, they can also be used to correct measurements of imaging devices on moving platforms, e.g. a laser scanner, camera or acoustic imaging device (sonar). One emerging application in this field is low-cost 3D mapping. The most well-known 3D mapping application is Google Street View. This application uses a high-grade GNSS receiver and a tactical grade IMU in order to find the position and direction of their 360 degree view camera setup. Systems like these typically cost hundreds of thousands of dollars, which prohibits the use of such inertial navigation systems (INSs) for many smaller-scale mobile mapping challenges. Many low-level government bodies, such as municipalities or fire brigades in forest fire areas, would like to have access to cost-effective systems, for example to monitor public works, or to find hot spots in the forest with an unmanned aerial vehicle.
Thanks to MEMS technology, complete entry-level mobile mapping solutions are now commercially available. For example, a low-cost laser scanner/thermal camera, a MEMS based INS that outputs position and orientation, and a laptop with control software can all be combined in a cargo box that fit on any ordinary car. These low-cost solutions are targeted to users with limited budgets and who are not willing or able to refurbish a whole car for the purpose of mobile mapping. For aerial mapping, the laser scanning solution is often deployed on a small UAV, weighing as little as 5-10 kg. Due to the small size and low weight of MEMS AHRSs, it is no longer needed to rent a special aircraft, and pilot, for aerial mapping.
Figure 3. Camera stabilization (in this case on a ship) is often performed with MEMS sensors.
Offshore and cargo
A market that has seen one of the largest growth areas in the use of MEMS AHRSs is the offshore and maritime market. The ability to cope with vibrations and magnetic distortions has enabled new applications. Shipping companies and other companies active in heavy industry are adopting low-cost MEMS AHRSs now.
Ship state estimation has many applications and one of them is cargo management. Using MEMS AHRSs, the system determines motions, accelerations and velocities of the ship and can calculate these values at any location on the ship as a virtual sensor. Combining this information with weather data, wave measurements, cargo information and many more measurements, the best course and velocity is determined, for example to minimize roll and to optimize fuel consumption. Without the accuracy and robustness of current day MEMS based motion trackers, an application such as this would not have been possible to achieve in such a cost-effective way.
Figure 4. A kite-propelled cargo ship, using MEMS AHRSs for control.
The upcoming market of green energy is rapidly adopting MEMS AHRSs. One emerging application is the use of large kites for propulsion of container ships or electricity generation in wind farms. In the application where the kites are attached to a cargo vessel, the system makes use of high-altitude winds to help propel the vessel. MEMS AHRSs are used to steer the kite for maximum efficiency. In the case of kites attached to generators, kites are installed in large offshore wind farms. Retracting the kite after a full cycle has to be done at exactly the right moment and MEMS AHRSs are used to determine this moment. Any weight or size added to these kites makes them less efficient and even with the high accelerations in these applications, MEMS AHRSs give sufficient accuracy.
MEMS based AHRSs: the next generation in motion tracking?
MEMS AHRSs are often associated only with unmanned or remotely operated vehicles. There is a rapid growth of new application areas where MEMS IMUs and AHRSs can also be used. These are enabled by advances in the technology: more accurate MEMS inertial sensors, better sensor fusion algorithms and miniaturization of components. System integrators who embrace this new class of IMUs and AHRSs will be able to make innovative products for their customers.
This article is a part of MEMS Journal's ongoing market research project in the area of MEMS based motion sensors. If you would like to receive our comprehensive market research report on this topic, please contact Dr. Mike Pinelis at [email protected] for more information about rates and report contents.
Copyright 2013 MEMS Journal, Inc.