Chen Dongyao's team at Shanghai Jiaotong University proposed a fully automatic compass calibration technology

Recently, the team of Assistant Professor Chen Dongyao of the Johns Hopcroft Center for Computer Science in the School of Electronic Information and Electrical Engineering has made breakthrough progress in the field of Internet of Things in the field of computer networks, and the research result "Automatic Calibration of Magnetic Tracking" has been accepted by ACM MobiCom 2022, a top conference in the field of network systems, with Chen Dongyao as the corresponding author
。 The study pioneered fully automated compass calibration and is important
for critical applications such as precise human-machine interaction.

Research background

As a technology that is gradually emerging, magnetic field-based perception is strongly promoting the realization of various interactive applications, among which representative application scenarios include high-precision attitude tracking, human-computer interaction, and tactile sensing
.
This technology utilizes multiple MEMS magnetic sensors to capture changes
in the magnetic field over close range.
However, magnetic sensors are very sensitive to real-life interference, especially the common hard and soft iron effects
.
As a result, users frequently need to perform tedious and lengthy calibration processes, which severely limits the usefulness
of magnetic field perception.

Typical application scenarios for magnetic field perception, such as pure gesture interaction under AR/VR, joint angle detection, etc

Research Results

To remove this limitation, the research team proposed the MAGIC automatic calibration system
.
MAGIC targets MEMS magnetic sensor arrays that automatically calibrate both soft and hard iron interference
at the same time.
To minimize user intervention, the research team introduced a new type of automatic triggering module
.
Unlike traditional manual calibration methods, MAGIC achieves superior calibration performance with minimal user involvement (e.
g.
, for tracking application scenarios
).
Through empirical research, the research team showed that the overhead generated by MAGIC is also extremely small, with a total energy consumption of only 0.
108J
.

The left figure shows the magnetic field drift, and the right figure shows the commonly used "around the 8-word" compass calibration algorithm

The largest source of external magnetic field interference is the hard iron and soft iron effects in the space, such as common headphones, and even ferromagnetic furniture will bring interference
.

(a) Demonstrated a miniaturized coil design

(b) Demonstrate the effect of assembly on a magnetic field sensing device

To verify the calibration performance of MAGIC, the research team designed and conducted a series of experiments
.

Calibration performance of MAGIC

The research team also tested more handwriting schemes and used them to judge the performance
of MAGIC.

MAGIC can calibrate the trajectory of the magnetic field movement

Given that MAGIC clears a key "roadblock" in the field of accurate magnetic field tracking, the technology has great application prospects
.

About the author

Chen Dongyao

Dongyao Chen, Assistant Professor of the School of Electricity, received his Ph.

About the International Conference on Mobile Computing and Web (MobiCom)

The International Conference on Mobile Computing and Networking (MobiCom) is the world's top academic conference in the field of wireless networks and mobile computing systems, is a CCF Class A conference, sponsored by ACM SIGMOBILE, dedicated to solving challenging challenges
in the direction of network communication and mobile computing.