MEMS Inertial Sensors Move Beyond Airbags

Oct. 1, 2004
The use of MEMS accelerometers to sense rapid deceleration for airbag deployment is well over a decade old and considered a well-known technology. However,

For the PDF version of this article, click here.

The use of MEMS accelerometers to sense rapid deceleration for airbag deployment is well over a decade old and considered a well-known technology. However, MEMS inertial sensors have found many other uses in automobile environments of late. This article discusses some of the many applications for MEMS inertial sensors and the performance requirements of those sensors that improve the safety, convenience and reliability of cars and trucks today.


Many types of systems are geared toward enhancement of vehicle stability and control. The most common is the almost ubiquitous automatic braking system (ABS). Until recently, most ABS systems did not use an inertial sensor. They simply read wheel speed and apply pulsed braking if the wheels are thought to be skidding. However, most all-wheel-drive systems and some newer high-performance ABS systems look at longitudinal acceleration to determine if the chassis is still moving. This is particularly important for all-wheel-drive-equipped vehicles where all four wheels may have lost traction due to the application of drive torque.

The most important performance parameters for accelerometers used for ABS is zero g bias and sensitivity stability. In general, it is assumed that the minimum available deceleration force available (even on slippery surfaces) will be about 100 mg (0.98 m/s2). So the combination of zero g bias drift and sensitivity variation must not vary more than 100 mg over the automotive temperature range (see Table 1 for more information regarding performance requirements). Analog Devices' ADXL103 iMEMS® accelerometer with a typical zero g bias stability of 16 mg and sensitivity drift of 0.3% over the automotive temperature range is ideal for this application.

Electronic stability control (variously known as ESC, VSC, VDC — each automaker seems to have its own acronym) assists the driver to regain control of the automobile just as it is starting to skid. A VDC system uses a yaw rate sensor (or gyroscope), a low g accelerometer, wheel speed sensors (which may also be used by the ABS system) and steering wheel angle input. Wheel speed from each wheel is measured, and the predicted yaw (or turn) rate of the car is compared to that measured by the gyroscope and the intentions of the driver (as predicted by the steering wheel angle). A low g accelerometer is also used to determine if the car is sliding laterally. If the measured yaw rate differs from the computed yaw rate or lateral sliding is detected, single wheel braking or torque reduction can be used to make the car “get back in line.”

ESC systems require a yaw rate sensor with fairly low noise (typically less than 0.5°/sec) and low sensitivity to mechanical vibration. Many types of yaw rate sensors actually generate vibration (most MEMS gyros use a vibrating mass to generate the velocity component used to sense angular rate). This is undesirable as the accelerometer is normally placed on the same PCB as the gyro. Just as in ABS, the accelerometer must be stable over temperature, as small amounts of lateral acceleration must be measured.

Roll stability control (RSC) changes the roll dynamics of the vehicle in response to road-handling demands. This is particularly important for vans, pickup trucks and SUVs where the higher center of gravity makes loss of control and potential rollover of the vehicle more likely during hard maneuvers. Roll stability control systems use a roll rate sensor (a gyro whose axis of sensitivity is in the roll axis), and a lateral accelerometer. Torque adjustment, selective braking, and suspension adjustment can be used to alter the roll performance of the vehicle in response to inertial measurements.

The performance requirements of the inertial sensors used for RSC are similar to that of ESC.

Table 1.Application Dynamic Range Initial Zero g Bias Zero g Bias Drift Over Temperature Initial Sensitivity Accuracy Sensitivity Drift Over Temperature RMS Noise Bandwidth ABS ±1.5g No Fixed Requirement 1mg/°C ±3% <1% 2 mg 40 Hz ESC ±1.7g ±50mg 1mg/°C ±5% <1% 2 mg 40 Hz RSC ±1.7g ±50mg 1mg/°C ±3% <1% 2 mg 40 Hz Rollover Detection ±6g No Fixed Requirement No Fixed Requirement ±10% <5% 20 mg 40 Hz Tilt Alarm±1gNo Fixed Requirement0.5mg/°C±10%1 mgIn-car Navigation±1gNo Fixed Requirement1mg/°C±5%10 HzElectronic Brake ±1g ±50mg 0.5mg/°C ±3% <1% 2 mg 10 Hz


In addition to the ubiquitous frontal and side impact air bags found in most cars today, rollover detection systems are becoming common in vans, pickup trucks and SUVs because of their greater tendency to roll over. In the event of rollover, side curtain air bags can be deployed to protect the occupants. Rollover detection systems read the roll angle and roll rate of the vehicle to determine if it is in the process of rolling over. If so, the side curtain air bags are fired to protect the occupants.

Rollover detection systems employ a roll rate sensor to read the roll rate. The roll rate is integrated to determine the roll angle of the vehicle. An accelerometer reading vertical acceleration (Z axis) is also required as large roll angles may be encountered in banked curves with no possibility of rollover. Better rollover detection systems also use another accelerometer to measure lateral acceleration as a vehicle striking a curb or other object while sliding sideways is much more likely to rollover.

Gyros used for rollover sensing do not require the same resolution as those used in VDC systems, but they must have excellent rejection of external shock and vibration and have a larger dynamic range. Analog Devices' ADXRS300 iMEMS gyroscope is commonly used in this application because of its insensitivity to external shock and vibration (Figure 2). Since rollover events happen over a fairly short period of time, the performance requirements for the accelerometer are not as severe as for ESC or RSC. However, better performance can be obtained using a more stable accelerometer. As we will see later, there are other reasons why a rollover detection system might use an ESC-grade accelerometer.


Many automakers (particularly European manufacturers) are currently including anti-theft systems incorporating tilt detection systems as standard equipment. A dual axis (lateral and longitudinal axis) accelerometer is used to detect changes in inclination as would occur if the car were being jacked up or towed. Zero g bias temperature stability of better than 50 mg over the automotive temperature range is required to ensure that temperature changes do not result in annoying false alarms.

Large electrolytic fluid tilt sensors have the required sensitivity for this application, but have several disadvantages. Their tilt range is generally limited, so it is possible to park a car at an inclination larger than the sensor can tolerate. In addition, the widely varying automotive temperature environment is not well tolerated by fluid tilt sensors. Analog Devices ADXL213 iMEMS accelerometer is ideal for this application because of its unique combination of stability, high sensitivity, wide range, compact size and high quality.

Electronic parking brake systems automatically actuate the parking brake sufficiently to prevent slipping when the automobile is parked on an incline. While driving, the system prevents the car from rolling backward on a hill when neither the gas nor brake pedals are pressed. The system measures the inclination of the vehicle, determines how much braking force is required, and applies it. Aside from safety and convenience advantages, electronic brake systems allow the carmakers to do away with the parking brake lever (or pedal) and associated hardware. This reduces the cost and weight of the vehicle, while improving reliability

As with ABS, a low g longitudinal accelerometer is used in this application. The performance requirements are similar to that for ABS.

In-car navigation systems are becoming an increasingly popular option. More than half of all cars sold in Japan today sport a navigation system. A global positioning system (GPS) is at the heart of a navigation system, but GPS information alone is insufficient for car navigation. The GPS can tell you where you are (position and altitude), but not what direction you are facing. Magnetometers (electronic compass) are highly reliable long-term heading sensors, but can give false readings in the short term because ferrous metal objects (such as the car next to you) perturb them.

When the system is first started the initial direction of travel is matched up with map data. Once initial direction is established, yaw rate sensor information is used to determine when and how much the car has turned until directional data is verified by map matching.

In urban settings it is not unusual to have the GPS signal obscured for short to moderate periods of time by tall buildings or tunnels. At these times, the navigation system relies on the yaw rate sensor for heading information and, in some cases, a low g longitudinal accelerometer for position information. The yaw rate information is integrated once to determine heading and acceleration signal is integrated twice to derive position (this technique is called dead reckoning).

Ideally, the noise performance requirements of the yaw rate sensor and accelerometer are much better than what is available at moderate cost. So software (in the form of map matching) comes to the rescue and makes certain assumptions about where the vehicle might be.

Many other accessories can (and currently do) use inertial sensors in their operation. However, an exhaustive list of them is beyond the scope of this article.


Figure 1 illustrates the many inertial sensors used in a fully featured car today. In some cases, up to 15 axes of inertial sensors (accelerometer and gyro) are used. Because there are only six possible degrees of mechanical freedom, it is obvious that many of these sensors are redundant. We have arrived at this situation because historically each system has been purchased from different suppliers. But today, the concept of a cluster of inertial sensors sending their information to whatever system needs it is becoming the goal of many automotive OEMs.

Clearly, there are cost savings to be had using a single sensor for multiple functions,. However, there are other advantages as well. Generally, the best place to mount an inertial sensor is near the center of gravity of the vehicle. The center of gravity of an automobile is normally somewhere near the center console in the passenger compartment — a rather valuable piece of real estate in an automobile. Using a reduced sensor count, and smaller sensors if possible, makes the sensor cluster practical from a packaging point of view.

There are two disadvantages to this approach, however. Each sensor must be as good as the most demanding application, so the individual cost of each sensor can be higher than the would otherwise be the case. This is tolerable since the overall vehicle cost is reduced. More significantly, a single sensor malfunction could take out several systems. Highly reliable sensors and high fault coverage built-in self-test systems can mitigate this hazard.

This concept may not extend to crash sensing, however. Some separate stand-alone accelerometers would still need to be placed at some locations around the car for crash sensing as necessitated by the required proximity to crash zone.


The proliferation of inertial MEMS sensor content in cars is driving system designers to rethink the current stand-alone sensor architecture in favor of an inertial sensing cluster. Once the inertial information for all systems is available, additional features will add little cost to the vehicle. While cost will still be a factor, the cost savings realized by elimination of redundant sensors will dominate the equation. Only those sensor technologies that can offer high performance, small size, and robust operation will eventually make the cut.


Harvey Weinberg is a senior applications engineer for inertial products at Analog Devices Micromachined Products division where he has worked for six years. Prior to that, he worked for 10 years as a circuit and systems designer specializing in process control instrumentation. He holds a Bachelor of Electrical Engineering degree from Concordia University in Montreal, Canada.

Sponsored Recommendations


To join the conversation, and become an exclusive member of Electronic Design, create an account today!