Hello- Recently we've installed a VR experience space on an exhibition by Jung Yeondoo in Mito, Japan.
How it works: - People walk through the corridor in which garbages are arranged on both sides, while holding an OVR DK2 connected to a laptop. - When people look the garbages through the OVR, they see the things of nature - trees, waterfalls, river, etc.
Collaborating with the team in Seoul National University (Music and Audio Research Group), positional tracking in the corridor could be achieved.
DK2 was so unstable on the laptop, but we finally succeeded integating it to our system. (I hope the stable SDK and runtime be released as soon as possible so we can stablize our work using it.)
However, as expected, yaw drift occurs as walking through the corridor. The garbages include much meterials which can interfere magnetic field, so can think MagYawCorrection does not work. But, when update DK2 firmware using OculusConfigUtil, yaw keeps correct until 4 or 5 times of moving through the corridor. (I tried to turn off MagYawCorrection by OVRHmd.ConfigureTracking, but the result was worse.)
How can I improve yaw drift correction? Any suggestion or idea will help us.
Interesting. Maybe the PlayStation Move? http://en.wikipedia.org/wiki/PlayStation_Move Or a Wii sensor bar placed on top of the HMD (lightweight, needs almost no power) and a Wii Remote at the end of the hall in the direction people are walking 😉 If (most of the time they won't) the dots appear on the infrared camera inside the Wii Remote you can calibrate the orientation accordingly. Or Microsoft Kinect. Maybe there are solutions in the motion capture field.