I’ve recently returned from a trip to China where I’ve attended the two Imagination Summits organized in Beijing and Shanghai, respectively. Alongside a series of exciting presentations from my colleagues as well many of our ecosystem partners, I also enjoyed seeing several demonstrations and prototypes of next-generation devices using Imagination IP in addition to many real-life products using our technologies.

One Imagination-powered demo I’d like to highlight today comes from Standing Egg and relates to a previous announcement we made earlier this year.

Standing Egg is a Korean fabless start-up that has integrated our MIPS M5100 CPU inside a low power chip designed for sensor fusion. A team of engineers representing the company was present at our summit showcasing two sensor fusion demos running on a compact Xilinx Artix-7 FPGA board.

In addition to the MIPS M-class CPU, the FPGA board also included a Standing Egg SGA100 accelerometer, a gyroscope and magnetometer.

Standing Egg - MIPS FPGA

The first demo consisted of a 6-axis IMU (Inertial Measurement Unit) running a filter algorithm that fuses the data from the SGA100 accelerometer and the gyroscope; this provides an efficient computational recursive method to estimate the state of a process at a given time and obtain feedback from the sensor measurements.

Standing Egg - MIPS MCU - Sensor data processingStanding Egg sensor data processing algorithm

After performing calibration and applying another low pass filter to the gyroscope and accelerometer signals (IMU configuration), the resulting signal is used to obtain a smooth measure of rotation in 3D space.

The second demo uses the readings provided by the calibrated and filtered accelerometer and magnetometer sensors to create a virtual gyroscope. The algorithm synthesizes the output in angular rates to approximate the behaviour of a hardware gyroscope.

Standing Egg - Demo screenshot

Substituting a real gyroscope inside IoT devices with a software-based implementation can eliminate the high power consumption and increased costs commonly associated with hardware gyroscopes.

You can see a comparison of the precision delivered by the hardware and virtual gyroscopes in the diagram below:

Standing Egg - Virtual Gyroscope Data Output

To find out more about upcoming events, please visit our dedicated page on our website.

Make sure to follow us on Twitter (@ImaginationPR@ImaginationTech), LinkedInFacebook and Google+ for the latest news and announcements from Imagination and our ecosystem.


  • Venu

    Asking as a layman without much knowledge in the area.. what is the cost and latency tradeoff of HW vs virtual gyro.

    • By using a virtual gyro, you save the cost of integrating more dedicated sensors into the device; depending on the product and volume shipments (smartphone, wearable, other IoT device), the cost savings could be significant.

      Thanks to the advanced algorithms used by Standing Egg and the MIPS MCU, the performance of the virtual gyro is very close to the dedicated hardware version.