MEMS accelerometers work by measuring the capacitance between a moving mass and the frame of reference (the die). The moving mass is made by etching away everything alongside and underneath it to physically isolate it from the die, except for the carefully constructed suspension beams. Maximizing the capacitance of the assembly, and therefore the sensitivity and range of the sensor, usually involves creating complicated interdigitated fingers of silicon. The mass is free to move in one axis and is held in position from the bulk silicon by supporting beams that constrain its movement in the perpendicular direction.
One and two dimensional, in plane accelerometers can be made in this way, but to integrate a 3D accelerometer a different structure has to be used. By calibrating for the different geometries and sensitivities of the springs, it is possible to produce figures for acceleration in all three axes. Typical low g accelerometers measure full scale accelerations up to about 16 g, depending on how they are calibrated and trimmed.
The common form of MEMS gyroscope is a vibrating, typically wheel shaped structure; when subject to rotation, it allows a transducer to detect a Coriolis term in its equations of motion.
There is also an emerging automotive market for microbolometers and CMOS image sensors to enable headsup displays with nightvision capabilities, according to Yole's Robin.
The principle of operation for a microbolometer is analogous to that of an image sensor, with an array of pixels that are sensitive to infrared radiation. But the microbolometer requires that the heatabsorbing array be isolated from a supporting silicon substrate. It is necessary to etch away a sacrificial layer, which can be up to 2 microns deep.
Thus the principle of manufacture for a microbolometer is similar to that for inertial MEMS sensors.
Mainstream adoption of what are effectively night vision cameras in cars is likely to be driven, again, by regulation. One idea is that information about pedestrians or animals moving in front of a vehicle can be displayed on the windshield as a form of augmented reality, to help the driver avoid obstacles—and, in extreme circumstances, to trigger automated braking.
Pending EU regulations making such pedestrian detection mandatory are expected to kick start a market around 2014 or 2015, Robin said.
Two European collaborative research projects (ADOSE and ICU+FNIR) pursuing this technology have recently wrapped up their work, and some topend vehicles from Toyota, BMW and Ford already offer nightvision systems, Robin said.
Scanning micromirror optical MEMS devices, used as projection display drivers, may have a role to play in projection headsup displays, he added.
@PeterClarke: good summary but I will get you on a technicality!! Your statement "MEMS accelerometers work by measuring the capacitance..." is only partially correct. The capacitive proof mass type of accelerometers are the most common but there are many other types using different mechanical and electromechanical behaviours. For example, you can have a piezoelectric type (again a proof mass type but senses piezo voltage differentials), gas flow type... In the late nineties, I worked on the latter type (single and dual axis) which uses the principle of a hotwire anemometer to sense change in flow pattern of encapsulated gas brought on by the g-field. The problem is the cavity size needs to be small enough to increase the frequency response of the system to an acceptable level. MEMSIC even now produces g-sensors that work on this principle.
Your last statement some what concerns me -IDM route is not the way to grow MEMS business when faster innovations can come from fabless startups. This is absolutely essential to see a healthy advancement of MEMS market.
Dr. MP Divakar
There are so many requirement required to be fitted in the cars depending the different terrains, locations, types of usage many more parameters. Basically cars/vehicles are the second most occupied places where a person will have to spend his/her time throwout the life. So it will go on and on in terms of developments in incar electronics.
It need not be an astronaut's suit but just a few wireless sensors fitted in the car itself around the driver's seat could do this job. A wireless breath analyzer, a facial recognition camera and such simple minded stuff would do
Intriguing but not very practical. The car environment is being enhanced for the masses; you are suggesting that the masses need to be roused to be alert. That time will sure come but it will be part of the sensors installed in a vehicle and on the road. Doubt anybody wants to "suit up" like an astronaut to take the car for a spin or a long trip.
More and more electronics is being added to assist the driver in parking, checking surrounding obstacles and so on. Why not simplify the matter by having some electronics mounted on the driver's body to check the driver's alertness and if found dozing, not looking at the front, drunken, yawning, not overspending etc then creating some stimulus to bring him back to his senses. Because if the driver is alert and attentive then all the other things are automatically taken care of
biaunm: I believe Peter mentions in his story on first page, TPMS stands for tire pressure monitoring system. I urge you to read on and take a gander at the rest of the special edition. Table of contents here: http://www.nxtbook.com/nxtbooks/cmp/eetimes_ai_20110609/#/3/OnePage
Peter Clarke, since a central point in your story is that regulations are fueling growth, it would have been a kindness to your readers to have explained briefly what the TPMS mandate is or at least to have expanded the TPMS acronym.
If you got into this later in the story, I apologize for missing it. I stopped reading at this point, on page 1 of 3.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.