editor's blog
Subscribe Now

Locus of (Con)Fusion

At the MEPTEC MEMS conference a couple weeks ago, one sensor fusion question kept coming up over and over: Who’s in charge of sensor fusion?

On the one hand, IMU makers in general are giving away sensor fusion packages that help integrate the data from the individual sensors in their combo units. Then there are guys like Movea that don’t make sensors themselves, but integrate across a wide variety of sensors for both high- and low-level motion artifacts (motion in their case, but the concept extends to anything).

So who’s job is it?

I happened to have a conversation with Movea’s Dave Rothenberg that same day, and I brought the topic up.

His first comment was that what most IMU makers refer to as sensor fusion is simply the software required to establish orientation, which is a relatively low-level characteristic. He said that this correlated to Movea’s Foundation series, which they’ve actually de-emphasized a bit since it is hard to sell against free software, even if they do think they do a better job.

The sensor guys say they’re the right place to do it because they know their sensors better than anyone else. That actually covers two separate things: the physical characteristics of the sensors and how they operate, and the low-level data details – formats etc. Dave mentioned that it is work for them to adapt their software to different sensors, since they don’t all look or speak alike. (Area for future possible standardization? Future topic…) But they have to get it right in order for the other pieces that lay over it to work properly: errors at the bottom level will compound as further algorithms manipulate them.

(This also ties into the question of loose vs tight coupling, since a sensor maker is in a better position to do things tightly.)

Of course, it’s unlikely that the sensor vendors will want to take on the higher-level algorithms since those, almost by definition, will, at some point, involve sensors that they don’t make. So it looks like things may go the way of the embedded world, where critical low-level drivers and other bits of firmware are provided by (or in close partnership with) the processor maker, with other companies layering higher-value stuff on top. That seems to be how the sensor world is shaping up, which leaves room both for the sensor guys and for the third-party folks.

Leave a Reply

featured blogs
Nov 15, 2024
Explore the benefits of Delta DFU (device firmware update), its impact on firmware update efficiency, and results from real ota updates in IoT devices....
Nov 13, 2024
Implementing the classic 'hand coming out of bowl' when you can see there's no one under the table is very tempting'¦...

featured video

Introducing FPGAi – Innovations Unlocked by AI-enabled FPGAs

Sponsored by Intel

Altera Innovators Day presentation by Ilya Ganusov showing the advantages of FPGAs for implementing AI-based Systems. See additional videos on AI and other Altera Innovators Day in Altera’s YouTube channel playlists.

Learn more about FPGAs for Artificial Intelligence here

featured paper

Quantized Neural Networks for FPGA Inference

Sponsored by Intel

Implementing a low precision network in FPGA hardware for efficient inferencing provides numerous advantages when it comes to meeting demanding specifications. The increased flexibility allows optimization of throughput, overall power consumption, resource usage, device size, TOPs/watt, and deterministic latency. These are important benefits where scaling and efficiency are inherent requirements of the application.

Click to read more

featured chalk talk

MEG-Array and M Series
Sponsored by Mouser Electronics and Amphenol
In this episode of Chalk Talk, Jeremy Ellis from Amphenol Communications Solutions and Amelia Dalton chat about the benefits of Amphenol’s MEG-Array and M-Series connector solutions. They also investigate the tooling configurations available for these solutions and how Amphenol’s ball and socket BGA interface can simplify board routing, eliminate press-fit constraints on via and provide excellent SI performance.
Oct 21, 2024
34,448 views