When Robots go Whoozy: Basic Advances, IMU

There is some truth to Urban dictionary defining IMU as the whoozy acronym  “I Miss You”. The unifying theme among robotics, AR/VR, and “the future of innovation” is the integration and efficacy of an inertial measurement unit (IMU).

An IMU  is a sensor packaging containing 3 discrete sensors that can be utilize to track moving things: gyroscopes that measure angular velocity, acceleratormers for linear acceleration and magnetometers to test magnetic fields. IMUs are used in gaming controllers VR headsets, vehicle tracking, altitude, reference and orientation systems. 

The  most common complaint in robotics and in VR/AR is the feeling of motion sickness. This phenomenon attributes to data lag, aka high latency, low resolution, and vitiation of the data garnered [1]. This article ascertains to delineate the caveats of various means of garnering data in autonomous vehicles, as well to explore research and investment themes redefining IMUs. 

There are four key market requirements for inertial systems [2]:

  1. Precision to ensure maximum safety 
  2. Repeatability to consistently meet high automotive standards
  3. Scalability to achieve higher production volumes 
  4. Affordability to meet automotive industry needs

These four guidelines are not only great areas for emerging markets in hardware, but for innovative research and improvement. The difficulty is cost-maximization with high accuracy of data. Currently autonomous manufacturers heavily employ Light Detection and Ranging (LiDAR), Radio Detection and Radio (RADAR)  and Micro-electro-mechanical systems (MEMS), and conjunction of thereof. These recent innovations are notable improvements since atavistic roots of SONAR innovation from WWII.

LiDAR depends on pulsating laser beams and garnering data from the reflected light through a scanner and GPS receiver, and is great for resolution of elevation. This process serves advantageous and of higher resolution via 3D images and topography as light travels exponentially faster than sound. However, light can be interfered with through weather conditions, warranted rain or snow, or even heavy sunlight can prevent accurate resolution of road markings. 

The principles of RADAR is similar to LiDAR and SONAR in that instead of light or sound waves, the medium of purport for reflective measurement is radio. useful for obscured objects, and travels much further than sound. RADAR is resilient through weather conditions, travels much further, but lacks in fastidiousness of image detail [3]. 

Latest innovations that expand the capaciousness of LiDAR and RADAR in hardware are micro-electrical mechanical systems (MEMs), and fiber optic gyroscopes (FOGs).

MEMS is particularly useful for IMUs in that they allow large scale computations within microchips, as it combines both mechanical and electrical components.These gyros use the Coriolis Effect. The Coriolis Effect states that if a mass is on a rotating frame of reference or platform, the mast must increase its angular rotation in order to continue its inertial course of motion. Applications of MEMs stem from their computational power combined with their microscopic scale, from inkjet printers to determining volumetrically molecules for DNA sequencing and autonomous lab on a chip technologies pave the way for pharmacy on a chip devices. MEMS is also explored within decreasing latency and for Moore’s law in optical switches to relay visual information, radio frequency (RF MEMS) for long range communication. However, due to the nature of microfabrication of MEMS they are limited via their properties and attached substrates such as quartz and silicon, which may leave residual stress and instability in the system. 

Fiber optic gyroscopes (FOG) are optical location sensors based off of the sagnac effect of light signals as supposed to converting mechanical signals to electrical data of a MEMS component. The largess of FOG relies on the Sagnac Effect, which proposes that in a closed path, light waves will travel in opposite directors but reach a sensor at different rates based off of their angular rotation. The main caveat of FOG is that fiber optics are currently very expensive due to the difficulty of polarized maintaining optical fibers to maintain clean light trajectory [5]. FOGS are too expensive for mass production of consumer use, but are currently utilized in marine, navy, and space exploration.

Currently MEMS applications have been utilized in inertial navigation systems (INS) at a sacrifice for 20-30% but at a fraction of the price for FOG applications [6]. Explorations are made in IMUs to detect position more accurately via combining various MEMS with LiDAR through sensor fusion to extract one measurement, as well as respective filters such as complementary, particle, Kalman and Madgwick for agglomerating data. Other limitations may be due to heat, and bandwidth in which either MEMS and FOG can function while maintaining precision. Furthermore, some gyroscopes have limited bandwidth below a certain frequency due to sensor processing and feedback, which may lead to error rippling. 

Future blog posts ascertain to delineate use cases of respective filters and various applications for VR/AR, IoT, autonomous vehicles and robotics as market applications expand. Future studies will ascertain how to minimize and understand the specificities of types of residual stress, the bottlenecking of sensor fusion, and alternatives to current method of data amalgamation. Examples include nascent developments of resonance fiber and photonic fiber optic gyros. 

Kiyoko S. Osone
CEO, Weclikd Inc.
Founder,  Knockout Capital LLC


[1] “How IMUs are used in VR applications.” Accessed September 12, 2019. https://www.hillcrestlabs.com/posts/how-imus-are-used-in-vr-applications.

[2] McCormack, Sean. Photonic Chip Technology-Based Inertial Systems: Disruptive Technology for Safe, Precise Autonomous Navigation. KVH Industries, 2019.

[3] “LIDAR vs RADAR Comparison. Which System Is Better for Automotive?” Archer Software, August 21, 2019. https://archer-soft.com/en/blog/lidar-vs-radar-comparison-which-system-better-automotive.

[4] Heaton, Chris. “Reducing the Size and Cost of Fiber Gyroscopes.” Reducing the size and cost of fiber gyroscopes. Accessed September 12, 2019. http://spie.org/news/0440-reducing-the-size-and-cost-of-fiber-gyroscopes?SSO=1.

[5] Goodall, Chris, Chris Goodall, Chris, and Trusted Positioning, Inc. “The Battle Between MEMS and FOGs for Precision Guidance.” The Battle Between MEMS and FOGs for Precision Guidance | Analog Devices. Accessed September 12, 2019. https://www.analog.com/en/technical-articles/the-battle-between-mems-and-fogs-for-precision-guidance.html.