Application of MEMS sensor in drone

According to reports, the market size and application range of drones are booming, and it seems that new applications will appear every other day. Whether it's delivery of mail or parcels, entertainment for all ages, security monitoring, management of agriculture or industry, or a new world of aerial photography, drones can be seen everywhere.

Initially, most drones were nothing but simple toys. However, their flight capabilities have recently increased significantly, making their controls safer, more stable, and easier, making them more widely used in real life.

A key factor in these increased capabilities is the use of high performance microelectromechanical systems (MEMS) sensors, so the sensor market in drones is growing rapidly:

According to Yole's "Sensors for UAVs and Robots" report data, the sensor market for UAVs and robots is growing strongly, and is expected to reach $709 million by 2021, with a compound annual growth rate of 2018-2021. 12.4%.

MEMS sensor affecting flight performance of drones

UAVs rely on inertial MEMS sensors to maintain a stable direction, be accurately manipulated by the user, or fly automatically. However, some of the challenges faced by drones complicate their system design: motor calibration is not perfect, system power varies with load, operating conditions change rapidly, or sensors introduce inaccurate information. These can lead to deviations in the positioning process, which eventually leads to positional errors during navigation and even to drone failure.

To make drones not just toys, or even "going to the next level," high-performance MEMS sensors and advanced software are essential. The high-precision inertial measurement unit (IMU), air pressure sensor, magnetometer, dedicated sensor node (ASSN) and data fusion between the sensors on the advanced drone have a direct and substantial impact on their flight performance.

Due to size constraints and harsh environmental and operating conditions, such as temperature fluctuations and vibrations, sensor requirements are taken to new levels. MEMS sensors must minimize these effects and provide accurate and reliable measurements.

The main methods to achieve superior flight performance are: software algorithms (such as sensor calibration and data fusion), mechanical system design (such as reducing vibration), and MEMS sensors based on the requirements and needs of drone manufacturers. Below we focus on MEMS sensors with some examples.

The Attitude Reference System (AHRS) is the "heart" of drones and includes inertial sensors, magnetometers, and processing units. AHRS estimates the direction of travel of the equipment, such as the angle of roll, pitch and yaw. Inaccuracies in the sensor, such as offset, sensitivity errors, or thermal drift, can cause orientation errors. Figure 1 shows the orientation error (rolling, pitch angle) as a function of accelerometer offset, which is usually the largest factor in the sensor error chain. For example, an acceleration offset of only 20 mg will cause the device to have a 1 degree directional error.

MEMS传感器在无人机中的应用

Figure 1: Direction error caused by accelerometer offset

Inertial Measurement Unit (IMU)

The IMU includes accelerometers and gyroscopes with embedded handlers. This enables it to determine the motion trajectory, including linear motion and rotation.

Bosch Sensortec's BMI088 is a 6-axis IMU with a 16-bit low noise accelerometer and a 16-bit low drift gyroscope. The technology of this high-precision device is derived from high-end automotive sensors, so it maintains excellent bias and temperature stability over a long period of time, as well as excellent vibration robustness, making it ideal for drones.

Figure 2 shows typical values of BMI088 drifting with temperature.

MEMS传感器在无人机中的应用

Figure 2: Typical values ​​of BMI088 drift with temperature (both acceleration and rate are 0)

The amount of drift shown indicates that the accelerometer offset is within 10 mg and the gyroscope offset is less than 0.5 dps. In addition, BMI088 behaves linearly with temperature and has almost no hysteresis. This makes the BMI088 extremely attractive for use in drones and robots.

Air pressure sensor

The high-performance air pressure sensor built into the drone accurately measures altitude and can be combined with IMU readings for height control. The air pressure sensor must minimize external influences and errors. Currently, combined with other sensors, such as GPS and optical flow sensors, and ranging sensors, it improves system reliability and reduces positional errors.

Bosch Sensortec's BMP388 air pressure sensor is used to provide height information for improved flight stability, altitude control, take-off and landing performance. This makes manipulating drones a breeze and therefore appeals to a wider range of users.

UAV requirements for air pressure sensors are often extremely demanding. Even with adverse weather and temperature effects, the high accuracy must be controlled within tight tolerances, and the sensor must maintain low latency and minimal drift over time. The BMP388 meets these demanding requirements with a relative accuracy of +/- 0.08 hPa (+/- 0.66 m) and an absolute accuracy of +/- 0.5 hPa between 300 and 1100 hPa. Low temperature coefficient compensation (TCO) is usually Less than 0.75 Pa/K. It has an attractive price/performance ratio, low power consumption and a very small package size of just 2.0mm x 2.0mm x 0.75mm.

In addition to improving the TCO, there are many factors that can improve overall accuracy: relative accuracy, noise, stability, and absolute accuracy. From clumsy toys to high-precision aircraft, the potential of today's drones in industrial and commercial innovation applications exceeds the imagination of engineers.

Magnetometer

The magnetometer is like a compass that can set the heading for the drone based on the Earth's magnetic field. Just like Bosch Sensortec's BMM150, it is a 3-axis digital magnetometer.

The BMM150, in combination with the BMI088 IMU, offers a 9-degree-of-freedom (DoF) solution for heading estimation and navigation. The BMM150 maintains good performance over a wide temperature range, with 16-bit resolution and high magnetic field resistance (no magnetization provides stable sensor offset), making it ideal for drone applications and minimizing calibration The amount of work required for sensor offset.

Dedicated sensor node

The Dedicated Sensor Node (ASSN) is a highly integrated, intelligent sensor hub that integrates multiple sensors in a single package and is equipped with a programmable microcontroller. It provides a flexible and low power solution for motion sensing applications.

For example, Bosch Sensortec's BMF055 is an ASSN that integrates an accelerometer, gyroscope, magnetometer, 32-bit Cortex M0+ microcontroller with processing software, and the outputs of each sensor. BMF055 can be used as AHRS in combination with directional processing software. The device is available in a small 5.2mm x 3.8mm x 1.1mm package, saving valuable space and weight. The sensor provides an all-in-one package for drone applications. Figure 3 illustrates the application of the BMF055 in a drone as a directional processing unit with integrated sensor fusion algorithms.

MEMS传感器在无人机中的应用

Figure 3: AMFS application of BMF055 (ASSN) in drones

Signal processing and software

In addition to the individual sensors, we can also look at the system diagram of the overall structure of the signal processing of the drone, as well as the software needed to integrate sensor readings and control.

Figure 4 shows the functional blocks for different signal processing in a typical consumer drone. The left column shows the individual sensors and the right column shows its derived software processing functions, such as direction processing and flight control algorithms. The dark blue square sensor is essential for optimal stability control of indoor and toy drones, while the gray squares represent optional sensors that can be used to extend outdoor flight and automatic waypoint navigation.

MEMS传感器在无人机中的应用

Figure 4: Schematic diagram of signal processing for consumer drones

By integrating various sensors and data fusion, software functions such as directed processing can be performed directly on the chip. In addition to MEMS sensors, Bosch Sensortec offers sensor data fusion software for directional processing, including sensor calibration, sensor data pre-processing and directional processing. For drone manufacturers, this can significantly reduce the complexity of engineering and software, reduce unnecessary risks and reduce time-to-market.

However, manufacturers still need to provide their own software and special code for the mechanical design and powertrain of drones, such as control loops and the use of specific functions.

Typical drone function

Let's take a look at how innovative MEMS sensor technology can be combined with software to implement modern drone functions. Nowadays, even low-cost toy drones are generally complex. First, by using the output of the IMU, the stabilizer can keep the drone level. By integrating data from the air pressure sensor, the drone can be maintained in height and position. For example, in a toy application, it controls the drone to flip when the height is constant. The result is that the operator no longer has to spend so many hours of practice to master the basic controls and significantly reduces the risk of accidental collisions.

Data fusion with the GPS module adds some interesting features to the UAV's outdoor flight, such as automatic flight between several waypoints, and the “Back Home” feature – the drone can automatically return and land safely to the start position.

Other novel features include "track mode" or "follow me mode", where the drone can rotate around a specific point or can follow a person autonomously. Combined with the camera, the operator can see himself from the “bird's eye view” while walking “with a drone” or interact with the drone through gestures.

Let me swim in the sky

With the development of robotics, semiconductors and current MEMS sensor technology, especially the increasing precision and miniaturization, it indicates that unmanned remote control aircraft will become more popular in the future. From weather or air pollutant monitoring, livestock management, security or express delivery systems to next-generation augmented reality games or the Internet of Things (IoT) platform, the role of high-tech aircraft and drones in our daily lives is growing. The more important, the Bosch MEMS sensors will cast a powerful "heart" for them.