LeadTech.co

Moore's Law applies to #AI ? In #China ? ! Who knew ? Evidently someone has translated Gordon Moore into Mandarin. Although Moore's Law originally applied to transistors, the concept holds true: every 24 months, computational power doubles in power at a lower cost, rendering the previous generation (and some formerly hot stocks and AI centers) technologically obsolete. This has been the case for only the last 60 years 😎 Raise your hand if you're surprised. Be honest We have 2 patents for the use of AI for determining the lead on a moving target - but the challenge is that the rifleman or gunner has to hold on the AI-recognized target long enough for the system to get a lock on it. And if the target is evasive, no lock, no fire, no bueno. That was the result of a recent field test of the #NotSoSmartShooter: the damn thing failed to fire 14% of the time against drones, which is an argument in favor of "OI" - Original Intelligence - and a shotgun or a Mk 19 airburst grenade, after sufficient training in a VR Marksmanship Trainer. One service is going to get this right. Rhymes with "sardines" 🤠

Misleading Training on Robotic Targets Novice riflemen have been taught a simplified mil rad (MRAD) hold technique in order to pass a basic test on tracked targets moving crosswise at known distances and speeds. The trick involves using windage marks on a scope reticle as predetermined hold points. Snipers with spotters use a MRAD technique effectively in combat, but that's been turned into a gimmick by civilian range instructors who give trainees the pre-set holds under ideal conditions. That range instructor's trick doesn't work on anything else: Oblique movers Movers that change speed or direction Second Focal Plane reticles in variable mag scopes Mechanical, red dot, or reflex sights High angle shots - from a roof or up a hill Movers at less than 100m or more than 300m Meaning most of the targets you'll see in the wild. A better way is available. Courtesy of the United States Marine Corps. Coming soon to a Trijicon, Inc. VCOG or ACOG near you against a MVP Robotics, Inc. robotic target moving at an uncertain distance at an uncertain speed at an uncertain angle. Like reality 😎

1st Draft of a Shotgun Marksmanship Training Simulator on FPV Drones Shotguns are being used against FPV drones in Ukraine as the last line of defense by both sides. This is a rough draft of VR shooting at FPV drones. Will change the VR gun to a semi-auto, require reloading, add display of a lead target, increase drone speed, and add evasive maneuvers. Shotgun will be changed to a military semi-auto like the Beretta Defense Technologies Benelli #AIDroneGuardian. The sim controller can be a Heckler & Koch USA 416 or a SIG SAUER, Inc. MCX (that is represented in the simulation as a Beretta semiautomatic shotgun) with a Mantis Tech eBolt using a HTC VIVE XR E or Varjo XR4 headset. The sight picture is a bit off in the video because it is not the POV of the dominant eye. The POV in the headset shows correct eye dominance and a good sight picture straight down the barrel. We can simulate a reflex optic on a military shotgun Trainee sees the lead displayed in VR, repeats drill until they have gotten a sense of how much lead is required, then they are tested in VR and, when proficiency plateaus, go to the IP CONVERGENCE CORPORATION range to see if they can hit the damn things. Ray Stephenson noticed that this VR Beretta 694 over/under magically fires like a semi-auto Beretta M 1014 with infinite reloads 😎 Stay tuned #VR #Simulators #drones

Why AR FoF LaserTag is Mission Improbable on Leading Movers We have seven patents on displaying a Lead Target, two of which are for AR, including live targets. We assume that for live targets, a remote sensor, such as Trillium Engineering camera or a Skydio drone is tracking the mover and we have time to allow for system latency in calculating the lead. Similar drill in AR if the target is a hologram; we can factor latency into the lead calculation when we build the simulation, as in VR or 2D. Shooting a real mover in AR LaserTag in Force on Force (FoF) presents a variable systemic latency that can be greater than the Time of Flight (ToF) of the bullet. This doubles if both players are moving. That complicates both the lead calculation and the virtual hit registration- since the laser hit sensors on the target are of no use when lead is required. Cumulative systemic latency in AR FoF LaserTag consists of: 1. Track POA of barrel 2. Track position of moving Shooter 3. Track position of moving Target 4. Compute ToF 5. Trigger signal Bluetooth lag 6. Virtual hit zone displayed in AR 7. Virtual hit registered (updated 1, 2, 3) 8. Sum latency of 1 - 7 That variable system latency effectively limits AR FoF LaserTag to hopping around in a warehouse playing Wack-a-Mole Hostage Rescue. Like Classic LaserTag 😎

COTS Drop/In VR Controller All it takes to turn an XM7 or M4A1 into a VR robot shooter is a Mantis Tech bolt and magazine and a HTC VIVE tracker on the rail attached with a bracket from 3D Simulations, Inc. Plus a VR HMD. Haptic optional. Simulator software from LeadTech.co. Barrel removed at the advice of my lawyers Dewey, Cheatham & Howe 😎