MEMS Sensors Are the Heart of a Drone

Sensors Insights by Sergej Scheiermann

The market for drones is booming in size and scope, with new applications appearing seemingly every other day. Drones are becoming increasingly ubiquitous, whether it be delivering mail or packages, entertaining children both young and old, security monitoring, management in agriculture or industry, or opening new horizons for aerial photography.

Initially, most drones were relatively simple toys. More recently, however, their flying capabilities have improved significantly, making them safer, more stable, and easier to control, and thus allowing them to be used for a much wider range of real life applications.

One of the key factors in this improvement has been the use of high-performance micro-electro-mechanical system (MEMS) sensors. The market for sensors in drones is growing fast:

Based upon data from IHS Markit (Motion Sensors for Consumer & Mobile Devices – 2017) the market for MEMS motion sensors (i.e. accelerometers, gyroscopes, IMUs and pressure sensors) in drones and toy helicopters is expected to reach in total approximately 70 million units until 2021, with a CAGR 2018-2021 of 17 %.

 

Impact of MEMS Sensors Drone Flying Performance

Thanks to inertial MEMS sensors, the drone keeps its orientation stable and can be precisely controlled by a user, or even fly autonomously. Nevertheless, there are some challenges which make drone system design complex: the motors are not perfectly calibrated, system dynamics can vary depending on the payload, operation conditions can rapidly change, or sensors can introduce inaccuracies. This can lead to deviations in the orientation processing and finally to position deviations during navigation, or even to drone fails.

For drones to be more than toys and to go to the next level, high-quality MEMS sensors and sophisticated software are essential. The precision of the drone’s Inertial Measurement Unit (IMU), barometric pressure sensor, geomagnetic sensor, Application Specific Sensor Node (ASSN), and sensor data fusion have a direct and substantial effect on its flying performance.

Size constraints as well as demanding environmental and operational conditions such as temperature changes and vibrations, bring the sensor requirements to the next level. MEMS sensors must reduce these influences as much as possible and deliver precise, reliable measurements.

There are several ways to achieve excellent flight performance: software algorithms such as sensor calibrations and data fusion, mechanical system design for example to reduce vibrations, and selection of the MEMS sensors themselves according to the drone manufacturers’ requirements and needs. Let’s focus on MEMS sensors and consider some examples.

The heart of the drone is its Attitude Heading Reference System (AHRS) which includes inertial sensors, magnetometers and processing units. The AHRS estimates the device’s orientation, for example roll, pitch and yaw angle. The sensor’s inaccuracies such as offset, sensitivity errors or thermal drifts lead to orientation errors. Figure 1 shows the orientation errors (roll, pitch angle) as a function of accelerometer offsets, which is usually the biggest contributor in the sensor errors chain. For example, accelerometer offsets of only 20 mg lead to a 1-degree error in device orientation.

Fig.1: Tilt error induced by offsets of accelerometer.
Fig.1: Tilt error induced by offsets of accelerometer.

 

Inertial Measurement Unit (IMU)

An IMU includes an acceleration sensor and a gyroscope, with embedded processing. This enables it to determine motion, both in terms of linear movement and rotation.

Bosch Sensortec’s BMI088 is a 6-axis IMU with a low-noise 16-bit accelerometer and a low-drift 16-bit gyroscope. This highly accurate device’s technology is derived from high-end automotive sensors, so it can provide outstanding bias and temperature stability over extended time periods and exhibits high vibration robustness, making it ideal for drones.

Figure 2 shows a typical offset drift over temperature of the BMI088.

Fig. 2: Typical zero-g and zero-rate offset drifts over temperature of the BMI088.
Fig. 2: Typical zero-g and zero-rate offset drifts over temperature of the BMI088.

The offset drifts shown are within a range of only 10 mg for the accelerometer and less than 0.5 dps for the gyroscope sensor. Furthermore, the BMI088 exhibits linear behavior versus temperature with very small hysteresis. This makes the BMI088 very interesting for drone and robotic applications.

 

Barometric Pressure Sensor

A high-performance barometric pressure sensor built into the drone precisely measures altitude and can be used in combination with the IMU’s readings for altitude control. The barometric pressure sensor must reduce external influences and inaccuracies as much as possible. Nowadays, in combination with additional sensors such as GPS and optical flow, distance sensors are used to improve reliability of the system and reduce positional errors.

Bosch Sensortec’s BMP388 barometric pressure sensor provides altitude information for improving flight stability, altitude control, take-off and landing performance. This makes steering a drone child's play, and thus more appealing to a broader range of users.

The requirements placed on pressure sensors in a drone are often extremely demanding.  Subject to the effect of less than ideal weather and temperature conditions, altitude accuracy must remain within a tight tolerance band, plus the sensor must exhibit low latency and negligible drift over time. The BMP388 meets these tough requirements with a relative accuracy of +/-0.08 hPa (+/-0.66m), an absolute accuracy between 300 and 1100 hPa of +/- 0.5 hPa and a low TCO of typically less than 0.75 Pa/K. It exhibits an attractive price-performance ratio, coupled with low power consumption and an extremely small package size of only 2.0 x 2.0 x 0.75 mm3.

In addition to TCO improvements, multiple factors contribute to an improved overall accuracy: relative accuracy, noise, stability and absolute accuracy. From clumsy toys to high precision aircraft; the present potential for innovative industrial and commercial drone applications is limited solely by the engineers' imagination.

 

Magnetometer

A magnetometer acts like a compass, enabling the heading of a drone to be established in relation to the earth’s magnetic field. An example is Bosch Sensortec’s BMM150, a 3-axis digital geomagnetic sensor.

The BMM150, in combination with the BMI088 IMU, provides a 9-degree of freedom (DoF) solution and can be used for heading estimation and navigation. Stable performance over a wide temperature range, 16-bit resolution and immunity against strong magnetic fields (no magnetization provides stable sensor offsets) make the BMM150 well suited for drone applications, and minimize the effort required for sensor offset calibration.

 

Application Specific Sensor Node

The Application Specific Sensor Node (ASSN) provides a highly-integrated, intelligent sensor hub, which combines multiple sensors in one package, along with a programmable microcontroller. It provides a flexible, low power solution for motion sensing applications.

For example, the BMF055 from Bosch Sensortec is an ASSN with an integrated accelerometer, gyroscope, magnetometer and a 32-bit Cortex M0+ microcontroller to manage the software, including sensor outputs. BMF055 combined with orientation processing software can be used as an AHRS. The device is housed in a small 5.2 x 3.8 x 1.1 mm3 package, saving valuable space and weight. This sensor provides an all-in-one package for drone applications. Figure 3 demonstrates the usage of the BMF055 in drone application as an orientation processing unit with integrated sensor fusion algorithms.

Fig. 3: BMF055 (ASSN) usage as AHRS in drone application.
Fig. 3: BMF055 (ASSN) usage as AHRS in drone application.

 

Signal Processing and Software

Beyond the individual sensors, we can also take a system-level view of a drone's overall signal processing structure, and what software is required to integrate sensor readings and controls.

Figure 4 shows the different signal processing functions in a typical consumer-grade drone. The left-hand column shows the individual sensors, with the columns to the right representing the derived software processing functions such as orientation processing and flight control algorithms. The dark blue sensor blocks show state-of-the-art sensors mostly used in indoor and toy drones for stabilization, and the grey sensor blocks indicate extended, optional functionality required for outdoor flights and autonomous waypoint navigation.

Fig. 4: Signal processing overview of a consumer-grade drone.
Fig. 4: Signal processing overview of a consumer-grade drone.

With integrated sensors, certain software functions like orientation processing can be carried out directly on-chip, primarily by the fusion of the various sensor readings. Besides MEMS sensors, Bosch Sensortec offers sensor data fusion software for orientation processing, which includes functionalities such as sensor calibration, preprocessing of sensor data and orientation processing. For a drone manufacturer, this can significantly reduce engineering and software complexity, mitigate unnecessary risk and reduce time to market.

Nevertheless, the manufacturer will still need to provide their own software, code specific to the mechanical design and dynamics of the drone, such as control loops and use case-specific functions.

 

Typical Drone Features

Let’s look at how innovative MEMS sensor technology combined with software enables modern drone features. Sophisticated features are now commonplace even in low-cost toy drones. Firstly, a stabilizer keeps the drone on a horizontal level by utilizing IMU output. Integrating data from the barometric pressure sensor enables a drone to maintain its altitude and position – and in toy applications, for example, it can be used to make the drone flip over without changing height. The result is that the pilot does not require as many hours of practice to master basic maneuvers and the risk of accidental crashes is significantly reduced.

Data fusion with a GPS module provides drones with some interesting features for outdoor flights, for example, autonomous flying between several waypoints, and a 'return-to-home' feature where the drone returns automatically to its starting position and safely lands there.

Other newer features include an 'orbit-mode' or 'follow me-mode', where the drone circles around a specific point or could autonomously follow a person. Combined with a camera, the pilot can now view themselves from a birds-eye-view perspective while 'taking the drone for a walk' or even interact with the drone by using gestures.

 

The Sky Is No Longer the Limit

Developments in robotics, semiconductors and now MEMS sensors – including their ever-increasing precision and miniaturization – promises a future where unmanned remotely controlled aircrafts will be more present in the future. From weather or pollution monitoring, cattle management, security or delivery systems to the next generation of augmented reality gaming or IoT platforms, high-tech aircrafts and drones will play an increasingly important role in our everyday lives. And Bosch MEMS sensors will be at the heart of it all. For more information, visit www.bosch-sensortec.com.

 

About the author

Sergej Scheiermann is a Product Manager for Sensor Solutions at Bosch Sensortec. Bosch Sensortec GmbH, a fully owned subsidiary of Robert Bosch GmbH, develops and markets a wide portfolio of microelectromechanical systems (MEMS) sensors and solutions tailored for smartphones, tablets, wearable devices and IoT (Internet of Things) applications. The product portfolio includes 3-axis acceleration, gyroscope and geomagnetic sensors, integrated 6- and 9-axis sensors, environmental sensors, optical microsystems and a comprehensive software portfolio.