New Skytrak with Club data coming?

[HappyGilmore102]

Hi Everyone

Looks like Rapsodo filed a patent in June 2021 regarding a unit that is camera and radar based, collects club data.

Maybe we will see it at January's PGA show, fingers crossed


Patents Assigned to Rapsodo Pte. Ltd

• MEASUREMENT AND RECONSTRUCTION OF THE GOLF LAUNCHING SCENE IN 3D
  
  
  Publication number: 20210299540
  Abstract: A method, including scanning a golf club to obtain scanning information; inputting the scanning information into a processing system; using at least one camera positioned behind and in-line to a golf swing direction and at least one lighting unit to obtain a series of images of a golf club during the golf swing; converting the series of images into parameterized motion representations; using at least one radar to obtain a radar signal; inputting the parameterized motion representations and the radar signal into the processing system; receiving golf club parameters and golf swing parameters as an output of the processing system; and generating a visual model of the golf club and the golf swing in a virtual space using the golf club parameters and the golf swing parameters.
  Type: Application
  Filed: June 11, 2021
  Publication date: September 30, 2021
  Applicant: RAPSODO PTE. LTD.
  Inventors: Jin XU, Batuhan OKUR, Lodiya Radhakrishnan VIJAYANAND

• Measurement and reconstruction of the golf launching scene in 3D
  
  
  Patent number: 11077351
  Abstract: A golf launching monitoring arrangement allows equipment to be placed at a position behind the player (i.e., behind the golf ball), to measure both club and ball movement. A 3D scan of the club head before the play serves two purposes: 1) 3D registration that enables accurate measurement of the club head position and orientation for the camera system measuring the club movement from the back; 2) for reconstruction of the launching scene. With a 3D model of the club head, a simple 3D model of the golf ball and accurate measurement of their movement during the play, a full 3D golf launching scene can be reconstructed authentically. With this reconstruction, the movement of both the club head and the resulting ball movement can be replayed at any viewing angle, with any frame rate and at whatever resolution for the players or the coaches to view and analyze.
  Type: Grant
  Filed: May 28, 2019
  Date of Patent: August 3, 2021
  Assignee: RAPSODO PTE. LTD.
  Inventors: Jin Xu, Batuhan Okur, Lodiya Radhakrishnan Vijayanand

• OBJECT TRAJECTORY SIMULATION
  
  
  Publication number: 20210060407
  Abstract: A method may include receiving a group of images taken by a camera over time in an environment, in which the camera may be oriented within the environment to capture images of an object in a substantially same direction as a launch direction of the object, and the group of images including a first image and a second image. The method may further include: identifying a first position of the object in the first image; identifying a second position of the object in the second image; generating a flight vector based on the first position of the object and the second position of the object; and determining one or more flight parameters using the flight vector. Additionally, the method may include: generating a simulated trajectory of the object based on the flight parameters; and providing the simulated trajectory of the object for presentation in a graphical user interface.
  Type: Application
  Filed: November 17, 2020
  Publication date: March 4, 2021
  Applicant: RAPSODO PTE. LTD.
  Inventors: Batuhan OKUR, Boyi ZHANG

[Jgold2790]

We wouldn't see anything from this for at least 2 years. Interesting though.

[Morini]

Yes, it was discussed on this forum earlier this year. My take on it is that it is not a Skytrak 2. It seems to be a Rapsodo MLM on steroids. I read through the patent at that time, and here is what I posted in June:

Ok, I read through the patent. It is very interesting, but I do not think it is a successor to Skytrak. Rather it seems to be a significant upgrade to the Rapsodo MLM, and will likely only be able to read full ball flight (ie. range) and not balls hit into a net.

It is a 4 camera system (two pairs) positioned behind the ball. Two high res slow frame rate cameras to read the ball. Then two low res high frame rate cameras to read the club head.

Ball: Quite simple. Two high res slow frame rate cameras (30 fps) track the ball as it flies down the range and will calculate 3d trajectory. It will not read spin directly, but rather will derive it from observed 3d ball movements. Therefore, this technology will not work if hit into a net etc as there isn't enough ball flight to derive spin.

Club head: The club head is "scanned" by the golfer prior to hitting by rotating it and a 3d model of the club head is created. The radar unit is used to turn on the low resolution high frame rate capture, and low res images of the club head is captured. The trick here is that the low res club head "blob" is compared to the 3d model, and it figures out club head orientation and position at each frame. This is then used to created a 3d recreation of the club head movements.

The unit can't see the ball until it flies, so ball speed is measured some distance away (third frame is 20 feet away for driver), and ball speed is then used to estimate time of impact, which is compared to 3d club head position at that exact point in time. Then face angle, AoA etc can be measured.

I know that Quad and TM4 often differ on face angle because the disagree on exact moment of impact. I think the same would apply here. The patent claims calculation of impact to 0.1 milliseconds. I think such accuracy would be very difficult to achieve calculating ball speed from low frame rate cameras, but who knows. Also, if the ball launch angles, club path and impact location is known with a high degree of accuracy, the face angle can be derived indirectly. Path and AoA would be very accurate with this system. Of course wind will affect ball flight, so not sure how accurate launch data will be if spin isn't directly measured.

Interesting, but clearly not an LM that is useful for sim. Brilliant, though, if the high frame rate 2d to 3d reference model works. Will allow for cheaper cameras, so competitive advantage, just the way Skytrak can use $30 dollar cameras to compete against $3000 ones. This seems to be Batuhan Okur's M.O. Clever use of cheaper technology to dominate lower segment of market.

edit:

Upon a second read, it does say that the high frame rate cameras will also read the ball. Maybe it will do so with high degree of accuracy. I am not sure. I believe it will be very difficult from behind with low res cameras. Maybe with tons of sub pixel image processing. VLA and HLA will be no problem. Spin I am not so sure.

[HappyGilmore102]

Hi

If you read the latest patents its the other way around.

Camera takes high res pictures of the club and the radar gets the ball data.

It looks like a working version of the ES16 tour but sits behind like a MLM.

This thing will be very good, I think we'll see it in January at the PGA show, 2 years time will be to late.

[Morini]

HappyGilmore102, in the originatl patent the radar was used solely to identify the backswing and ready the unit to capture clubhead and ball. "The radar detects a reversal of club head speed and triggers the cameras 101A, 101B to run in a high-FPS mode for continuous image capture to capture the club head in the field of view (FOV) (1905)."

I will read through latest to see if there is a change.

[Morini]

If you read the latest patents its the other way around.

Camera takes high res pictures of the club and the radar gets the ball data.

It looks like a working version of the ES16 tour but sits behind like a MLM.

I read the most recent patent, and I disagree with your take on it. The radar triggers cameras. The low res high FPS cameras uses AI to determine 3d club head position. After club is out of view, the low res cams also capture some images of the ball, but likely can't pick up spin given distance and low res. The high res low FPS cameras track the ball during flight, just like the MLM. This will obviously only work outside, and as such useless for home sim. It is possible that the unit tries to use impact position and club data to back out spin axis, but I have doubts about how accurate that would be, as retroactive impact estimation needs to be so precise.

[HappyGilmore102]

HI Morini

You are probably right, but i don't see the reason for another MLM if they can't get spin right.

That makes me think, they probably have something up their sleeve.

Garmin R10 will see updates and has blown MLM out of the water in my view, so another MLM is dead with no sim play.