From ascen@world.std.com Tue Feb 21 08:21:25 1995 Return-Path: Received: from mx5.u.washington.edu by stein2.u.washington.edu (5.65+UW94.10/UW-NDC Revision: 2.32 ) id AA09387; Tue, 21 Feb 95 08:21:25 -0800 Received: from hitl-new.hitl.washington.edu by mx5.u.washington.edu (5.65+UW94.10/UW-NDC Revision: 2.31 ) id AA08131; Tue, 21 Feb 95 08:21:24 -0800 Received: by hitl.hitl.washington.edu; id AA12838; Tue, 21 Feb 1995 08:17:12 -0800 Received: from uucp6.UU.NET by relay3.UU.NET with SMTP id QQyebh17560; Tue, 21 Feb 1995 11:21:17 -0500 Received: from world.UUCP by uucp6.UU.NET with UUCP/RMAIL ; Tue, 21 Feb 1995 11:21:11 -0500 Received: by world.std.com (5.65c/Spike-2.0) id AA20446; Tue, 21 Feb 1995 11:15:17 -0500 Newsgroups: sci.virtual-worlds Path: ascen From: ascen@world.std.com (Ascension Technology) Subject: Re: TECH: Questions about Polhemus Fastrak Message-Id: Organization: The World Public Access UNIX, Brookline, MA X-Newsreader: TIN [version 1.2 PL2] References: <3ib0qb$lvc@nntp1.u.washington.edu> Date: Tue, 21 Feb 1995 16:15:15 GMT Lines: 118 Apparently-To: uunet!sci-virtual-worlds@uunet.uu.net Status: OR Tsung-chieh Tsay (ttsay@magnus.acs.ohio-state.edu) wrote: : Anybody has ever used Polhemus 3SPACE Fastrak? : I found when I translated(moved without rotation) the receiver : about only 20in. from the transmitter. The rotation value from Polhemus : increased, too. But I didn't rotate the receiver. How can I calibrate it? The effect you are seeing is most likely caused by conductive or ferrous metal in your working environment. One test of this is if your rotation value (and position values) deviate from expected as the receiver moves closer to metal or as the receiver moves further away from the transmitter then metal is from the transmitter. Check your operating environment for metal. We have a demo we show to people which is pretty impressive in which the 3SPACE and Flock of Birds are side-by-side on a table and we place sheets of metal between the transmitter and receiver while reading the position and angles of the two system's receivers. Of course the Polhemus tracking system goes wild, and generally our system will show some small responses (from which I launch into a discussion about taking steps to minimizing these distortions such as running our system at a different frequency, which - surprise! - works) One of the first things that our side-by-side shows is the large effect a metal strip in the table has on the Polhemus. 20 degree constant change in Elevation values from the small amount of metal built into the table. This is explained to let the user know the Polhemus system is not broken, and also to introduce the concept of metal problems there application will have to consider. Once you become aware of metal, you will notice it in furniture, in the walls, and with an AC magnetic based system you can detect metal very well. The reason.. Eddy Currents: Polhemus tracking systems use constantly varying magnetic fields to 'connect' the transmitter and receiver. These fields generate eddy currents in electrically conductive material. Eddy currents generate secondary magnetic fields which distort the shape of the field around the transmitter. Magnetic based tracking systems relies on the symmetry of these fields to calculate the position and orientation of the receiver. Ascension, on the other hand, uses a patented DC based magnetic system to generate the fields. Remember, it's CHANGING magnetic fields that cause these eddy currents. The Flock of Birds turns the field on, then holds the field on for a length of time. The eddy currents from the pulse quickly decrease as time goes by, so by the time we measure the field, most eddy currents have dissipated. Ferrous Metal: Ferrous metal is a problem for anyone using assumptions about magnetic fields symmetry to calculate position/orientation. Ferrous metal is magnetic and always distort generated magnetic fields. This aspect is a problem for all magnetic based tracking systems. Luckily for us, most applications environments may be modified to remove or replace ferrous metal with non-ferrous metal or non- metallic substances. Changing metal environment: Consider changing your tracking environment. Move what metal you can away from your Polhemus tracking system. Try to remove metal from anywhere in it's operation range, not just in the area between your transmitter and receiver. Remember, these fields are all around the transmitter, and that metal directly behind the transmitter will distort the field at that distance and further in ALL directions. Magnetic Mapping: You can try tocharacterizee the distortion in your system and implement a correctionalgorithmm. This is a standard solution and pretty common for the Polhemus system and they should be able to provide details on implementing this solution. Mapping the Ascension Flock of Birds is not unheard-of, and we recommend it to anyone who is looking for more accuracy then the 0.1" RMS transnational, or the 0.5 degree RMS angular accuracy our system has when it goes out the door, or to increase accuracy in our long range transmitter (ERT) system It can increase the accuracy to about double what the systems specs already are. The basic mapping steps are very similar to compensating for metal, and I will be happy to send you what information we have. If you would like this information, or have any magnetic tracking based technical questions please contact me at: ascension@world.std.com, (802) 860-6439 fax, or (802) 860-6440 voice. Also, feel free to download and examine the information on our FTP site: ftp.std.com /ftp/vendors/Ascension. Our latest user manual, user software, and short technical description is there, as well as some marketing info. Use binary for all files that are not *.txt. Best of luck on your project. Regards, Steve Work Ascension Technology Corporation Technical Support : Tsung-chieh Tsay : Construction Laboratory for Automation and System Simulation(CLASS) : Civil Eng., Ohio State Univ.