I want to measure velocity or acceleration of a moving body(E.g: Car,bicycle,etc) without using a GPS module. Is it possible to calculate coordinate acceleration or velocity using an IMU.
Posted by Jonathanrjpereira 2 years ago
Hello, I found here on https://www.instructables.com/ a lot of topics talking about gyroscope, accelrometer, & magnetometer sensors. How they work, the idea behind them, and how to make use of the data returned by different sensors. Then stays the ultimate problem of filtering the data and getting reliable data. Then there's an online algorithm for that written by PhD student which is a very good algorithm. It'll be used on PC side as I understood from some gurus that such algorithms will get the arduino to its knees. There're some sensors out there that you may cost you a little more but they have the option to get filtered data directly from the sensors. Then comes the question itself. Is it possible? Well that was my bachelor project to read how other people did it and redo it. Then I remember the first time I was searching for how to work with these sensors .. it was a mess. Then I made a decision to make this thing available for everyone to play with. So far I implemented "Arm Tracking" and based on the idea .. anything else will be possible actually. I wanted someone (or some people) to help or brainstorm with me how can we make it easier for anyone who wants to start with the sensors .. from writing arduino library to simplify this .. to the "Skeleton Tracking" thingy. Even if it'll only be explanation but I wanted it to be one big guide for everyone just to pay back those who helped me with other stuff they were good at. Even if the guide is about gathering resources in one big topic and linking them together. I hope I could deliver what I wanted to say properly :D Thanks in advance
Posted by tdk4 5 years ago
Hi all, It's not a new question but the available technology keeps changing so I'll ask this again, having not found any current answers online that are useable at a diy, non electrical engineering level... I'm working on a project in which a body worn camera is tracked in space, then recreated as a virtual camera in the corresponding virtual 3d space (kind of like a lo-fi version of the cameras used on the "Avatar" set). This is to be done over 15-60 minute durations in an indoors area of about 30X30 feet. What I need for this is tracking of the camera's position and orientation in space, precise and continuous (as much as I can get, but at least within and inch or 2, and a few degrees, of the true values), to be either logged on the tracking device and/or transmitted to a computer on set. Since it is body worn, it needs to be light and wireless. Since it's on a moving body in a space with some objects in it, optical tracking isn't really a good option. I looked at a wide range of solutions, from "flock of birds" and the inertiaCube on the crazy-expensive side to sparkfun's IMU sensors, but frankly am a bit bewildered by the selection and don't know how to figure out which is the best to try first, given my needs. I'm really not looking to reinvent the wheel or rebuild everything from scratch, just find a solution for this not costing more than a few hundred dollars that is fairly reliable. Software-wise I want to be able to get 3D XYZ position and rotation values from it that I'll be able to use in Realtime 3D software such as Max/MSP Jitter and the like. Any ideas, stories of previous experience with such setups, links, etc would be extremely welcome. Many thanks! nadav
Posted by nadvass 7 years ago