Almost a year ago, I posted here about how VR technology could help control and perform research on UAVs in an indoor environment and at a more affordable cost. Currently, the most advanced indoor UAV research lab in the USA and Switzerland are using camera-based motion capture systems. Unfortunately, those systems cost tens of thousands of dollars, require calibration, and are not very portable. I think the lack of availability of motion capture equipment is a factor significantly limiting UAV research and development.
My team and I are now releasing the Otus Tracker, a motion capture system and software package designed specifically for indoor UAVs and robots! The Otus Tracker has similar capabilities to camera-based motion capture, but is 5-10 times cheaper, portable, and does not require calibration.
The Otus Tracker and the RCbenchmark tracking lab have the following characteristics at launch:
Sub-millimeter accuracy tracking
250 Hz refresh rate
Very low latency (<5 ms)
5 m x 5 m x 5 m tracking area
Plugins for C, C++, ROS, Matlab, LabView and Python. You can also code your own plugins in other languages that support UDP communication.
- Control code examples (including the code for the demo in the launch video)
Controlling a drone is hard. You need infrastructure to make it work. We hope that with the Otus Tracker, software, and examples we provide, we are significantly lowering the financial and technical barrier needed to develop and do research with UAVs, whether it is in SLAM, artificial intelligence, control, or swarm. We are planning to add to our documentation to help new researchers get started with UAV development.
The Otus tracker is developed by the same team that worked on RCBenchmark’s motor testing tools: the Series 1580 (5 kgf) and the upcoming Series 1780 (25 kgf). Our dynamometers are used by hundreds of hobbyists, businesses and universities worldwide. I think we have quite a unique position in the UAV industry, as we have worked with many of the important players. Based on their feedback and the fact that measuring the position and orientation of a drone was one of our biggest problems when we developed UAVs (the other problem being measurement of thrust and motor efficiency), we decided to develop the Otus Tracker.
The tracker is shipping now! Please let me know if you have questions in the comments, or contact our team at info@rcbenchmark.com.
Comments
@Daniel: Our goal is to allow the development of affordable quads with ultra precise indoors control, eventually affordable for the community. Unfortunately, right now, just the BOM (without assembly, R&D and everything else) is many, many times 50$/units in the low thousands of units. Still, when this kind of technology drops by an order of magnitude in price, more universities, startups and businesses can access it, which hopefully leads to better tech for everybody.
@Georacer: We had bad UX experience as researchers with so many software for technical tools, we are trying our best not add another software to the list. We are trying to make the experience as seamless as possible for the users, as we did for our thrust stands.
If it works as seamlessly and as intended, then $1000 sounds like a deal to me, given the current state of the art.