We created a racquet mount for our ML2 sports app. When the ML2 controller is mounted there is a delay, which causes problems with our physics engine. When the controller is unmounted, the controller tracks just fine. The sensors of the controller are not covered when mounted.
I am in the bubble, dimmer lights and it is better, but the racquet follow is delayed when full swings - causes the physics to be off, due to the VR racquet not following the physical. I am not sure what to do on this, we have tried a bunch of things. I tried disconnecting and re-connecting (11 second of the home button), same thing. I think the physics itself is good, just cannot get the racquet to follow the physical is some conditions well.
The controller is losing the tracking when doing swings.
Rac is attached to the controller model and its following the controller inside unity .
Physics we are using Box casting to check for the ball collision to return it.
Main problem occurs the lose of tracking when doing hard swings. There is a delay or latency for the controller to follow, when attaching to the mount.
However, it's not happening when we hold the controller in our hands .
The issue might be with the tracking technology of the controller. The controller uses the two cameras at the front of the device to track it’s position and rotation in 3D space. Additionally there are IR lights at the top of the controller that are tracked by the headset.
You may be running into the limitation of the tracking technology. Have you tried reading the controller rotation directly using the input system? Or do you find that the controller stops reporting position and rotation when swinging the controller quickly?
My guess is its getting affected by the cameras and IR light .
Some times in backswing when the controller is away from the head set , we have this issue .
This isnt happening when we are holding the controller in hand.
We make sure the mount is not disturbing the camera and the led
Yes at times controller stops reporting position when its far away from the camera with full motion swings.
Yes we are using the controller rotation and position from the Input system too
No, I don't believe that the controller continues reporting once the position is no longer reporting. Do you have any advice that can help us? Will there be any updates to the controller tracking in 2025? If we can get the same tracking results while the controller is mounted as we do when the controller is not mounted, we will be in great shape.
I think you are running into a limitation of the IR visibility and the update rate of the visual positioning of the controller. You may need to consider different mounts (tiling it up more so the ring around the touch pad is more visible to the headset cameras) or techniques for your usecase.
Maybe there is another Bluetooth peripheral device you can use to track the racket.
I'm not sure how you would align the racquet with the physical one. You may want to use a combination of both the controller (mounted so that the touchpad points more towards the headset) and another device that can capture IMU data.
That said, this would be more of a research project as I have not tried to do this type of tracking, and the results might not be robust enough depending on your use case.
You can access the World Camera via the Pixel Sensor API and view the secondary stream, which would show which cameras are able to visualize the IR lights on the controller when it is mounted to the racquet.
